The IBM PC Convertible, Model 5140, was IBM's first attempt at a battery-powered laptop. Of course, it was thick, had a small screen, and expansions made it longer. However, it does share more than a few similarities with an earlier of the PC family, the IBM PCjr. While it took ideas from the PCjr. and improved on them in many respects, the system was not a great success due to its own issues, namely its high price and less-than impressive specifications compared with competitor's products of the time. The introduction of the 3.5" disk drive on this system, when all software had previously been released on 5.25" disks, meant that it took time for the market to release PC compatible software for the machine. Also, if the chicklet keyboard was the Achilles Heel of the PCjr, the Convertible's initial low-contrast LCD gave it a bad reputation. Later machines used a super twist LCD and a backlit LCD, both offering much improved readability, but like the PCjr., it was too little, too late.
Keyboard - The original IBM PC Keyboard used 83 keys. The PCjr. Cordless Keyboards use 62 keys and the PC Convertible uses 78 keys (U.S. style). The fewer keys make handling these keyboards with low-level routines quite different from the IBM PC keyboard. Both the PC Convertible and the PCjr. use the Non-Maskable Interrupt to generate IBM PC Keyboard compatible scan codes.
In an IBM PC, serial data from the keyboard is sent to the system unit, deserialized by a shift register and then placed in I/O port 60, the 8255 PPI Port A0. IRQ1 is used to signal the CPU that there is a key to be read. A program can read the raw scancode directly or execute INT 09H/INT16H to obtain an extended ASCII scancode.
In the IBM PCjr., serial data sent from the keyboard is sent to bit 6 of I/O Port 62, 8255 PPI Port C0. The NMI is generated to alter the CPU that there is a key to be serviced and the CPU deserializes the data stream, then INT 48H is executed to translate the raw scancode into codes compatible with INT 09H/INT16H. IRQ1 is never used, but the translated scancodes appear at I/O port 60.
In the PC Convertible, the raw scancode is placed in I/O Port 07D. An NMI is generated and the NMI handler will translate the raw scancode into a compatible scancode and stores the result in I/O Port 60 and generates IRQ1 and the key can be read with INT 09H/INT16H. While obviously derived from the PCjr., this is an obvious improvement over the PCjr.'s slow routine and should be much more compatible with programs expecting PC keyboard handling.
Memory Expansion - Neither the PC Convertible nor the PCjr. were intended by IBM to reach 640KB, both were intended to max out with 512KB. Memory expansions for both machines came in 128KB. IBM officially only supported the Memory Expansions for the PCjr. (128KB in system unit + 3 x 128KB expansions), but the fourth expansion was easy enough to add. IBM only physically supported four memory modules in the PC Convertible, each with 128KB, and the system always came with two installed from the factory. Third party expansions were required for the PC Convertible to reach 640KB and recommended for the PCjr. to reach 640KB.
Feature Expansion - Both machines were expanded by hardware attachments, the PCjr. via sidecars that fit onto a connector on the right side of the machine, the PC Convertible by a connector on the rear of the machine. IBM released five PCjr. attachment options, parallel, power, memory, speech and cluster, and four for the PC Convertible, serial/parallel, CRT display, speech and portable printer. If customers did not particularly care for adding onto the length of the PCjr. they must have positively detested the added weight and length that the add-ons for the Convertible added.
Modem - IBM included a modem slot in the PCjr. It sold a modem based on the Novation Smart 103 modem for the PCjr, which could run at 110 or 300 baud. It did not sell well because it was not Hayes compatible. IBM also included an internal modem socket in the PC Convertible. Apparently ignoring the market again, IBM used the same style of modem for the PC Convertible internal modem upgrade. This time the modem could also support 1200 baud.
The COM ports were similarly assigned on both systems. The PCjr. had a serial port built in and it was assigned to COM1, but the resources were at 278/03. Once the modem was installed, the serial port became COM2 and the modem COM1 at 3F8/04. The PC Convertible Internal Modem always used 3F8/04 and would be assigned to COM1 when it was installed. This time, IBM knew enough that not everyone would install the internal modem, so the PC Convertible had a serial adapter that could be set to COM1 3F8/04 if no internal modem was installed or COM2 2F8/03 if it was
CGA Support - The PCjr. had only basic, BIOS CGA support since it did not implement the crucial Color Select and Mode Control registers. Although the PC Convertible uses a CGA-friendly 640x200 LCD resolution, the built in video is not particularly CGA compatible either. Color text-mode attributes were not supported on the monochromeLCD, it was either normal video, reverse video or no display. 320x200 graphics would display three shades of gray plus black, but the graphics take up the whole screen, meaning they are twice as wide as they should be. The built in LCD does not distinguish between 640x200 and 320x200 in this instance. The MC6845 was only emulated enough for the basic modes, the color select register is not present and the mode control register has missing bits compared to real CGA.
The built-in display adapter can also be configured to emulate a MDA card. When the CRT Display adapter is attached to the PC Convertible, it will function like an MDA card and the CRT Display adapter should provide 100% CGA compatibility. The Display adapter has three video output ports, a Direct Drive port using an 18-pin BERG connector with pinouts identical to the PCjr's Direct Drive video port (except no audio). The adapter can connect directly to the PCjr. Color Display, Model 4863 or the PC Convertible Color Display, Model 5145. It can connect to an IBM PC Color Display, Model 5153 or IBM PC Enhanced Color Display, Model 5154 or any other compatible display via the PCjr. 18-BERG to 9-pin DE-9 adapter. It also has a composite video output like the PCjr. and a 6-pin BERG connector for an RF adapter. The BERG connector has a compatible pinout with the PCjr. RF switch.
Printer - Both systems had a printer designed specifically for it, the IBM PC Compact Printer for the PCjr. and the IBM Portable Printer for the PC Convertible. Both are unidirectional thermal serial printers operating at a maximum of 1200 baud and have a 2K buffer. Both use a transmission format of 1 start bit, 8 data bits and 2 stop bits. The Convertible's character set includes text-drawing characters at ASCII 176-223 which the Compact Printer does not. The PC Compact Printer can print in standard, condensed, double-wide and underline styles and 480-bit graphics, while the Portable Printer adds emphasized, superscript and subscript styles and 960 and 1920 bit graphic modes. The Portable Printer supports virtually all the control codes of the Compact Printer plus some more for its enhanced features.
Speech Attachment - The Convertible's Speech Attachment is supposed to be functionally identical to the PCjr's.
Sunday, February 16, 2014
Saturday, February 8, 2014
Introduction to the PCjr., or How NOT to Destroy It!
.png) |
| IBM PCjr. Startup Screen - Maximum RAM Count (but you can have more) |
First Things First : To use a PCjr., you need, at a minimum the system unit, the keyboard and the power cable. If you are missing any one of these three, the PCjr. is useless. I am going to go through each in terms of simple to complex :
The External Power Adapter: This is a large black power converter. It has wires on either end. One the end of one wire is a three-pronged plug. On the other end is a rectangular three pin connector that connects to the back of the system unit. The port is in between the A and the C ports. On early bricks, the cable with the three-pronged plug is detachable from the large brick. A standard three-pronged computer power cable should fit into the brick. The brick only supports 120v/60Hz AC power, so in Europe you will need a step-down converter. It outputs AC power, the AC to DC conversion is done inside the system unit. The power cable can fit into the system either way (polarity is irrelevant for AC).
The Cordless Keyboard : The original Cordless Keyboard, later known by everybody, even IBM as the "chicklet" keyboard is the less common of the two IBM PCjr. Cordless Keyboards, especially after IBM offered a free trade in to the non-chicklet keyboard. IBM called the replacement keyboard the "Freeboard", presumably because it offered some freedom from the frustration of the chicklet and was a free replacement for PCjr. owners who hated the chicklet keyboard. Internally, they function identically.
The PCjr. Cordless Keyboards use rubber-dome technology and can feel stiff yet squishy. The keyboard communicates with the system via an infrared sensor or via an optional keyboard cable. It takes 4 x AA batteries, and you should check the battery compartment to see if there are old corroded batteries. If there are, you should clean the contacts with cotton swabs or Q-tips with apple cider vinegar or lemon juice and follow it up with baking soda. You may need to unscrew the board to inspect for leakage on the PCB. The infrared can work up to twenty feet away according to IBM, but line of sight must be maintained and the sensor on the keyboard should be as perpendicular as possible to the sensor on the computer. Rechargeable batteries may or may not provide reliable power for the keyboard.
The System Unit : The computer is inside on a large system board. Most PCjr's you will find these days are the Enhanced Model, which come with a 5.25" diskette (floppy disk) drive and a 64KB memory expansion. These have 128KB of RAM total, not counting any extra memory added via sidecar. The Basic Model did not come with a diskette drive or the 64KB memory expansion. If you are getting a boxed unit, it should come with all the above and two books, the first being the PCjr. Guide to Operations and the second being the "Hands On BASIC for the IBM PCjr." A complete unit should also come with a plastic plate to cover the sidecar expansion and a Basic Model also comes with a plastic cutout to cover the floppy bay hole.
Checking for damage : The expansion ports on the rear of the PCjr. use BERG-style connectors, which little more than pins bent at right angles sticking out from the system board. These are more prone to damage than connectors set in a D-shell, but it was cheaper for IBM to implement them in this way. Check for any bent, broken or pushed in pins. Most of the connectors have a missing "key pin", so check below to discover pins that are supposed to be missing. Gently straighten or pull out any pins with a small pair of pliers or tweezers. Expansion cables are held in the port by friction alone, so do not try yanking cables out. Similarly, the sidecar expansion port also uses pins, but at least these are cased in a plastic connector like an IDE port.
Opening the case : The case top is held down by friction and tabs, there are no screws to deal with. Using a small screwdriver, insert it into each of the three indentations at top rear edge and gently pop that portion of the case off.
Inside the case : There are up to four internal expansion boards. Fortunately each can only fit one way. There are no jumpers or dipswitches inside or on the internal expansion board. IBM did not use jumpers (officially) on its sidecar expansion board either. Everything is configured in software, which was rather progressive for the time and technology. If you want a true definition of plug and play, here it is!
On the left side of the unit (with the cartridge ports facing you), is the power board, which is always present. The power board comes in either a 33W short card or a 45W long card. If the power card is about 2/3rds the length of the case, it is the short card. If it extends nearly to the front of the case, it is the long card. There is a 2-pin staking connector which the power board must be inserted into when it is put back in its slot. There are grooves in the case to stabilize the card, make sure the card fits into them.
The second board is the 64KB Memory and Display Option. As its name suggests, this adds 64KB to the 64KB on the motherboard for a total of 128KB inside the System Unit. Early memory boards are covered in a metal shield, later boards are just a bare PCB. If this board is not present, you will only have 64KB of RAM in the system and a big hole if any sidecar memory expansions exist.
The third board is the PCjr. Internal Modem Option, and as it was always an optional upgrade, it probably won't be present in your average PCjr. System Unit. IBM's modem was not very popular since it only communicated at 300 baud and did not use the industry standard Hayes command set. Third party Hayes-compatible modems were also made to fit in this slot. Like the power board, this card connects to a 2-pin staking connector and grooves in addition to its slot. This is the only internal upgrade not to come automatically bundled with the PCjr. System Unit.
The fourth board is the PCjr. Diskette Adapter. This connects to the Diskette Drive via a 34-pin ribbon cable with a pin header on one end and a card edge connector on the other. It only supports the one internal drive. Note the side of the cable with the colored wire, that side must correspond to pin 1 on the board if the diskette drive is going to work properly. Pin 5 is missing on the adapter pin header as a key pin to prevent wrongful insertion.
The diskette drive is mounted on a unique plastic drive support. It is connected to the board by two standoffs that push through the bottom of the case. To remove the drive support mechanism (and the drive itself), first unplug the cables for the diskette drive power connector and the fan power connector from the power board. Next, disconnect the drive data cable. Next, with a pair of pliers, squeeze the tabs in and push upwards on the standoffs until the constrict and pop up from the bottom of the case. These are not screws, so do not twist a screwdriver in them. Finally, pull up on the back of the diskette drive and it should be able to be lifted out of the system with no further difficulty.
If you have removed all the internal boards and the diskette drive support, you will have complete access to the interior of the PCjr. Check for any signs of damage and be prepared to dust. There is a raised metal box on a small PCB, that is the Infrared Receiver. Make sure it is firmly seated on its slot. The receiver is connected to the System Unit by its connector and two plastic standoffs, so you won't be able to remove it without removing the System Unit itself. The System Unit can be removed from the case by unscrewing four screws at the rear of the system, at least one screw holding down the cartridge ports and removing whatever sidecars are attached.
Rear Expansion Ports :
Except for the RCA ports for video and audio and the modem port (if one is installed), there is nothing standard about the PCjr.'s expansion ports. However, the hardware that the PCjr. is designed to connect to is standard with one exception. Looking at the rear of the machine, the ports are lettered as follows :
J J L K LP T V M D S C P A
J = Joystick - There are two joystick ports in the PCjr., with the first port, sometimes called the right joystick being closest to the edge of the System Unit. The inner port is sometimes called the left joystick. These connect to the PCjr. Joystick, which functions just like a standard PC joystick. The PCjr. joystick just uses a different connector. River Raid, Demon Attack and Microsurgeon will not work without a joystick connected. 8-pins, with Pin 1 missing as the key pin.
L = Spare - These pins are not connected to anything on the system board. 6-pins, with pin 4 missing as the key pin.
K = Keyboard - For the optional Keyboard Connector. The keyboard connector has a phone jack on one end and a large RFI ferrite choke. It plugs into the PCjr. keyboard and allows you to use the keyboard without IR and batteries. Third party keyboard replacements also plug into this port, but must be specifically designed for the PCjr. If there is anything plugged into this port, then the system unit will not receive the input from the IR receiver, even if its just an unconnected wire. 6-pins, with pin 1 missing as the key pin.
LP = Light Pen - IBM never marketed a light pen with a PCjr. style connector, but PC Enterprises did. 6-pins with pin 6 missing as the key pin.
T = Television - For the TV Connector, an RF switchbox to connect to an antenna screw terminal on a TV. You may need a 300 to 75 Ohm balun connector. External Audio can be heard from this connector. 6-pins, with pin 2 missing as the key pin.
M = Modem - A standard RJ-11 phone jack connector will be found here if you have an internal modem installed, otherwise it is just a hole.
V = Video - RCA Composite Video jack, better than RF, but not as sharp as :
D = Direct Drive Video - This connector supplies a CGA-compatible video signal and a mono audio signal. The standard IBM PC Color Display Model 5153 or IBM Enhanced Color Display Model 5154 or compatible monitors can connect to it via a 9-pin adapter. IBM released one because its PCjr. Color Display was not available at launch. The IBM PCjr. Color Display Model 4863 has a connector that plugs directly into this port. The IBM PCjr. Color Display has a built-in speaker with a volume control wheel for the external audio output. The IBM PC Convertible Color Display Model 5145 also uses the same connector as the PCjr. Color display, but may or may not have a speaker for the external audio. 18-pins. An adapter exists to allow the PCjr. and Convertible displays to work with a 9-pin D-shell port.
S = Serial - Any serial device can use this port with a proper adapter, such as from IBM. The only device I know of that has the connector for this port is the IBM PC Compact Printer. An adapter exists to allow the Compact Printer to be used with a standard serial port. 16-pins.
C = Cassette - IBM made a cable to interface with cassette drives and cassettes for data storage and retrieval. It has two mini-jacks and one micro-jack. The cassette interface is functionally identical to the interface in the IBM PC Model 5150. 8-pins, with pin 5 missing as the key pin.
P = Power Connector - See above
A = Audio - RCA Mono Audio Jack, necessary to hear sounds from the external audio unless you are using the TV Connector or the 4863 PCjr. Color Display.
Cartridge Ports :
On the front of the machine, there are two cartridge slots. These slots are functionally identical. Like any cartridge slot, they can accumulate dust and should be cleaned with a Q-tip and 99% Isophropyl Alcohol. The most common cartridge that can be found is the BASIC cartridge (IBM Cartridge BASIC). This cartridge is intended to supplement the Cassette BASIC already in the system ROM and add support for the advanced sound and video capabilities of the PCjr. and diskette drive support. Game cartridges are generally rare, especially the good ones from Activision and Imagic. Lotus 1-2-3 is a rare example of a cartridge program with a cartridge for both slots.
Sidecar Port :
On the side of the PCjr., there is the expansion port of the system, the sidecar port. If there are no sidecars installed, there may be a plastic faceplate, which snaps into two holes and can be easily pried off with a screw driver or fingers. Sidecars are pushed onto the pins, then screwed into system unit or the next sidecar with a flat head screwdriver. You can extend the length of your system quite considerably with expansion sidecars.
Most PCjr.s have at least an IBM Parallel Printer sidecar attached, as the PCjr. did not provide a parallel port in the system unit. The Printer sidecar provides one unidirectional parallel port using the standard DB-25 connector.
Many PCjr.s also have a Memory Expansion Attachment sidecar, from IBM or a third party company. These sidecars add at least 128KB of RAM to the system. Few self-booting programs will recognize more than 128KB of RAM in a PCjr. The memory expansion's device driver must be loaded in DOS for DOS to report more than 128KB available to programs. A device driver called JRCONFIG will work with any memory expansion.
If you have a 33W power board in the system unit and use more than two sidecars, you will need the IBM PCjr. Power Attachment sidecar, which provides more power for sidecars. This sidecar must be inserted before any sidecars it needs to power (closer to the system unit) and has a separate power supply identical to the PCjr. system unit power supply. They use the same connector. With a 45W power board, you can use three sidecars comfortably.
Turning the System On :
When you turn on the screen, you will immediately see, if the system is working at all, the IBM logo and the fifteen RGBI colors that a CGA monitor can display in addition to black. The memory will be counted and tested. Once the memory has been counted, the screen will change to a blinking cursor, the diskette drive will seek and try to boot a disk in the drive. Any expansions with a bootable ROM, like a hard drive or SCSI expansion interface, will attempt to boot at this time. If neither the diskette drive nor any bootable interface is available or functioning properly, then the PCjr. will boot a cartridge, and if no cartridge is inserted, then Cassette BASIC.
Fan and Heat :
The fan inside the system blows cool air on the floppy drive and diskette. IBM felt this was necessary to prevent a disk from melting inside the drive after heavy usage. The fan is not intended to cool the system, that is what all the vents are for. You can disconnect the fan without immediate apparent issue if you do not often use the disk drive, but the fan improves airflow inside the case. This should help the Power Module (the hottest element in the system by far) and the chips underneath the floppy drive (including the CPU). Although you may be tempted by the prospect of silent operation, leave the fan on.
Keyboard Adventure and System Diagnostics :
The PCjr. has a built-in program called Keyboard Adventure. This can be loaded by pressing the Esc key just after Cartridge or Cassette BASIC has loaded. You cannot have pressed any other key before pressing the Esc key. This program was made to help introduce first-time computer users to the keyboard, and can be used to test the keyboard. A full description of the program can be found in the PCjr. Guide to Operations manual, which came with every system. Press Ctrl + Alt + Del to exit and reboot the system.
To start system diagnostics, any time after the PC has booted, whether to BASIC or to a disk, press Ctrl + Alt + Ins. This will appear to reboot the system, but the memory will be counted and you will be shown a screen with several ASCII graphics and numbers underneath them. The tests are rather cryptic and are fully explained in the Guide to Operations with the exception of Test Tags H, I, L, M, N, O & P. Here are a list of the Test Tags and what they test :
Test Tag 1 - Diskette Drive (requires spare floppy disk)
Test Tag 4 - 40-Column Display modes and 16KB graphics modes
Test Tag 8 - 80-Column Display modes and 32KB graphics modes (requires IBM PCjr. 64KB Memory and Display Expansion Option)
Test Tag 5 - Light Pen
Test Tag 6 - Joysticks
Test Tag 9 - internal PC Speaker and 3-voice sound (external speaker, TV Connector or PCjr. Color Monitor required)
Test Tag J & K - Cordless Keyboard (Test Tag J treats the keyboard as a 62-key keyboard, Test Tag K treats the keyboard as if it were an IBM PC 83-key keyboard and allows you to use key combinations for the missing keys)
Test Tag G - IBM PCjr. Internal Modem
Test Tag H - IBM PC Compact Printer
Test Tag I - RS-232 (with loopback Plug)
Test Tag L & M - IBM PCjr. Cluster Adapter Attachment
Test Tag N, O & P - IBM PCjr. Speech Adapter Attachment
To escape from any test, press Fn + Break. Press Ctrl + Alt + Del to exit and reboot the system.
Error Messages :
If you are unfortunate, you may encounter an error message. Error messages are assigned a letter, here are a list of error messages and possible solutions :
Error A - Memory Problem, remove memory expansion sidecars and internal memory option to try to isolate which one is causing the program. Will cause the system to hang.
Error B - Keyboard Problem, move the keyboard away from any extremely bright light source, replace the batteries, move the keyboard closer to the system if using IR and do not press any keys during the bootup process. May cause system to hang.
Error C - Cassette Problem, press Enter to continue booting without using the cassette. Usually caused when the cassette motor relay goes bad.
Error D - Serial Port Problem, may hang system. Unplug any device from the serial port.
Error E - Modem Problem, press enter to continue booting without using the Modem. Make sure the modem is properly inserted.
Error G - Cartridge Problem, may hang system if cartridge is inserted, remove cartridge.
Error H - Diskette Problem, press enter to continue booting without using the Diskette Drive. Make sure the diskette drive adapter card is properly plugged into the system, the diskette drive is properly connected via the cable to the diskette drive adapter, the diskette drive is plugged into the power connector.
Error X - Unknown, may hang system. May not actually exist.
Video and Audio Connections :
The PCjr. is very video friendly in that it has a composite video output jack that can connect to almost any modern TV. You do not need a hard to find CGA compatible monitor to get video output. The colors may not be as pure, the pixels may not be as distinct, but graphics should be visible. If you boot to DOS and you see 80-column color fringed text, use the DOS command MODE BW80 to switch the text to black and white. If you want more legible text, then use MODE BW40 to use a 40-column text mode.
If you find that your PCjr.'s display is off center or partially cut off, press Alt Left Arrow to shift the visible display to the left and Alt Right Arrow to shift the visible display to the right. You can also enable or disable the keyboard audio click by pressing Ctrl + Alt + Caps Lock
The internal beeper inside the PCjr system unit will only generate PC Speaker sounds. If you want to hear sounds from the 3-voice chip, the PCjr. Speech Adapter or from the Cassette Port, you will need to connect the PCjr. to an amplified speaker via the RCA audio output jack, to a TV using the TV Connector switchbox, or to a 4863 PCjr. Color Display.
COM and LPT assignment.
The built in serial port is assigned to COM1 and LPT1 in the BIOS if there is no Internal Modem or Parallel Printer Attachment installed. However, the resources used by the serial port are I/O 278H and IRQ3, which are generally assigned to COM2. The Internal Modem uses I/O 3F8 and IRQ4, the BIOS assigns the Modem to COM1 if it is installed and the serial port becomes COM2. If the Parallel Printer Attachment is installed, it becomes LPT1 and you may need to redirect LPT1 to COM1 or COM2 to print to the serial-based Compact Printer.
Recommended Programs :
Most DOS programs will require a 256KB PCjr., so if you don't have one you are probably going to be limited to self-booting (PC Booter) programs. King's Quest I & II, The Black Cauldron and Below the Root are excellent adventure games that take advantage of the enhanced graphics and sound features of the PCjr. Super Boulderdash and Jumpman are solid ports of these classic games with PCjr. features. Microsoft Flight Simulator 2.0 looks best on the PCjr. and can take advantage of a 256KB PCjr.
The Device Driver JRCONFIG is a must if you are running a 256KB or more PCjr. in DOS. The PCjr. was released with PC-DOS 2.1, and was officially last supported in PC-DOS 4.0. DOS 5.0 and above require patching via DEBUG to run on the PCjr. While DOS 3.3 and below can run in a 128KB PCjr., DOS 4.0 and above are not likely to leave much memory left for programs unless 256KB or more is in the system.
Sunday, February 2, 2014
PCjr. Post-Cancellation, or the Afterlife of a Flop
November 1, 1983 - IBM announced the PCjr. to great fanfare. Commentators predicted and expected that IBM's new system would bring order and compatibility to the fractured home computer market just as it had the business market with the PC and XT. Magazines are published geared to the new system and software is updated to support the new features and otherwise run on the PCjr.
March 18, 1985 - IBM announced that it was discontinuing manufacturing the PCjr. system in April, intending to sell the remaining stock at a discount and not release any further peripherals or upgrades for the system. It would continue to service the machines within their warranty period.
The reasons for the PCjr.'s failure have been well-documented elsewhere. Had it not been discontinued, it would have probably have been sold until April 1, 1987, when the rest of the PC family was retired to make way for the PS/2 machines. However, the public apathy for the machine a result of the machine's high price, competition from other home computers, miserable chicklet keyboard, incompatibilities with PC software and the high cost of the upgrades meant that despite IBM's weight in the industry, it had to be withdrawn.
What happens when a buyer, believing in the solid name of IBM, buys a computer whose lifespan had just been reduced from years to months or even days? Most PCjr.s had been sold during the 1984 Christmas season. IBM not only abandoned the PCjr, but also abandoned its family and entertainment software products. With one exception, the last games IBM marketed for the PC or PCjr. were released in 1985. While it tried to put on a brave face by announcing a new and improved successor to the Jr., nothing came of it in the United States. The closest thing to a PCjr. replacement was the PC JX, essentially the next generation of PCjr. released only in Japan, Australia and New Zealand. It sold very poorly. IBM would not return to the market with a home-friendly machine until the PS/1 in 1990.
When IBM announced the PCjr was being discontinued, software developers were the first to jump ship. Bill Gates' Microsoft, which had heavily supported the Jr. with the Microsoft Booster sidecar and an updated Flight Simulator 2.0 which took advantage of the Jr.'s capabilities, eliminated its PCjr. development almost overnight. Spinnaker Software tried to burn its PCjr.s in a great bonfire, only to find out that they wouldn't burn!. Sierra On-line quickly focused its efforts to the PC and the Tandy platform, and many other game companies did the same. Any software companies who had not yet jumped on the PCjr. bandwagon certainly were not about to start supporting the machine once its discontinuance had been announced.
All PCjr. magazines had already folded by December, 1984, in response to the weak market for the PCjr. before the 1984 Christmas season. Peter Norton had written a trilogy of how-to books for the PCjr., Discovering the PCjr, Exploring the PCjr and Mastering the PCjr, but the third book was never released due to the discontinuing of the PCjr.
Every game that had support for the PCjr.'s graphics and sound, as identified in an earlier post on this blog, were all self-booting games released in 1984 or 1985. In 1986, games were more and more beginning to run in DOS and very few advertised PCjr. support. Any DOS games from 1986 or later that did run on the PCjr., with expanded memory, tended to treat it just like an IBM PC with CGA. Sierra Online was the main exception, because its AGI engine had been originally built on the PCjr. and thus it was no great effort to make sure its games continued to run on it. All Sierra AGI DOS games thus will work on the PCjr. Some of the AGI booters as marketed by Tandy may not. So will Thexder, 3-D Helicopter Simulator, later versions of Silpheed but later Game Arts/Falcom ports will not. Early versions of King's Quest IV, Leisure Suit Larry II and perhaps Police Quest II specifically did not run on the PCjr., but later versions of these SCI engine games and other 16-color SCI0 engine games would run on a PCjr.
For the home user that had almost certainly dropped over $1,000 on the PCjr., it was not so easy to discard the computer like a bad memory as the mainstream industry had done. Some undoubtedly felt that their purchase was now useless and consigned it to the attic or the basement. Others probably only used it until they could afford something better. But for some of the owners of the approximately 500,000 PCjr.s IBM shipped out of its factories, they were not going to just let the machine gather dust and mildew.
Those who decided to stick with it and make a go with the Jr. were not completely left out in the cold, hardware wise. The hardware expansion market for the PCjr. had dwindled considerably, but there were companies still trying to fulfill the need for upgrades. Comapnies like Racore and Legacy and eventually PC Enterprises stepped up to offer an assortment of sidecars, cartridges, peripherals and expansion chassis to expand the capabilities of the comparatively crippled PCjr. You could obtain a second floppy drive, a hard drive, 640KB of RAM, a clock/calendar chip, replacement keyboards, CPU accelerator boards, a coprocessor upgrade, Hayes compatible modems, and eventually SCSI adapters, VGA and Sound Blaster clones. They sold compatibility cartridges that would fool software programs into thinking they were running on a PC. In short, you could upgrade your jr. almost as much as you could a PC or XT, but you had to do your research and purchase via mail order. However, these expansions were sold to a limited market and pricey, often pricier than comparative PC expansions. Also, with the monitor, sidecars, expansion chassis and many adapters and cables to connect to standard PC peripherals, the PCjr. would eventually begin to look like a big "L" shaped spider. PC Enterprises was in business the longest, finally closing its doors around the turn of the century.
For software, after 1985 things became really iffy for PCjr. owners. Most PC game software (except Sierra), either only supported PC CGA, Hercules or EGA graphics and PC Speaker sound, and most of it required 256KB or more. That software was indifferent to the PCjr. but if DOS-based, often would run if the PCjr had sufficient memory. All IBM PCs and most clones by 1985 came with 256KB. While the Tandy 1000 also only came with 128KB, it was designed for 640KB, the PCjr. was designed for 128KB. While Tandy's memory expansions were pricey at first, they also added DMA, a widely appreciated feature which was difficult to implement on the PCjr.
For software now supporting Tandy graphics and sound, often the extra effort at trying to support a "dead" computer meant that developers did not bother, even though IBM made half a million PCjr.s and sold 400,000 of them. Many developers did not have a PCjr. to test with. Even though the IBM PCjr. and the Tandy 1000 operate almost identically, there are enough differences that making sure that software ran reliably on both was a real headache. Besides, with DMA, a relocatable video buffer, no extra latency between video and RAM access in the first 128KB, PC compatible floppy I/O addresses, two floppy drive support, easier hard disk support, better keyboard, more standard peripheral ports and faster Tandys, what time-pressed programmer wanted even to test their code on a PCjr.?
Here are some of the issues facing the programmer who wanted to make his software, designed on a Tandy 1000, work on a PCjr. :
Memory Issues - In the IBM PCjr., the first 128KB of memory is governed by the video controller. This was done to simplify the design, eliminate the need for a DMA controller or separate circuitry to refresh the DRAM. The video controller would read all the RAM sixty times a second. However, this came at the expense of performance, as the video circuitry and the CPU had to contend with each other's access to RAM. Any RAM added via sidecar above 128KB had its own refresh controller and thus accesses to and from this RAM was as fast as a PC. So in the PCjr. there was a combination of "slow" RAM and "fast" RAM. Tandy implemented this same feature in its machine, but unlike the PCjr, Tandy made sure that there was no adverse impact on performance. So there is no "slow" and "fast" distinction in a Tandy, and any programmer desiring a consistent speed for their operations would have this headache.
The addition of memory above 128KB on a PCjr. does not give you as much memory as you would think. Adding more memory requires a DOS memory driver to configure the system to use the memory. Otherwise programs will not recognize or use it (with the rare exception of the Flight Simulator II booter.) The memory driver will set the free RAM from 128KB upward and leave the slow first 128KB to the Video, BIOS, DOS and may configure the rest as a RAM disk. A Tandy does not require this and can have more free RAM available to programs because its DOS does not need memory drivers or treat the first 128KB as non-user memory. Generally speaking, a Tandy without the 768 RAM upgrade will report a maximum of 624KB available to DOS programs while a PCjr. will report 608KB.
Video Issues - Both the Tandy 1000 and IBM PCjr. have a video memory window at B8000 to maintain CGA compatibility. Until the (upgraded) Tandy 1000 TX, there would be no actual memory in this area, it would redirect to the designated address of the main memory where the video buffer was located. However, IBM only provided a 16KB window at B8000-BBFFF while Tandy allowed access to a 32KB window at B8000-BFFFF. For the 160x200x16 mode or any of the CGA modes, this made no difference because one screen's worth of these graphics modes use 16KB. The 320x200x16 and 640x200x4 modes require a 32KB buffer. On any Tandy 1000, the programmer could use the B8000 window and not have to bother about figuring out where the video buffer really began. If the programmer made no changes to his code for the PCjr., only every other line would be visible. This gave rise to the "Tandy-mod" for the PCjr., where a chip would be soldered into the PCjr. to show all the lines. To support the PCjr., the programmer would have to write directly to the portion of memory which the video circuitry was using when using a 32KB mode.
The IBM PCjr. had a chip called the Video Gate Array (VGA) that controlled the graphics modes in conjunction with the MC6845. Tandy implemented this VGA, but its implementation was a little different. The PCjr.'s VGA had an address and data write register at port 3DA, with the write mode switched by a read to port 3DA. Tandy must have decided this was inefficient, so address writes in its VGA went to port 3DA and data writes to port 3DD. Any program that writes to the VGA must be aware of this difference.
Moreover, the actual registers that controlled the modes were implemented differently in the Tandy and PCjr. In the Tandy, to restore compatibility with the PC CGA, Tandy's VGA supports I/O ports 3D8 (Mode Control) and 3D9 (Color Select). The PCjr. does not implement these ports, and handles all of the Mode Control functions in the VGA internal ports, most important is port 00. Tandy's VGA does not implement port 00 in its VGA, instead relying mainly on 3D9. The differences in the VGA ports (and there are others) is another headache for the programmer who wants to directly program the graphics registers.
Floppy Drive Issues - Tandy followed the IBM PC and used addresses 3F0-3F7 to communicate with the Floppy Disk Controller. The IBM PCjr. uses addresses F0-F7. While software using DOS or BIOS routines should not care, since using DOS or BIOS takes care of disk transfers, programs accessing the disk controller directly would need to be modified for the PCjr. Booter programs and copy protection routines, important in the 1980s, frequently accessed the hardware directly to obfuscate the code or to improve performance. Moreover, the PCjr. had no provision for DMA while every Tandy 1000 had it or could be upgraded to it. DMA made for faster disk transfers, and timing critical disk functions would be affected by this.
Keyboard Issues - Tandy uses a 90-key keyboard, and its raw scan codes are identical or similar to the 83-key IBM PC keyboard, except for those keys unique to the Tandy keyboard. The 62-key PCjr. keyboard with its different arrangement uses different scancodes for the keys that exist on the 83-key PC keyboard. This is typically with the Fn key, and the scancode pair is then translated by the BIOS for the PC scancode. Tandy's keyboard uses IRQ1, like the PC's, while the PCjr's keyboard uses NMI. All these make for issues with low level keyboard reading routines.
Even with these issues, there was considerable life left in the old Peanut. It was quite possible to purchase a PCjr. in 1987, often at a heavily discounted price. IBM provided some support for the Jr. until that time at least.
In the 1980s, there were often users groups created in a geographical area dedicated to a particular home computer. Perhaps spurred in part by the abandonment of the platform, PCjr. users groups were formed in several parts of the country and operated for many years, some well into the 1990s. The most famous ones were the Eugene, Oregon PCjr. Users Group and the Metro PCjr. Users Group in the metro-Washington D.C. area. These groups provided advice, support and meetings where PCjr. owners could trade the latest news and gossip, review hardware and software and try to solve problems. The groups also had their own Bulletin Board Services (BBSs), where people could post on message boards and download patches and programs. Newsletters in print and on disk were also a feature of these clubs.
Many issues of one particular newsletter, the jr Newsletter, are available online. The jr Newsletter ran from June 1986 to at least October 1993. This newsletter provided tips, fixes and advertising for PCjr. specific peripherals. IBM's BBS also had PCjr. specific files for a long time. Some games, sometimes unintentionally, still supported PCjr. graphics and sound or were hacked to do so. Electronic Arts, for example, included a PCjr. patch for Starflight made by a PCjr. owner who submitted it to the company.
The PCjr.'s life was not totally brought to an abrupt end in March 1985, despite the views of many at the time and since. Extending the lifespan of the machine required the user to become intimately familiar with this machine's hardware and the software available for it. It is an important example of how ordinary people, despite mainstream indifference (and often contempt) and market forces which should have condemned the machine into a nearly-forgotten footnote of computing history, allowed a unique computer to achieve some measure of the hoped-for potential intended for the machine in 1983.
March 18, 1985 - IBM announced that it was discontinuing manufacturing the PCjr. system in April, intending to sell the remaining stock at a discount and not release any further peripherals or upgrades for the system. It would continue to service the machines within their warranty period.
The reasons for the PCjr.'s failure have been well-documented elsewhere. Had it not been discontinued, it would have probably have been sold until April 1, 1987, when the rest of the PC family was retired to make way for the PS/2 machines. However, the public apathy for the machine a result of the machine's high price, competition from other home computers, miserable chicklet keyboard, incompatibilities with PC software and the high cost of the upgrades meant that despite IBM's weight in the industry, it had to be withdrawn.
What happens when a buyer, believing in the solid name of IBM, buys a computer whose lifespan had just been reduced from years to months or even days? Most PCjr.s had been sold during the 1984 Christmas season. IBM not only abandoned the PCjr, but also abandoned its family and entertainment software products. With one exception, the last games IBM marketed for the PC or PCjr. were released in 1985. While it tried to put on a brave face by announcing a new and improved successor to the Jr., nothing came of it in the United States. The closest thing to a PCjr. replacement was the PC JX, essentially the next generation of PCjr. released only in Japan, Australia and New Zealand. It sold very poorly. IBM would not return to the market with a home-friendly machine until the PS/1 in 1990.
When IBM announced the PCjr was being discontinued, software developers were the first to jump ship. Bill Gates' Microsoft, which had heavily supported the Jr. with the Microsoft Booster sidecar and an updated Flight Simulator 2.0 which took advantage of the Jr.'s capabilities, eliminated its PCjr. development almost overnight. Spinnaker Software tried to burn its PCjr.s in a great bonfire, only to find out that they wouldn't burn!. Sierra On-line quickly focused its efforts to the PC and the Tandy platform, and many other game companies did the same. Any software companies who had not yet jumped on the PCjr. bandwagon certainly were not about to start supporting the machine once its discontinuance had been announced.
All PCjr. magazines had already folded by December, 1984, in response to the weak market for the PCjr. before the 1984 Christmas season. Peter Norton had written a trilogy of how-to books for the PCjr., Discovering the PCjr, Exploring the PCjr and Mastering the PCjr, but the third book was never released due to the discontinuing of the PCjr.
Every game that had support for the PCjr.'s graphics and sound, as identified in an earlier post on this blog, were all self-booting games released in 1984 or 1985. In 1986, games were more and more beginning to run in DOS and very few advertised PCjr. support. Any DOS games from 1986 or later that did run on the PCjr., with expanded memory, tended to treat it just like an IBM PC with CGA. Sierra Online was the main exception, because its AGI engine had been originally built on the PCjr. and thus it was no great effort to make sure its games continued to run on it. All Sierra AGI DOS games thus will work on the PCjr. Some of the AGI booters as marketed by Tandy may not. So will Thexder, 3-D Helicopter Simulator, later versions of Silpheed but later Game Arts/Falcom ports will not. Early versions of King's Quest IV, Leisure Suit Larry II and perhaps Police Quest II specifically did not run on the PCjr., but later versions of these SCI engine games and other 16-color SCI0 engine games would run on a PCjr.
For the home user that had almost certainly dropped over $1,000 on the PCjr., it was not so easy to discard the computer like a bad memory as the mainstream industry had done. Some undoubtedly felt that their purchase was now useless and consigned it to the attic or the basement. Others probably only used it until they could afford something better. But for some of the owners of the approximately 500,000 PCjr.s IBM shipped out of its factories, they were not going to just let the machine gather dust and mildew.
Those who decided to stick with it and make a go with the Jr. were not completely left out in the cold, hardware wise. The hardware expansion market for the PCjr. had dwindled considerably, but there were companies still trying to fulfill the need for upgrades. Comapnies like Racore and Legacy and eventually PC Enterprises stepped up to offer an assortment of sidecars, cartridges, peripherals and expansion chassis to expand the capabilities of the comparatively crippled PCjr. You could obtain a second floppy drive, a hard drive, 640KB of RAM, a clock/calendar chip, replacement keyboards, CPU accelerator boards, a coprocessor upgrade, Hayes compatible modems, and eventually SCSI adapters, VGA and Sound Blaster clones. They sold compatibility cartridges that would fool software programs into thinking they were running on a PC. In short, you could upgrade your jr. almost as much as you could a PC or XT, but you had to do your research and purchase via mail order. However, these expansions were sold to a limited market and pricey, often pricier than comparative PC expansions. Also, with the monitor, sidecars, expansion chassis and many adapters and cables to connect to standard PC peripherals, the PCjr. would eventually begin to look like a big "L" shaped spider. PC Enterprises was in business the longest, finally closing its doors around the turn of the century.
For software, after 1985 things became really iffy for PCjr. owners. Most PC game software (except Sierra), either only supported PC CGA, Hercules or EGA graphics and PC Speaker sound, and most of it required 256KB or more. That software was indifferent to the PCjr. but if DOS-based, often would run if the PCjr had sufficient memory. All IBM PCs and most clones by 1985 came with 256KB. While the Tandy 1000 also only came with 128KB, it was designed for 640KB, the PCjr. was designed for 128KB. While Tandy's memory expansions were pricey at first, they also added DMA, a widely appreciated feature which was difficult to implement on the PCjr.
For software now supporting Tandy graphics and sound, often the extra effort at trying to support a "dead" computer meant that developers did not bother, even though IBM made half a million PCjr.s and sold 400,000 of them. Many developers did not have a PCjr. to test with. Even though the IBM PCjr. and the Tandy 1000 operate almost identically, there are enough differences that making sure that software ran reliably on both was a real headache. Besides, with DMA, a relocatable video buffer, no extra latency between video and RAM access in the first 128KB, PC compatible floppy I/O addresses, two floppy drive support, easier hard disk support, better keyboard, more standard peripheral ports and faster Tandys, what time-pressed programmer wanted even to test their code on a PCjr.?
Here are some of the issues facing the programmer who wanted to make his software, designed on a Tandy 1000, work on a PCjr. :
Memory Issues - In the IBM PCjr., the first 128KB of memory is governed by the video controller. This was done to simplify the design, eliminate the need for a DMA controller or separate circuitry to refresh the DRAM. The video controller would read all the RAM sixty times a second. However, this came at the expense of performance, as the video circuitry and the CPU had to contend with each other's access to RAM. Any RAM added via sidecar above 128KB had its own refresh controller and thus accesses to and from this RAM was as fast as a PC. So in the PCjr. there was a combination of "slow" RAM and "fast" RAM. Tandy implemented this same feature in its machine, but unlike the PCjr, Tandy made sure that there was no adverse impact on performance. So there is no "slow" and "fast" distinction in a Tandy, and any programmer desiring a consistent speed for their operations would have this headache.
The addition of memory above 128KB on a PCjr. does not give you as much memory as you would think. Adding more memory requires a DOS memory driver to configure the system to use the memory. Otherwise programs will not recognize or use it (with the rare exception of the Flight Simulator II booter.) The memory driver will set the free RAM from 128KB upward and leave the slow first 128KB to the Video, BIOS, DOS and may configure the rest as a RAM disk. A Tandy does not require this and can have more free RAM available to programs because its DOS does not need memory drivers or treat the first 128KB as non-user memory. Generally speaking, a Tandy without the 768 RAM upgrade will report a maximum of 624KB available to DOS programs while a PCjr. will report 608KB.
Video Issues - Both the Tandy 1000 and IBM PCjr. have a video memory window at B8000 to maintain CGA compatibility. Until the (upgraded) Tandy 1000 TX, there would be no actual memory in this area, it would redirect to the designated address of the main memory where the video buffer was located. However, IBM only provided a 16KB window at B8000-BBFFF while Tandy allowed access to a 32KB window at B8000-BFFFF. For the 160x200x16 mode or any of the CGA modes, this made no difference because one screen's worth of these graphics modes use 16KB. The 320x200x16 and 640x200x4 modes require a 32KB buffer. On any Tandy 1000, the programmer could use the B8000 window and not have to bother about figuring out where the video buffer really began. If the programmer made no changes to his code for the PCjr., only every other line would be visible. This gave rise to the "Tandy-mod" for the PCjr., where a chip would be soldered into the PCjr. to show all the lines. To support the PCjr., the programmer would have to write directly to the portion of memory which the video circuitry was using when using a 32KB mode.
The IBM PCjr. had a chip called the Video Gate Array (VGA) that controlled the graphics modes in conjunction with the MC6845. Tandy implemented this VGA, but its implementation was a little different. The PCjr.'s VGA had an address and data write register at port 3DA, with the write mode switched by a read to port 3DA. Tandy must have decided this was inefficient, so address writes in its VGA went to port 3DA and data writes to port 3DD. Any program that writes to the VGA must be aware of this difference.
Moreover, the actual registers that controlled the modes were implemented differently in the Tandy and PCjr. In the Tandy, to restore compatibility with the PC CGA, Tandy's VGA supports I/O ports 3D8 (Mode Control) and 3D9 (Color Select). The PCjr. does not implement these ports, and handles all of the Mode Control functions in the VGA internal ports, most important is port 00. Tandy's VGA does not implement port 00 in its VGA, instead relying mainly on 3D9. The differences in the VGA ports (and there are others) is another headache for the programmer who wants to directly program the graphics registers.
Floppy Drive Issues - Tandy followed the IBM PC and used addresses 3F0-3F7 to communicate with the Floppy Disk Controller. The IBM PCjr. uses addresses F0-F7. While software using DOS or BIOS routines should not care, since using DOS or BIOS takes care of disk transfers, programs accessing the disk controller directly would need to be modified for the PCjr. Booter programs and copy protection routines, important in the 1980s, frequently accessed the hardware directly to obfuscate the code or to improve performance. Moreover, the PCjr. had no provision for DMA while every Tandy 1000 had it or could be upgraded to it. DMA made for faster disk transfers, and timing critical disk functions would be affected by this.
Keyboard Issues - Tandy uses a 90-key keyboard, and its raw scan codes are identical or similar to the 83-key IBM PC keyboard, except for those keys unique to the Tandy keyboard. The 62-key PCjr. keyboard with its different arrangement uses different scancodes for the keys that exist on the 83-key PC keyboard. This is typically with the Fn key, and the scancode pair is then translated by the BIOS for the PC scancode. Tandy's keyboard uses IRQ1, like the PC's, while the PCjr's keyboard uses NMI. All these make for issues with low level keyboard reading routines.
Even with these issues, there was considerable life left in the old Peanut. It was quite possible to purchase a PCjr. in 1987, often at a heavily discounted price. IBM provided some support for the Jr. until that time at least.
In the 1980s, there were often users groups created in a geographical area dedicated to a particular home computer. Perhaps spurred in part by the abandonment of the platform, PCjr. users groups were formed in several parts of the country and operated for many years, some well into the 1990s. The most famous ones were the Eugene, Oregon PCjr. Users Group and the Metro PCjr. Users Group in the metro-Washington D.C. area. These groups provided advice, support and meetings where PCjr. owners could trade the latest news and gossip, review hardware and software and try to solve problems. The groups also had their own Bulletin Board Services (BBSs), where people could post on message boards and download patches and programs. Newsletters in print and on disk were also a feature of these clubs.
Many issues of one particular newsletter, the jr Newsletter, are available online. The jr Newsletter ran from June 1986 to at least October 1993. This newsletter provided tips, fixes and advertising for PCjr. specific peripherals. IBM's BBS also had PCjr. specific files for a long time. Some games, sometimes unintentionally, still supported PCjr. graphics and sound or were hacked to do so. Electronic Arts, for example, included a PCjr. patch for Starflight made by a PCjr. owner who submitted it to the company.
The PCjr.'s life was not totally brought to an abrupt end in March 1985, despite the views of many at the time and since. Extending the lifespan of the machine required the user to become intimately familiar with this machine's hardware and the software available for it. It is an important example of how ordinary people, despite mainstream indifference (and often contempt) and market forces which should have condemned the machine into a nearly-forgotten footnote of computing history, allowed a unique computer to achieve some measure of the hoped-for potential intended for the machine in 1983.
Saturday, January 25, 2014
Meet another Video Card - The Prolink Systems, Inc. MVGA-AVGA4VL VLB Card
Recently, I found a card nearly identical to this one from http://www.vgamuseum.info/ in an old system and decided to try it out :
I have tried it out and have come to like it more than my S3-805 based Diamond Stealth VLB card. Its slightly faster, more compatible, supports more RAM and video modes. I have no complaints about the DOS video quality. It uses the Cirrus Logic CL-GD5429, which supports true color modes and Windows acceleration. It uses memory-mapped I/O for increased speed. It comes with 1MB of RAM and can be upgraded to 2MB with two 256x16-70ns Fast Page Mode SOJ chips.
The card has two jumpers, JP1 when closed allows for IRQ2/9 usage, and JP2 when closed allows for operation when the bus speed is greater than 33MHz. The data sheet for the family of VGA controllers, for which this is perhaps its most advanced member, is readily available. It reports that the card's BIOS supports the following SVGA/VESA modes :
14h - 132x25T
54h/10Ah - 132x43T
55h/109h - 132x25T
5Eh/100h - 640x400x256
5Fh/101h - 640x480x256
58h,6Ah/102h - 800x600x16
5Ch/103h - 800x600x256
5Dh/104h - 1024x768x16i/p
60h/105h - 1024x768x256i/p
6Ch/106h - 1280x1024x16i
6Dh/107h - 1280x1024x256i*
66h/110h - 640x480x32K
64h/111h - 640x480x64K
71h/112h - 640x480x16M
67h/113h - 800x600x32K
65h/114h - 800x600x64K
68h/116h - 1024x768x32Ki
74h/117h - 1024x768x64Ki
* - Requires 2MB of video RAM.
i - Interlaced Mode
T - Text Mode
This card has better compatibility that the S3 card. EGA compatibility is very good. Commander Keen 4, 5 and 6 and Dangerous Dave do not need the SVGA compatibility switch, Keen 1-3 and Keen Dreams do not have scrolling problems, SimCity's text fonts look correct in the EGA high resolution mode and there are almost no flickering lines in the Silpheed intro.
VGA and SVGA compatibility is outstanding. System Shock CD allows the 640x400 mode to be used, although the speed on my 486DX2/66 is not the speed at which I would like to play the game. It has no problem with games with unusual Mode-X VGA modes like 320x240 (Epic Pinball), 320x400 (System Shock CD), 320x199 (Jazz Jackrabbit), 360x350 (Pinball Illusions), 312x200 (Prehistorik 2), 320x184 (Jurassic Park) and 320x350 (Pinball Fantasies). It supports SVGA 640x480 and 800x600 resolution modes in Pinball Illusions perfectly. It works with Prehistorik 1 & 2, Duff and Lollypop, both of which use a tweaked 320x200 mode. It is among the supported 640x400 SVGA modes of Microsoft Flight Simulator 5.x. It can even support the old Paradise SVGA 800x600 mode that Wonderland uses. DOOM and DOOM II run just as fast on this card as the S3 card.
The DOS refresh utility is called CLMODE and there are Windows 3.1 drivers for the chipset. The BIOS revision on my card is 1.00A.
Prompted by a friend of mine, I decided to perform a more demanding test, namely testing this card's support with UniVBE 5.3 and 6.53. UniVBE 6.53 reports that Linear Frame Buffer will not work reliably with CL 542x chips with more than 14MB of RAM. Support for LFB was thus dsiabled in my system, but 16MB of system RAM is more useful in a 486 than 12MB of system RAM and LFB. With 8MB of RAM, UniVBE will install LFB at 14MB. UniVBE also reports that Multi Buffering and Virtual Scrolling are also available, but 8-bit DACs are not present.
The card supports lots and lots of modes with UniVBE loaded, so I didn't test each and every resolution at each and every frame rate. With UniVBE 5.3 all the modes I tested worked. There are more modes supported with UniVBE than with the card's BIOS, and UniVBE offers superior compatibility with some modes than the BIOS alone. UniVBE requires 13KB of RAM and it may not be possible to load it into upper memory, so it should only be used when necessary. Here are the list of modes the card supports with UniVBE loaded :
4-bit Banked Only
640x480
800x600
1024x768
8-bit Banked and Linear
320x200
320x240
400x300
320x400
320x480
512x384
640x350
640x400
640x480
800x600
1024x768
1152x864
15-bit Banked and Linear
320x200
320x240
400x300
320x400
320x480
512x384
640x350
640x400
640x480
800x600
16-bit Banked and Linear
320x200
320x240
400x300
320x400
320x480
512x384
640x350
640x400
640x480
800x600
24-bit Banked and Linear
320x200
320x240
400x300
320x400
320x480
512x384
640x350
640x400
640x480
32-bit Modes
None
I have tried it out and have come to like it more than my S3-805 based Diamond Stealth VLB card. Its slightly faster, more compatible, supports more RAM and video modes. I have no complaints about the DOS video quality. It uses the Cirrus Logic CL-GD5429, which supports true color modes and Windows acceleration. It uses memory-mapped I/O for increased speed. It comes with 1MB of RAM and can be upgraded to 2MB with two 256x16-70ns Fast Page Mode SOJ chips.
The card has two jumpers, JP1 when closed allows for IRQ2/9 usage, and JP2 when closed allows for operation when the bus speed is greater than 33MHz. The data sheet for the family of VGA controllers, for which this is perhaps its most advanced member, is readily available. It reports that the card's BIOS supports the following SVGA/VESA modes :
14h - 132x25T
54h/10Ah - 132x43T
55h/109h - 132x25T
5Eh/100h - 640x400x256
5Fh/101h - 640x480x256
58h,6Ah/102h - 800x600x16
5Ch/103h - 800x600x256
5Dh/104h - 1024x768x16i/p
60h/105h - 1024x768x256i/p
6Ch/106h - 1280x1024x16i
6Dh/107h - 1280x1024x256i*
66h/110h - 640x480x32K
64h/111h - 640x480x64K
71h/112h - 640x480x16M
67h/113h - 800x600x32K
65h/114h - 800x600x64K
68h/116h - 1024x768x32Ki
74h/117h - 1024x768x64Ki
* - Requires 2MB of video RAM.
i - Interlaced Mode
T - Text Mode
This card has better compatibility that the S3 card. EGA compatibility is very good. Commander Keen 4, 5 and 6 and Dangerous Dave do not need the SVGA compatibility switch, Keen 1-3 and Keen Dreams do not have scrolling problems, SimCity's text fonts look correct in the EGA high resolution mode and there are almost no flickering lines in the Silpheed intro.
VGA and SVGA compatibility is outstanding. System Shock CD allows the 640x400 mode to be used, although the speed on my 486DX2/66 is not the speed at which I would like to play the game. It has no problem with games with unusual Mode-X VGA modes like 320x240 (Epic Pinball), 320x400 (System Shock CD), 320x199 (Jazz Jackrabbit), 360x350 (Pinball Illusions), 312x200 (Prehistorik 2), 320x184 (Jurassic Park) and 320x350 (Pinball Fantasies). It supports SVGA 640x480 and 800x600 resolution modes in Pinball Illusions perfectly. It works with Prehistorik 1 & 2, Duff and Lollypop, both of which use a tweaked 320x200 mode. It is among the supported 640x400 SVGA modes of Microsoft Flight Simulator 5.x. It can even support the old Paradise SVGA 800x600 mode that Wonderland uses. DOOM and DOOM II run just as fast on this card as the S3 card.
The DOS refresh utility is called CLMODE and there are Windows 3.1 drivers for the chipset. The BIOS revision on my card is 1.00A.
Prompted by a friend of mine, I decided to perform a more demanding test, namely testing this card's support with UniVBE 5.3 and 6.53. UniVBE 6.53 reports that Linear Frame Buffer will not work reliably with CL 542x chips with more than 14MB of RAM. Support for LFB was thus dsiabled in my system, but 16MB of system RAM is more useful in a 486 than 12MB of system RAM and LFB. With 8MB of RAM, UniVBE will install LFB at 14MB. UniVBE also reports that Multi Buffering and Virtual Scrolling are also available, but 8-bit DACs are not present.
The card supports lots and lots of modes with UniVBE loaded, so I didn't test each and every resolution at each and every frame rate. With UniVBE 5.3 all the modes I tested worked. There are more modes supported with UniVBE than with the card's BIOS, and UniVBE offers superior compatibility with some modes than the BIOS alone. UniVBE requires 13KB of RAM and it may not be possible to load it into upper memory, so it should only be used when necessary. Here are the list of modes the card supports with UniVBE loaded :
4-bit Banked Only
640x480
800x600
1024x768
8-bit Banked and Linear
320x200
320x240
400x300
320x400
320x480
512x384
640x350
640x400
640x480
800x600
1024x768
1152x864
15-bit Banked and Linear
320x200
320x240
400x300
320x400
320x480
512x384
640x350
640x400
640x480
800x600
16-bit Banked and Linear
320x200
320x240
400x300
320x400
320x480
512x384
640x350
640x400
640x480
800x600
24-bit Banked and Linear
320x200
320x240
400x300
320x400
320x480
512x384
640x350
640x400
640x480
32-bit Modes
None
Tuesday, January 21, 2014
SimCity for DOS - The Swiss Army Knife of Video Mode Support
The original version of SimCity for DOS, 1.02, supported a variety of video modes. Here is a screenshot of the title screen and an in-game screenshot of the Detroit scenario with a maxed out main window for each mode:
CGA Graphics Mode 06H 640x200 Monochrome :
Hercules Graphics Mode 720x348 Monochrome :
Tandy Graphics Adapter Graphics Mode 09H 320x200x16 / EGA Graphics Mode 0DH 320x200x16 :
(note : screenshots are 200% of original)
EGA Graphics Mode 0FH 640x350 Monochrome :
EGA Graphics Mode 10H 640x350x16 :
Version 1.07 of SimCity added the following :
MCGA/VGA Graphics Mode 11H 640x480 Monochrome :
SimCity Classic for DOS dropped the CGA and Tandy/EGA 320x200 support, but added the following :
MCGA/VGA Graphics Mode 12H 640x480x16 :
MCGA/VGA Graphics Mode 13H 320x200x256 :
(note : screenshots are 200% of original)
The modes common to SimCity and SimCity Classic look virtually identical in-game, but the title screens are different. Here they are :
Finally SimCity Enhanced CD-ROM for DOS only supported the following :
SVGA/VESA Graphics Mode 101H 640x480x256 :
Between the various incarnations of the same basic SimCity game for DOS, the game supported every major DOS video standard and an unusually wide variety of Graphics Modes. Sound support was also very broad. The original SimCity only supported sound effects, and supported the only available digitized sound hardware available in 1989 that would output digital samples without seriously compromising performance, the Tandy DAC and the Covox Sound Master. Unfortunately, not until SimCity Classic did the game support Sound Blaster cards.
CGA Graphics Mode 06H 640x200 Monochrome :
Hercules Graphics Mode 720x348 Monochrome :
Tandy Graphics Adapter Graphics Mode 09H 320x200x16 / EGA Graphics Mode 0DH 320x200x16 :
(note : screenshots are 200% of original)
EGA Graphics Mode 10H 640x350x16 :
Version 1.07 of SimCity added the following :
MCGA/VGA Graphics Mode 11H 640x480 Monochrome :
SimCity Classic for DOS dropped the CGA and Tandy/EGA 320x200 support, but added the following :
MCGA/VGA Graphics Mode 12H 640x480x16 :
MCGA/VGA Graphics Mode 13H 320x200x256 :
(note : screenshots are 200% of original)
The modes common to SimCity and SimCity Classic look virtually identical in-game, but the title screens are different. Here they are :
 |
| Hercules Graphics Mode 720x348 Monochrome |
 |
| EGA Mode 0FH 640x350 Monochrome |
 |
| EGA Mode 10H 640x350x16 |
 |
| MCGA/VGA Mode 11H 640x480 Monochrome : |
Finally SimCity Enhanced CD-ROM for DOS only supported the following :
SVGA/VESA Graphics Mode 101H 640x480x256 :
 |
| Note : Title Screen uses MCGA/VGA Graphics Mode 13H 320x200x256 (screenshot is 200% of original size) |
Between the various incarnations of the same basic SimCity game for DOS, the game supported every major DOS video standard and an unusually wide variety of Graphics Modes. Sound support was also very broad. The original SimCity only supported sound effects, and supported the only available digitized sound hardware available in 1989 that would output digital samples without seriously compromising performance, the Tandy DAC and the Covox Sound Master. Unfortunately, not until SimCity Classic did the game support Sound Blaster cards.
Thursday, January 16, 2014
A Tale of Two Configs
In an MS-DOS system with a 386 or better processor, when it comes to configuring the OS, typically the issue is whether to load or not to load EMM386.EXE. Loading EMM386.EXE puts the machine into Virtual 8086 mode, and using EMM386 to make expanded memory available substantially decreases the amount of upper memory available to load drivers. Fortunately, you need not make expanded memory available unless a program requires it. EMM386 is used to make upper memory available for loading device drivers, and without it, those drivers must be loaded in conventional memory. If those drivers cause the amount of free memory to be lower than the amount a program requires, the program will not load. Use of EMM386.EXE is utterly incompatible certain programs like Ultima VII Parts I & II. On the other hand, many other contemporary games require expanded memory to support sound. Other games don't care either way.
HIMEM.SYS, which provides access to all the memory above 640KB, is essential to load and should always be loaded . EMM386.EXE requires HIMEM.SYS to be loaded first. HIMEM.SYS provides access to the high memory area and allows DOS to load most of itself there, significantly cutting down on the conventional memory DOS would otherwise take up. For programs that require or can use it, it also makes extended memory available.
For MS-DOS 6.22, a typical 486 computer setup with a mouse, CD-ROM drive and a Sound Blaster 16 may have a CONFIG.SYS like this :
DEVICE=C:\DOS\HIMEM.SYS /TESTMEM:OFF
DEVICE=C:\DOS\EMM386.EXE I=B700-BFFF FRAME=E000
DOS=HIGH,UMB
FILES=40
BUFFERS=30
DEVICEHIGH=C:\DRIVERS\VIDE-CDD.SYS /D:MSCD001
and a typical AUTOEXEC.BAT file may look like this :
@ECHO OFF
PROMPT $P$G
PATH=C:\;C:\DOS;C:\SB16
SET BLASTER=A220 I5 D1 H5 P330 T6
SET SOUND=C:\SB16
SET MIDI=SYNTH:1 MAP:E
C:\SB16\DIAGNOSE /S
C:\SB16\MIXERSET /P /Q
LH C:\DOS\MSCDEX.EXE /D:MSCD001
LH CTMOUSE.EXE
My 486 uses roughly these lines in its AUTOEXEC.BAT and CONFIG.SYS. Lets describe what each line does, starting with the first line of CONFIG.SYS :
1. DEVICE=C:\DOS\HIMEM.SYS /TESTMEM:OFF
HIMEM.SYS is a Device Driver that must be loaded in CONFIG.SYS on startup. The config.sys command to load a device driver is DEVICE=. The full path where a device driver can be found must be included. HIMEM.SYS is primarily used to allow DOS to access the High Memory Area, but it also functions as an eXtended Memory Manager (XMM), giving your programs access to eXtended Memory Specification (XMS) memory. The /TESTMEM:OFF parameter bypasses a memory test that adds to the boot time. It is only useful on 286 or above systems.
2. DEVICE=C:\DOS\EMM386.EXE I=B700-BFFF FRAME=E000
EMM386.EXE can be used as a command or as a device driver. It's primary function is to act as an Expanded Memory Manager (EMM), giving your programs access to Expanded Memory Specification (EMS) memory. It also allows the creation of Upper Memory Blocks (UMB) in the Upper Memory Area (UMA). It is only useful on 386 or above systems. This device driver switches DOS from using real mode into Virtual 8086 mode, and some programs and games refuse to run in this mode. By removing the EMM386.EXE line, those programs that complained about EMM386.EXE and Virtual 8086 mode and refuse to load now will load.
3. DOS=HIGH,UMB
This line instructs DOS to load itself high, saving about 55K of Conventional Memory. It also instructs DOS to create UMBs. By loading drivers within UMBs, you can save Conventional Memory from being used. You must load HIMEM.SYS to use the HIGH parameter and EMM386.SYS to use the UMB parameter.
4. FILES=40
DOS's default maximum number of files being open at one time is 15, but some games require more. 40 is sufficient for just about any program.
5. BUFFERS=30
DOS's default maximum number of disk buffers to allocate is 8, but some games require more. 30 is sufficient for just about any program.
6. DEVICEHIGH=C:\DRIVERS\VIDE-CDD.SYS /D:MSCD001
DEVICEHIGH tells DOS to load a device driver into an UMB. It does not work with HIMEM.SYS or EMM386.EXE and will not work until both have been loaded.
VIDE-CDD.SYS is a generic IDE CD-ROM device driver that works with pretty much everything and only takes up 5KB of RAM. This device driver acts as an interpreter between DOS calls and the low level hardware. If the CD-ROM drive used a SCSI or proprietary interface, you would load another kind of driver here. You cannot actually use a CD-ROM as a drive until you load MSCDEX.EXE in AUTOEXEC.BAT The /D:MSCD001 is the name given to the CD-ROM and must be used with MSCDEX.EXE.
Alternatives to EMM386.EXE for UMBs
EMM386.EXE generally causes slightly decreased performance compared with an environment when it is not loaded. If you want to create UMBs without it, you will have to try to find a program that will work with your chipset. The program UMBPCI.SYS works with chipsets that provide a properly-functional PCI implementation. Pentium/586 and later systems are the mostly likely systems to work with UMBPCI.SYS. Programs that can work with 386 and 486 computers include HiRAM, URAM, DOSMAX and RDOSUMB. These programs only support specific motherboard chipsets, whereas EMM386.EXE is universally supported on PC compatible motherboard chipsets. I have a 486 SiS 85C471 chipset and URAM supports it, so I use that program. My CONFIG.SYS looks like this :
DEVICE=C:\DRIVERS\URAM.COM R=..........7777777777.... I Q
DEVICE=C:\DRIVERS\UMB.SYS C800-EFFF
DEVICEHIGH=C:\DOS\HIMEM.SYS /TESTMEM:OFF
DOS=HIGH,UMB
FILES=40
DEVICEHIGH=C:\DRIVERS\VIDE-CDD.SYS /D:MSCD001
My AUTOEXEC.BAT stays the same.
DOS 6.0 and above allows you to select your configuration properties via a boot-up menu system. The Help in DOS 6.0 and above gives examples on how to set up a menu so that one selection will give load EMM386 and a second selection will load URAM.
HIMEM.SYS, which provides access to all the memory above 640KB, is essential to load and should always be loaded . EMM386.EXE requires HIMEM.SYS to be loaded first. HIMEM.SYS provides access to the high memory area and allows DOS to load most of itself there, significantly cutting down on the conventional memory DOS would otherwise take up. For programs that require or can use it, it also makes extended memory available.
For MS-DOS 6.22, a typical 486 computer setup with a mouse, CD-ROM drive and a Sound Blaster 16 may have a CONFIG.SYS like this :
DEVICE=C:\DOS\HIMEM.SYS /TESTMEM:OFF
DEVICE=C:\DOS\EMM386.EXE I=B700-BFFF FRAME=E000
DOS=HIGH,UMB
FILES=40
BUFFERS=30
DEVICEHIGH=C:\DRIVERS\VIDE-CDD.SYS /D:MSCD001
and a typical AUTOEXEC.BAT file may look like this :
@ECHO OFF
PROMPT $P$G
PATH=C:\;C:\DOS;C:\SB16
SET BLASTER=A220 I5 D1 H5 P330 T6
SET SOUND=C:\SB16
SET MIDI=SYNTH:1 MAP:E
C:\SB16\DIAGNOSE /S
C:\SB16\MIXERSET /P /Q
LH C:\DOS\MSCDEX.EXE /D:MSCD001
LH CTMOUSE.EXE
My 486 uses roughly these lines in its AUTOEXEC.BAT and CONFIG.SYS. Lets describe what each line does, starting with the first line of CONFIG.SYS :
1. DEVICE=C:\DOS\HIMEM.SYS /TESTMEM:OFF
HIMEM.SYS is a Device Driver that must be loaded in CONFIG.SYS on startup. The config.sys command to load a device driver is DEVICE=. The full path where a device driver can be found must be included. HIMEM.SYS is primarily used to allow DOS to access the High Memory Area, but it also functions as an eXtended Memory Manager (XMM), giving your programs access to eXtended Memory Specification (XMS) memory. The /TESTMEM:OFF parameter bypasses a memory test that adds to the boot time. It is only useful on 286 or above systems.
2. DEVICE=C:\DOS\EMM386.EXE I=B700-BFFF FRAME=E000
EMM386.EXE can be used as a command or as a device driver. It's primary function is to act as an Expanded Memory Manager (EMM), giving your programs access to Expanded Memory Specification (EMS) memory. It also allows the creation of Upper Memory Blocks (UMB) in the Upper Memory Area (UMA). It is only useful on 386 or above systems. This device driver switches DOS from using real mode into Virtual 8086 mode, and some programs and games refuse to run in this mode. By removing the EMM386.EXE line, those programs that complained about EMM386.EXE and Virtual 8086 mode and refuse to load now will load.
3. DOS=HIGH,UMB
This line instructs DOS to load itself high, saving about 55K of Conventional Memory. It also instructs DOS to create UMBs. By loading drivers within UMBs, you can save Conventional Memory from being used. You must load HIMEM.SYS to use the HIGH parameter and EMM386.SYS to use the UMB parameter.
4. FILES=40
DOS's default maximum number of files being open at one time is 15, but some games require more. 40 is sufficient for just about any program.
5. BUFFERS=30
DOS's default maximum number of disk buffers to allocate is 8, but some games require more. 30 is sufficient for just about any program.
6. DEVICEHIGH=C:\DRIVERS\VIDE-CDD.SYS /D:MSCD001
DEVICEHIGH tells DOS to load a device driver into an UMB. It does not work with HIMEM.SYS or EMM386.EXE and will not work until both have been loaded.
VIDE-CDD.SYS is a generic IDE CD-ROM device driver that works with pretty much everything and only takes up 5KB of RAM. This device driver acts as an interpreter between DOS calls and the low level hardware. If the CD-ROM drive used a SCSI or proprietary interface, you would load another kind of driver here. You cannot actually use a CD-ROM as a drive until you load MSCDEX.EXE in AUTOEXEC.BAT The /D:MSCD001 is the name given to the CD-ROM and must be used with MSCDEX.EXE.
Alternatives to EMM386.EXE for UMBs
EMM386.EXE generally causes slightly decreased performance compared with an environment when it is not loaded. If you want to create UMBs without it, you will have to try to find a program that will work with your chipset. The program UMBPCI.SYS works with chipsets that provide a properly-functional PCI implementation. Pentium/586 and later systems are the mostly likely systems to work with UMBPCI.SYS. Programs that can work with 386 and 486 computers include HiRAM, URAM, DOSMAX and RDOSUMB. These programs only support specific motherboard chipsets, whereas EMM386.EXE is universally supported on PC compatible motherboard chipsets. I have a 486 SiS 85C471 chipset and URAM supports it, so I use that program. My CONFIG.SYS looks like this :
DEVICE=C:\DRIVERS\URAM.COM R=..........7777777777.... I Q
DEVICE=C:\DRIVERS\UMB.SYS C800-EFFF
DEVICEHIGH=C:\DOS\HIMEM.SYS /TESTMEM:OFF
DOS=HIGH,UMB
FILES=40
DEVICEHIGH=C:\DRIVERS\VIDE-CDD.SYS /D:MSCD001
My AUTOEXEC.BAT stays the same.
DOS 6.0 and above allows you to select your configuration properties via a boot-up menu system. The Help in DOS 6.0 and above gives examples on how to set up a menu so that one selection will give load EMM386 and a second selection will load URAM.
Tuesday, January 7, 2014
SID and DOS - Unlikely but True Bedfellows
The Commodore MOS 6581 Sound Interface Device, the SID, is one of the most famous synthesizer chips of all time. Its inclusion in the Commodore 64 and 128 computers helped to popularize computer music, mostly in Europe. Like virtually every other popular synthesizer chip of the time, it made its way to the PC in some form. Commodore had no problem in selling its chips to outside and technically rival companies. During its market life, the SID appeared in two different but related PC sound cards.
I. The Entertainer
In 1987, Microprose was looking for a way to enhance the sound of some of its DOS games. These games had access to superior audio hardware on the Commodore 64, Amiga and Atari ST, but the U.S. market was increasingly turning to the PCs. Unless a PC owner had a Tandy 1000 with its 3-voice chip, he had to settle for the PC speaker, as there was no popular and cheap sound expansion hardware available at the time. Since Microprose was still developing games for the Commodore 64, it seemed a natural idea to interface its SID chip with the PC. They advertised a PC sound board called The Entertainer, which included a 6581 and a joystick port on an ISA card. The only place this advertisement has ever been known to be seen is in an introduction screen and the readme to Gunship v429.04. The readme discusses the card with the following text :
Support has been added for the MicroProse Sound Board - "The Entertainer".
Presence of the board is detected by the program and enhanced sound is
automatically generated. Two additional control keys have been added to
control the volume. Ctrl-V turns the sound down one notch. Alt-V turns
it up. Just pressing "V" alone still toggles the sound on & off.
Microprose, according to the ad, sold the card directly and it was available through "your local retailer". Two high-level Microprose employees have been contacted and have no memory of selling the card at that time, so it probably was not actually sold and if it was sold, it did not sell well. The Entertainer card, if it was actually released by Microprose, was not supported by Microprose except in Gunship, v429.04 & 429.05 and Sid Meier's Pirates!, v432.1, 432.2 and 432.3. It may also be supported in 432.4, which was a DOS conversion of the floppy PC booter from a Best of Microprose CD. It was not known to have been intentionally supported in any other company's games. The in-game ad and the text quoted above was removed for Gunship v429.05, so it is very possible that this was an idea that Microprose considered but which it did not follow through to market.
II. The Innovation Sound Standard SSI-2001
Not too long after the Adlib Music Synthesizer Card began to be adopted by Sierra and other companies, and the idea (and more, as we will see) of a SID-on-a-chip card was resurrected by a company called Innovation. It released a card around April, 1989 called the Innovation Sound Standard SSI-2001, which is a 6581 plus a speed-adjustable gameport. It cost $129.00, but with discounts and coupons you could have purchased one for $69.00. It may have only been available directly through Innovation. Support was poor, the following games are the only ones known to support it :
Airball - SFX only, slow PC req'd
Bad Blood - Music only
Battle Chess II - SFX only
BattleTech: The Crescent Hawks' Revenge
F-19 Stealth Fighter
Falcon A.T.
Harpoon
Joe Montana Football - SFX only
Lord of the Rings Volume 1 - Floppy version only, SFX only
Red Storm Rising
Super Jeopardy
Ultima VI - Music only
Windwalker
I have verified that all the above games work with the Innovation emulation in DOSBox except for Super Jeopardy. While Mobygames claims that Lexi-Cross and Castles and its expansion Castles : The Northern Campaign support Innovation, I have personally verified that these programs do not have install options for Innovation, despite what the system requirements labels on their boxes may indicate. Asterix: Operation Getafix is also listed by Mobygames as supporting the Innovation, but it actually supports the InterSound MDO, a Covox Speech Thing-like device released by the French gaming company Coktel Vision.
The best informational article describing the card can be found in Computer Gaming World, Number 63, September 1989. It mentions that only Origin and possibly Mindscape were planning to support the card. Interplay and Microprose eventually did with more than one title. Although Commodore made PC clones, it never put its sound chips in them. The SSI-2001 is the only non-Commodore product I am aware of that used the SID chip prior to Commodore's failure in 1994.
I. The Entertainer
In 1987, Microprose was looking for a way to enhance the sound of some of its DOS games. These games had access to superior audio hardware on the Commodore 64, Amiga and Atari ST, but the U.S. market was increasingly turning to the PCs. Unless a PC owner had a Tandy 1000 with its 3-voice chip, he had to settle for the PC speaker, as there was no popular and cheap sound expansion hardware available at the time. Since Microprose was still developing games for the Commodore 64, it seemed a natural idea to interface its SID chip with the PC. They advertised a PC sound board called The Entertainer, which included a 6581 and a joystick port on an ISA card. The only place this advertisement has ever been known to be seen is in an introduction screen and the readme to Gunship v429.04. The readme discusses the card with the following text :
Support has been added for the MicroProse Sound Board - "The Entertainer".
Presence of the board is detected by the program and enhanced sound is
automatically generated. Two additional control keys have been added to
control the volume. Ctrl-V turns the sound down one notch. Alt-V turns
it up. Just pressing "V" alone still toggles the sound on & off.
Microprose, according to the ad, sold the card directly and it was available through "your local retailer". Two high-level Microprose employees have been contacted and have no memory of selling the card at that time, so it probably was not actually sold and if it was sold, it did not sell well. The Entertainer card, if it was actually released by Microprose, was not supported by Microprose except in Gunship, v429.04 & 429.05 and Sid Meier's Pirates!, v432.1, 432.2 and 432.3. It may also be supported in 432.4, which was a DOS conversion of the floppy PC booter from a Best of Microprose CD. It was not known to have been intentionally supported in any other company's games. The in-game ad and the text quoted above was removed for Gunship v429.05, so it is very possible that this was an idea that Microprose considered but which it did not follow through to market.
II. The Innovation Sound Standard SSI-2001
Not too long after the Adlib Music Synthesizer Card began to be adopted by Sierra and other companies, and the idea (and more, as we will see) of a SID-on-a-chip card was resurrected by a company called Innovation. It released a card around April, 1989 called the Innovation Sound Standard SSI-2001, which is a 6581 plus a speed-adjustable gameport. It cost $129.00, but with discounts and coupons you could have purchased one for $69.00. It may have only been available directly through Innovation. Support was poor, the following games are the only ones known to support it :
Airball - SFX only, slow PC req'd
Bad Blood - Music only
Battle Chess II - SFX only
BattleTech: The Crescent Hawks' Revenge
F-19 Stealth Fighter
Falcon A.T.
Harpoon
Joe Montana Football - SFX only
Lord of the Rings Volume 1 - Floppy version only, SFX only
Red Storm Rising
Super Jeopardy
Ultima VI - Music only
Windwalker
I have verified that all the above games work with the Innovation emulation in DOSBox except for Super Jeopardy. While Mobygames claims that Lexi-Cross and Castles and its expansion Castles : The Northern Campaign support Innovation, I have personally verified that these programs do not have install options for Innovation, despite what the system requirements labels on their boxes may indicate. Asterix: Operation Getafix is also listed by Mobygames as supporting the Innovation, but it actually supports the InterSound MDO, a Covox Speech Thing-like device released by the French gaming company Coktel Vision.
The best informational article describing the card can be found in Computer Gaming World, Number 63, September 1989. It mentions that only Origin and possibly Mindscape were planning to support the card. Interplay and Microprose eventually did with more than one title. Although Commodore made PC clones, it never put its sound chips in them. The SSI-2001 is the only non-Commodore product I am aware of that used the SID chip prior to Commodore's failure in 1994.
Technical details about the Innovation card and its SID implementation are as follows. The card puts the SID chip and its 29 registers directly on the ISA bus. It can use ports starting at 280, 2A0, 2C0 or 2E0. 280 is the default, and it takes up 32 consecutive I/O ports (3 presumably mirrored or unused). The SID is clocked using the 14.318180 MHz clock from the ISA bus. This clock signal is divided by the counter and flip flops on the board by 16 to produce a base frequency of .89488625 MHz. By contrast, the base frequency of the SID in an NTSC C64 is 1.02272714 MHz and a PAL C64 is .98524851 MHz.
The filter capacitors on the Innovation board use a value of 1nF. This was proper according to the datasheet of the 6581, but in practice in the Commodore 64, different values were used. By comparison, the 6581 SID in a C64 uses 470pF capacitors and an 8580 in a C64C or C128D use 22nF capacitors (1nF = 1000pF). If Innovation had used the later 8580 chips in their card, the datasheet for the 8580 specifies 6n8/6.8nF/6800pF capacitors. The POTX and POTY pins are not connected on the Innovation board, so the two registers which correspond to the functionality provided by these pins are useless.
The gameport is addressable at the standard port 201, supports 4 axes and 4 buttons, can be disabled with a jumper and supports three levels of "sensitivity", also selectable by jumper. The Innovation card is the first sound card known to incorporate a standard gameport, a feature popularized by the Sound Blaster in 1990 and then ubiquitous on sound cards for the rest of the decade.
By using the joystick enable jumper and the four port range selections, one can install four boards and hence four SID chips in one computer, if you had the slots to spare. Two boards would provide fully-independent stereo channels, but no software was known to take advantage of this capability.
By using the joystick enable jumper and the four port range selections, one can install four boards and hence four SID chips in one computer, if you had the slots to spare. Two boards would provide fully-independent stereo channels, but no software was known to take advantage of this capability.
I know of exactly two Innovation boards known to be in existence, and each has minor layout differences compared with the other. Both boards clearly show the use of 6581 chips, not the later 6582/8580 chips. Here are images of both boards known to be in existence :
Earlier Board :
Later Board :
Earlier Board :
Later Board :
No one I know has ever seen a "The Entertainer" branded card. It works just like the Innovation board hard coded at port 280, but it also has autodetection port functionality at port 200. Gunship and Pirates! will only produce SID sound if they read a value of 45h from this port. The code string BA 00 02 EC in Gunship 429.05 executables EGAME.EXE/CGAME.EXE/TGAME.EXE and TITLE.EXE (the only version that uses unencrypted executables) must be replaced with B8 A5 00 90. Ditto for the disk images of Pirates! Once the code is bypassed and an Innovation card is in the system or the emulation is enabled in DOSBox, you will hear SID sound. Unfortunately, DOSBox uses the 1.02MHz frequency and 470pF Caps for the SID, so it will not sound exactly like the real boards would.
III. The Modern Replica SSI-2001
In 2015, the prayers of many were answered when members of the VOGONS forum produced a replica of the SSI-2001 using a redesigned ISA board built mainly from photos of the card. This is what my card looks like :
Improvements over the old card include :
Can use 6581 or 8580 SIDs.
Jumper based speed setting for NTSC clock frequency or canonical SID clock frequency
Audio input jack
4-pin audio output header (dual mono on pins 1 & 4, pin 2 is ground)
Plated holes for POTX and POTY inputs
Fully labeled jumper settings
The cards have been made by a VOGONS user named Fagear, who is based in the Russian Federation. His work is top quality and he had the good sense to use mini-jacks instead of the RCA jack on the original. This meant that you can use any old junky ISA sound card for a bracket instead of having to make a custom bracket. My card works great, although I had to find my 6581 SID chip from another source.
You may notice, if you order from a card from Fagear, that the markings on the ICs, except for the SID, are not the familiar 74 series designations. The chips on this card are clones or work-alikes of the 74 series from several manufacturers exported by V/O Elektronorgtechnica. This company was a part of Elorg, the Soviet state owned entity that controlled imports and exports of computer hardware and where Tetris was born. However, with one apparent exception, the Russian parts appear to be functionally equivalent and use the same pinouts as the western chips. The identifying information for every part, western or Russian is silkscreened underneath the part, as can be seen here :
The sole exception is the chip just below the silkscreened SID CLOCK on the replica board, hence the jumper if the Russian or Western part is used.
You cannot just swap a 6581 for an 8580 in these boards whenever you like. The 6581 uses +12v, which the ISA bus provides. The 8580 uses +9v, which is not present on the ISA bus. In order to use an 8580, you must install a 78L09 voltage converter. The location is just to the left of the POT X holes. You would probably need to remove it if you revert to a 6581.
Thanks to VOGONS user form member bristlehog, you can play back .sid files through this card through a DOS utility called SIDPLAY. However, you will need a fast PC, around at least a 500MHz Pentium III, for proper speed. This is because .sid files use 6510 CPU instructions and rely on the 6526 CIA or the 6569 VIC-II for their timing, so a fair amount of these chips must be emulated. That means CPU power.
If you want to order a Replica SSI-2001 and download the player software, you can find it in this thread : http://www.vogons.org/viewtopic.php?f=46&t=27045
With this Board, I was able to verify that Super Jeopardy does output digitized sound. The output volume is so low as to make it almost useless. Activision got it right in the BattleTech: The Crescent Hawks' Revenge, the voice in the introduction is nice and clear with an Innovation.
III. The Modern Replica SSI-2001
In 2015, the prayers of many were answered when members of the VOGONS forum produced a replica of the SSI-2001 using a redesigned ISA board built mainly from photos of the card. This is what my card looks like :
Improvements over the old card include :
Can use 6581 or 8580 SIDs.
Jumper based speed setting for NTSC clock frequency or canonical SID clock frequency
Audio input jack
4-pin audio output header (dual mono on pins 1 & 4, pin 2 is ground)
Plated holes for POTX and POTY inputs
Fully labeled jumper settings
The cards have been made by a VOGONS user named Fagear, who is based in the Russian Federation. His work is top quality and he had the good sense to use mini-jacks instead of the RCA jack on the original. This meant that you can use any old junky ISA sound card for a bracket instead of having to make a custom bracket. My card works great, although I had to find my 6581 SID chip from another source.
You may notice, if you order from a card from Fagear, that the markings on the ICs, except for the SID, are not the familiar 74 series designations. The chips on this card are clones or work-alikes of the 74 series from several manufacturers exported by V/O Elektronorgtechnica. This company was a part of Elorg, the Soviet state owned entity that controlled imports and exports of computer hardware and where Tetris was born. However, with one apparent exception, the Russian parts appear to be functionally equivalent and use the same pinouts as the western chips. The identifying information for every part, western or Russian is silkscreened underneath the part, as can be seen here :
The sole exception is the chip just below the silkscreened SID CLOCK on the replica board, hence the jumper if the Russian or Western part is used.
You cannot just swap a 6581 for an 8580 in these boards whenever you like. The 6581 uses +12v, which the ISA bus provides. The 8580 uses +9v, which is not present on the ISA bus. In order to use an 8580, you must install a 78L09 voltage converter. The location is just to the left of the POT X holes. You would probably need to remove it if you revert to a 6581.
Thanks to VOGONS user form member bristlehog, you can play back .sid files through this card through a DOS utility called SIDPLAY. However, you will need a fast PC, around at least a 500MHz Pentium III, for proper speed. This is because .sid files use 6510 CPU instructions and rely on the 6526 CIA or the 6569 VIC-II for their timing, so a fair amount of these chips must be emulated. That means CPU power.
If you want to order a Replica SSI-2001 and download the player software, you can find it in this thread : http://www.vogons.org/viewtopic.php?f=46&t=27045
With this Board, I was able to verify that Super Jeopardy does output digitized sound. The output volume is so low as to make it almost useless. Activision got it right in the BattleTech: The Crescent Hawks' Revenge, the voice in the introduction is nice and clear with an Innovation.
Tuesday, December 31, 2013
Versions and Sources of the Phantom of the Opera
I. What was Actually Released? - Theatrical Release Versions
A. The Los Angeles Preview Version
As its name implies, this was a rough cut of Phantom of the Opera shown to test audiences in Los Angeles on January 7 and 26, 1925. At this point, the direction was totally credited to Rupert Julian. The critical and audience response was largely negative, and led to substantial changes. The ending is the closest to the book, as the Phantom is shown to die of a broken heart after he learns that Christine can never truly love him. An alternate ending where the Persian shoots the Phantom in his lair was shot and may also have been shown in some screenings of this version. Only this version also includes a scene at the Graveyard and Madame Valerius, both of which were in the novel by Gaston Leroux. Also like in the novel, Christine tells Raoul of her unmasking of the Phantom in flashback as they talk on the Opera Roof. At this point, Arthur Edmund Carewe's character is The Persian, just like in Leroux' book, and the Phantom's backstory is consistent with the one given in the book. This version was long enough at 22 reels to have an Intermission.
B. The San Francisco World Premiere Version
This was the second version of Phantom of the Opera shown to the public, in this case at the Curran Theater in San Francisco for a four week engagement from April 26 into May 23, 1925. This version includes newly-shot scenes directed by Edward Sedgewick. It too suffered a critical drubbing and audience apathy, and again the film was sent back for further editing. This version includes a newly-shot substantial subplot with Raoul and a Count Ruboff, played by Ward Crane, competing for Christine's affections, leading up to a duel. Added for mostly comic relief were Chester Conklin as Raoul's valet and Viola Vale as Christine's maid. The subplot occurs outside and separate from the happenings at the Opera and has been criticized as being cliched. The film's ending where the Phantom tries to escape the mob by carriage and is beaten and thrown into the Seine was shot by Sedgewick and first appears in this version. This and all subsequent versions could come in at 10 reels.
C. The New York General Release Version (General Release)
The third time was the charm for the Phantom, this version premiered at the Astor Theater in New York City on September 6, 1925 and shown well into 1926 across the United States. All of the material directed by Sedgewick for the San Francisco version was eliminated except for the new ending. At this stage, The Persian has become Inspector Ledoux of the Secret Police and the Phantom is a madman escaped from Devil's Island. Also, this version has the famously inept shot of Norman Kerry playing with his gloves as the he is eavesdropping on Phantom talking to Christine behind the wall. To the extent that any of the versions from 1925 have survived, this is the version that still exists in some substantial form (see part II-B, below). This version was reported to contain sequences of the Opera stage and stairwell, including the Bal Masque, in Prizma color and Process 2 Technicolor. It also uses the Handschliegl process for at least the Phantom's cape when he spies on Raoul and Christine on the roof of the opera.
D. The Talkie Re-release Version (Talkie Re-release)
Four years after the last footage had been shot for the silent versions of the Phantom of the Opera, Universal decided to make reissue it as a partial talkie. To do this, they reengaged Mary Philbin and Norman Kerry and some of the other original actors alongside some new actors to shoot new sound sequences that would be edited with the silent footage. Carlotta was now played by Mary Fabian, who could sing, and her mother was played by Fay Holderness for the talking scenes with the managers. Edward Martindel replaced John Sainpolis as Phillipe de Chagny except in a few long shots, and These sequences were directed in 1929 by Ernst Laemmle. Lon Chaney was under contract to MGM and Universal could not dub his voice. To get around this problem, they had another actor act as the Phantom's servant for those scenes where the Phantom talks to other people behind a wall or mirror. Once the Phantom as portrayed by Lon Chaney appears in person to Christine, intertitles resume and his servant disappears. The servant was played by Phillips Smalley. This was first released in Dallas and Detroit on January 3, 1930, but is mostly lost (see II-E, below, regarding its survival). It came in 10 reels.
II. What Survives? - Film and Disk
A. The George Eastman House Print (Eastman Print)
The best visual representation of the film in terms of quality survives in a 35mm print struck by Universal at the request of James Card, first curator of the George Eastman House in 1950. This print runs to ten reels and is about 90 minutes at 20fps and 75 minutes when run at 24fps (sound speed). It is not tinted and has no technicolor sequences. It also is apparently cropped on the left side, probably because it was duplicated on machinery intended for sound film. A sound negative in the Academy Ratio has the optical soundtrack running down the left side of the film. Although the original aspect ratio of the film is 1.33:1, due to the blocking on the left side of the image, the resulting Eastman Print has more of a Movietone aspect ratio close to 1.2:1. With sound film, the width of the film stip is the same but the left side of the film strip is taken up by the optical soundtrack. It has some moderate to severe mottling damage during the scene where Christine wakes up in her "bedroom" in the Phantom's lair.
This Print has been erroneously believed to be the Talkie Re-release, but it is not the case. Virtually all the footage in this print was shot in 1924-1925. The opening titles for the Eastman Print are from the Talkie Re-release in 1929. The "Phantom" skulking through the set during the titles is not Lon Chaney. The surviving sound discs for the Talkie Re-release do not synchronize with this footage except during Carlotta's signing from Faust.
The Eastman Print contains a puzzling sequence at the beginning of the film where a man with a lantern is walking in the cellars of the opera and "talks" to the audience. The Phantom's shadow in cape and hat also appears in this scene. The scene lasts for approximately one minute at sound speed are there are no intertitles. This caused film historian Scott MacQueen to believe that this sequence was shot so Universal could release the film internationally as a sound film. The Man with a Lantern could be attached to any foreign print speaking in the local language, and combined with the singing sequences, a foreign distributor could technically release it as a talking film. What he is saying has never been definitively established, but Phillip Riley in the Phantom of the Opera, Magic Image Filmbooks Series, Classic Silents - Volume 1, showed the sequence to a deaf person and a lip reader and one of them opined that the Man with a Lantern was speaking in German. The other person thought he could make out "Gaston Leroux". The other alternatives, albeit without any documentary support, is that this was a U.S. silent version of the Talkie Re-release for theaters that had not converted to sound or it was an internal reference print for Universal itself. No advertisements indicate that the Phantom was being offered for rental as a silent film in 1929-1930.
The actress who played Carlotta in 1925, Virginia Pearson, was redesignated as Carlotta's Mother in the Eastman Print. Carlotta was played by Mary Fabian, using footage shot for the Talkie Re-release. The opening credits give listings for both actresses, but the closing credits are from the General Release of 1925 and only list Virginia Pearson as Carlotta. The shots where Christine sings on the Opera Stage just before she is abducted by the Phantom is also from the Talkie Re-release. The notes and intertitles were also from the Talkie Re-release, although some new ones were necessary for the scenes which had talk but were silent in this print.
B. The John Hampton Show-at-Home Duplications Compilation Print (Hampton Print)
The Eastman Print is a mix of mostly outtakes and second camera takes shot in 1924-1925, B&W alternate takes of color footage and some Talkie Re-release footage. The only other near complete source for the Phantom of the Opera came from the collection of John Hampton, an early silent film preservationist who ran the Old Time Movie theater from 1940 to 1979. He showed only silent films at this theater which is still in operation and now known under new management as The Silent Theatre (and plays sound films as well, but all its silent films are played at sound speed). Back in the 1930s, Universal marketed 16mm reduction prints of some of their silent movies to the mass market. This is how 1923 film of The Hunchback of Notre Dame starring Lon Chaney survived. Hampton was able to collect some of these Show-at-Home prints to compile a mostly complete representation of the 1925 General Release of the film. According to one source, Hampton took the best elements of eight separate prints and spliced the best parts of each to make the print that is available today. Hampton's materials found their way to the Stanton Theater Foundation and are now stored at the UCLA Film and Television Archive.
This version is not at the same level of quality as the Eastman print. Detail is softer, contrast is poorer and scratches abound. It is much longer at about 115 minutes when run at 20 frames per second. No tinting or color sequences are present on the film elements. When compared, The Hunchback of Notre Dame has a better quality picture than Phantom as represented in the Hampton Print because it survives as a true Show-at-Home print and is fewer generations away from the negative. The releases of the Hampton Print currently available use scans of dupes from Hampton's prints. Hampton's original materials, which are stored at the UCLA Film and Television Archive, are superior to what is available but they have not been made generally available. UCLA announced a restoration of Hampton's original materials about fifteen years ago, but no progress has been made.
While the Hampton Print is almost totally comprised of footage from the 1925 General Release, the title sequence is borrowed from the Talkie Re-release until the card "Produced in its entirety at Universal City, California." The Bal Masque footage seen in the print is from the B&W camera, not the Technicolor camera which the 1925 General Release Version used (see below).
There is short sequence with the man with the lantern at the beginning of the Hampton Print, but it only lasts for about 13 seconds in this Print and included a medium shot of him not in the Eastman Print.
C. Trailers for the 1925 and 1929 Releases
The trailers for both the 1925 General Release and 1929 Talkie Re-release versions survive. The trailer for the Talkie Re-release is believed to contain much of the original title sequence for the 1925 film. If there was a sound disk (see section I-E., below) for this trailer then the disk has not survived. Ironically, the trailer for the 1925 General Release on most discs uses the title sequence from the 1929 Talkie Re-release. The surviving trailer film for the 1925 General Release should probably have begun with the Phantom shown over the Opera house with the names of the principal stars. Also, all the footage shown in the Talkie Re-release trailer was shot in 1924-25, no actual footage shot in 1929 was used.
The trailers contain some footage not seen in the Eastman or Hampton prints. For example, the trailers show the Phantom riding up the counterweight of the pulley mechanism that controls the mirror, which gives a more convincing explanation why Christine does not see him at first. It is easy to see why this was cut, however. With that shot, the impact of seeing the Phantom as masked is diminished.
D. Technicolor Bal Masque Sequence
In the 1970s, Film Archivist David Shepard, working for Blackhawk Films, discovered a sequence in two-strip Technicolor of the opening scenes at the Bal Masque. This film runs from the intertitle introducing the Bal Masque to the point when Christine and Raoul go onto the roof. The sequence was shot with Technicolor equipment and is different visually from the same sequence in the Eastman Print, shot using the B&W cameras. Two subsequent sequences, the first being Christine and Raoul descending from the roof and Florine encountering the Red Death have not been located but do exist in the Eastman Print in B&W.
The Eastman Print uses 1929 Talkie Re-release style intertitles and B&W footage. The Hampton Print also uses the B&W footage, but the intertitles have a curtain-style background. The Technicolor footage uses a third set of intertitles, in color with a different typeface from the first two and a different curtain image. Essentially there are three surviving versions of this sequence : Eastman B&W, Hampton Technicolor-to-B&W & Shepard Technicolor.
E. Talkie Re-Release
The 1929 Talkie Re-Release does not survive in its entirety. It had a Vitaphone soundtrack where the sound was stored on vinyl records or discs synchronized to the film projector. The 1929 Talkie Re-Release came on 10 reels and the ten accompanying sound disks survive. Of the film itself, only Reel 5 has been found and is currently stored in the Library of Congress, as are the sound disks. The reel is missing some footage, but Christine and Raoul have aged, especially the former. The Phantom's servant is visible in the newly-discovered Reel 5. Overall, there is about 1 minute and 15 seconds of new footage showing Christine and the Phantom's servant shot in 1929 and about 12 seconds of Raoul in footage shot in 1929 from this reel.
Here are the scenes with no corresponding video on the discs :
Scene 1 - Manager's Office: the Managers and the House Manager discuss why Box 12 has not been rented recently
Scene 2 - Outside and Inside Box 12: the Managers interrogate Madame Giry and investigate the occupant of Box 12
Scene 3 - Manager's Office: the Managers discuss the Phantom's demand to have Christine sign in Faust
Scene 4 - Christine's Dressing Room: Raoul asks Christine about her new-found fortune and her new admirer
Scene 5 - Opera Stage & Box: Christine sings in Faust and Phillipe warns Raoul about the Phantom
Scene 6 - Christine's Dressing Room: Raoul asks Christine again about her admirer and her Master's servant tells Christine that she should be prepared to go soon
Scene 7 - Manager's Office: Carlotta's Mother tells the Managers that her daughter will sign in Faust.
Scene 8 - Christine's Dressing Room: The Master's servant tells Christine to come through the mirror. The film for this scene has been mostly recovered.
Scene 9 - Roof of the Opera: Christine tells Raoul that she has seen the Phantom and they make plans
F. "Scream Scenes"
This is a compilation of clips from various films which was released in 1959. It shows a cut-down version of Christine's encounter with the Phantom in his lair and the unmasking scene. It's importance lies in that it contains three title cards, none of which exist in the Eastman or Hampton prints using the same text, suggesting that these scenes were taken from the Los Angeles or San Francisco versions of the film. It also contains a shot, not found on any other version, of the Phantom warming up on the organ before Christine exits the boudoir. The shots in the boudouir do not suffer from any nitrate decomposition as they do on the Eastman print. The Phantom sequence of "Scream Scenes lasts approximately 4:42.
III. What's Available - Home Video Releases
As the Phantom of the Opera is in the public domain in its raw format, anyone can scan the film and release it on some kind of home video format. There have been cheap VHS and DVD copies found in bargain bins for years. The cheapest releases are usually from an ancient transfer of the Eastman Print, sometimes a 16mm reduction print or derived from the Hampton Print.
All quality releases (with one exception) since the first Laserdisc in 1990 have two pedigrees. The first is from David Shepard, Blackhawk Films and Film Associates. The second is from Kevin Brownlow and Photoplay Productions. Shepard took the Eastman Print and inserted the recovered Technicolor Bal Masque sequence for the B&W footage in the Eastman Print. The work dates from 1990 but has been tinkered with since then.
Brownlow's version dates from 1996 and in addition to substituting the Technicolor footage, makes other changes. First, the Man with a Lantern and the closing cast credits are excised completely from this version. The justification for the first edit was likely that the sequence was superfluous to a silent film and would be puzzling for a modern audience. The second edit was done because the credits mistakenly credit Virginia Pearson as Carlotta when she was re-cast as Carlotta's mother. This version was broadcast to great fanfare on Channel 4 in the U.K. that same year and the general public would not be familiar with the issues surrounding this footage. Second, the Handschiegl effect on the Phantom's cape has been simulated using computer colorization. Third, the Bal Masque sequences after Christine and Raoul have descended from the roof have been computer colorized to approximate the two-color Technicolor look, even though the Technicolor film would have used different shots. Fourth, for the nitrate-decomposition affected scenes set in Christine's intended bedroom in the Phantom's lair, footage from the 16mm Hampton Print was substituted as it does not demonstrate the heavy mottling captured in the Eastman Print.
Shepard's version would eventually follow Brownlow's in part on the Blu-ray by recreating the Handschiegl effect on the Phantom's cape and in the Torture Chamber. The tinting schemes used by both versions have substantial differences, as described below. In their presentations of the Eastman Print, both versions leave out the B&W variation of the Technicolor Bal Masque Sequence.
The main score for the Shepard editions is from French-Canadian composer Gabriel Thibaudeau and composed in 1992, while British composer Carl Davis performed the honors for the Brownlow version in 1996. If one of these composers is identified in the version you are watching, you know from which pedigree your version comes from.
Shepard's version has been officially released on five occasions on an optical disc format, on Laserdisc in 1990, on Laserdisc and DVD in 1997 and on Blu-ray in 2011, all by Image Entertainment, and in 2015 by Kino International. Brownlow's version has been released twice, first by Milestone Film (distributed by Image) in 2003 and by the BFI in 2013. The 1997 DVD and BFI discs are region free, the 2003 DVD is Region 1 and the 2011 and 2015 Blu-rays are Region A. The latest Blu-rays are all still in print or easily available. All releases are single disc unless otherwise stated.
A. Discs Released
1. 1990 Image Laserdisc
This version runs the Eastman Print at 24fps with the Gaylord Carter Organ Score (probably the same recording as on the Image Blu-ray), but also includes the Hampton Print as an extra without a score. It contains the 1925 Trailer. This disc has long since been superseded.
2. 1997 Image Laserdisc and DVD
This version runs the Eastman Print at 20fps with the Thibeadeau score but only includes some scenes from the Hampton Print not included in the Eastman Print. It also includes the 1925 Trailer and an essay. This may have been the first silent film to be released on DVD.
3. 2003 Milestone DVD
This two-disc version runs the Eastman and Hampton Prints at 20fps. The Hampton Print is accompanied by a score from Jon C. Mirsalis. It contains the 1925 and 1929 Trailers, the latter with sound from the Talkie re-release discs, a video interview with Carla Laemmle and an audio archival interview with cameraman Charles Van Enger, still reconstructions of the Los Angeles Preview and San Francisco World Premiere versions, an extract of Faust from the 1929 feature Tiffany, nine dialogue selections from the Talkie re-release discs, the soundtrack from the talkie re-release edited to the Eastman Print, audio commentary by Scott MacQueen. The only major film sequence missing and available at the time of this release was the Man with a Lantern from the Eastman Print. The Eastman Print exhibited here has a problem with ghosting and motion artifacts believed to be the result of an improper PAL to NTSC transfer. While this DVD is long out of print, it contains many extras not found elsewhere.
4. 2009 Reelclassics DVD
This four disc set offers the Eastman Print with the Technicolor Bal Masque sequence and the Eastman Print with the B&W Bal Masque sequence. The score included for the Eastman Print is an organ score from Lee Erwin and a voice for the Man with a Lantern by John Griggs. It also offers the Hampton Print with the Technicolor Bal Masque sequence and a score by Ben Model. The print scanned for the Eastman Print is 16mm reduction from Griggs Moviedrome. Its features include a "widescreen" side by side comparison of the Eastman and Hampton Prints and the Technicolor and B&W Bal Masque sequences. Finally, it is the only release that features the "Scream Scenes" footage. These discs are printed on burned DVDs.
The contents of the DVDs are as follows :
Disc 1 - Eastman Print (from Griggs Moviedrome) with Technicolor Bal Masque sequence, The Light of Faith (short film starring Lon Chaney from 1922), Scream Scenes
Disc 2 - Eastman Print (from Griggs Moviedrome) with B&W Bal Masque sequence, 1925 Trailer
Disc 3 - Hampton Print
Disc 4 - Eastman vs. Hampton Print side-by-side Comparison, Bal Masque Color vs. B&W Comparison with commentary from, Keith Paynter
5. 2011 Image Blu-ray
This Blu-ray disc contains two HD versions of the Eastman Print. One version runs the film at a progressive 24fps at 1080p and shows noticeably less print damage. The other version runs the film at 20fps. Since Blu-ray spec only officially supports progressive frame rates at 24p or 23.976p, the 20fps version uses a 1080/29.97i resolution and telecine pulldown to support the unusual frame rate. A score from the Alloy Orchestra or the Gaylord Carter Organ score can be selected for the 24p version. The Carter score was composed in the 1970s and is synchronized to the film running at 24fps.
The 20fps version on the 2011 Image Blu-ray was interlaced using a pulldown process. The actual frame rate is 30fps, and this is achieved through interlacing. The field rate is here becomes 60fps. So, in between each two pure frames of the video, there is a "dirty frame" consisting of the odd lines of one frame of the film and the even lines of the next frame of the film. You can see this illustrated here :
o stands for odd, e stands for even and the highlighted pair are the dirty frames.
The Thibaudeau score or an audio commentary from Jon C. Mirsalis can be selected for the 20fps version. This Blu-ray contains a tinted version of the Hampton Print, but it is presented in Standard Definition NTSC resolution (720x480) only. It is accompanied by a piano score from Frederick Hodges. The 1925 Trailer is included and an interview with composer Thibeadeau, both also in SD. The Fifth Revised Shooting Script and Souvenir Program are also included as a succession of still images. The contents of this disc were released by Park Circus in the U.K. on a Region B disc.
The first pressing of this disc had a non-intuitive menu where you could only select the versions of the prints you wanted to watch by selecting the score attached to that version. To change to another version you had to eject and reinsert the disc or jump ahead to the end. Also, all stereo scores were output in mono and the Carter score was out of sync for Reel 1. The ballet dancers in the first ballet sequence also appear to be moving in slow motion in both the 24 and 20fps versions. A replacement disc fixes the stereo/mono issue and uses a new, more intuitive and technically impressive menu.
6. 2013 BFI Blu-ray Dual-Format
This package contains one Blu-ray disc and two DVDs. Although this post and blog is written from the perspective of someone in the United States, I am not adverse to including information about a new and substantial release elsewhere in the world. The Blu-ray disc and the two DVDs included are region free, but the DVDs are PAL and may not play in a U.S. player. The Blu-ray and the first DVD have identical content, with the Eastman and Hampton Prints and the extras, the second DVD just has the documentary Lon Chaney : A Thousand Faces.
Both the Eastman and Hampton Prints are run at 20fps, but in both cases the 1080p/24fps mode is used on Blu-ray. This is a dual-format release, so the prints are presented on the Blu-ray and one of the DVDs. On the DVDs, all video is shown at the 50i PAL standard. The film is sped up 4% as a result, but the audio is pitch corrected. This means that one out of every five frames is repeated to make up the missing 4fps. The Hampton Print is in HD, unlike the Image Blu-ray, but is not tinted. The Technicolor Bal Masque sequence in the Eastman Print is windowboxed, as is the sequence where they used a 16mm print for the scene with Christine in the boudoir, presumably to match the Movietone-like aspect ratio of the rest of the Eastman Print. The shots of the Phantom on the roof with the Handscliegl effect applied are cropped and somewhat grainier than the rest of the film.
The BFI Eastman Print is based on the 1996 Brownlow restoration. This restoration was done at standard definition at the PAL format. For those sequences where the 16mm Hampton Print is used to replace the damaged scene with Christine in her bedroom in the Phantom's lair, the three color titles at the Bal Masque, the Handscliegl process on the Phantom's cape and the colorized portion of the Bal Masque, all these sequences have been upconverted from SD materials.
The two trailers are included, and the recently discovered material from Reel 5 of the 1929 Talkie Re-release is included synchronized with its sound disc. The Man with a Lantern from the Eastman Print is included as an extra, the sequence remains excised from the main presentation. All video footage on the Blu-ray is presented in HD, scanned at 2K. The DVD uses the same scans, downscaled from HD into SD PAL. The short sequence of the Man with a Lantern from the Hampton print is deleted in this version. Also, the closing credits card in the Eastman Print for the cast is deleted.
The second DVD contains the documentary Lon Chaney: A Thousand Faces, first broadcast for Turner Classic Movies in 2000. It is directed by Kevin Brownlow and narrated by Kenneth Branagh. A booklet accompanies the discs describing the production and restoration of the film.
7. 2015 Kino-Lorber Blu-ray or DVD
This Blu-ray is something like a re-issue of the now out of print Image Blu-ray, but comes on two Blu-rays. A DVD version is also available. The first Blu-ray has 24fps and 20fps versions of the Eastman Print like the Image Blu-ray, but nothing else. The 20fps version is now using the 24p progressive frame rate supported by the Blu-ray format. The 20fps version also appears to have been noticeably cleaned up compared to the Image Blu-ray and even the 24fps version has minor improvements and corrects tinting errors. Because these versions share a dual-layered disc, their bitrates are higher than the older Image versions.
The 2015 20fps version is displayed progressively at 24p (or 23.976p). In order to do this, every fifth frame is repeated once to make up the extra 4 frames needed by the format. The reason this is done for this and many other silent films is because the Blu-ray format does not officially recognize any other progressive frame rate except for 59.94p at 1280x720.
The 24fps version on the Image and Kino Blu-rays came from Paul Killiam's film collection. Killiam was a comedian in vogue in the 1950s who did much to help preserve silent film. His print was duplicated directly from the Eastman Print and I believe eventually found its way into David Sherpard's Blackhawk Film archive.
The 20fps version on the Kino Blu-ray is a slowed down version of Killiam's print. The old 20fps version on the Image Blu-ray was duplicated from Killiam's copy, making it as a source a generation further away from the Eastman Print. This is one reason why it looks rather rough. Neither version on the Kino disc shows the tint colors in the pillarboxes on the sides of the video, which was a flaw on the Image Blu-ray.
There was a sychronization issue with the Thibedeau score on the Image Blu-ray which was mostly fixed on the Kino Blu-ray. The only issue that was not fixed was the slow motion ballerinas. The ballerina sequence at the beginning of the film was stretch printed on the print Image and Kino used. Stretch printing slow downs the action shown on film by duplicating frames when prints are made. Fortunately this issue lasts only for a minute or two (3:04 to 4:42 on the Kino 20fps version for example) and does not affect any other portion of the film. I think the ballet music used in this scene goes on for too long and spills into the next scene. The Alloy Orchestra's score on the 24fps version is much better timed to the action.
Unfortunately, the Carter score on the 24fps version is totally out of sync for the first 11 minutes of the film. The correct music plays about a minute and a forty-five seconds later than it should. The music for the ballerinas start to play at the end of the next scene with the managers. This sync issue may have occurred on the first version of the Image Blu-ray but did not on the second.
All the extras and the Hampton print are on the second Blu-ray disc, which is also dual layered. The extras are mostly the same as on the Image Blu-ray with three extras.
The Hampton Print is now in 1080i, but it is an upscaled version of the 480i version found on the Image Blu-ray. However, it is better to have the upscaling done by professional software instead of the Blu-ray player or the TV, and the film uses much more of the disc on the Kino than the Image.
Essentially, if the old Image disc is version 1.0, the newer image disc is version 2.0, then the Kino disc is essentially version 3.0.
All discs except for the 1990 Laserdisc have a speed-corrected transfer of the film to 20 frames per second. except where specified as running at 24 frames per second. Sound films use 24 frames per seconds, but silent films generally were projected at 16-22 frames per second. Milestone's Eastman Print has a serious issue with ghosting and jittery movement, probably due to frame rate differences between the PAL video master and its conversion to NTSC video. Image's Blu-ray versions suffer from some unnatural slowdown during the opening ballet sequences and speedup in the first scene in the manager's office, but are otherwise without motion issues (unless they are in the original print).
As befitting Blu-ray, the Eastman Print on either disc looks much sharper than any DVD or Laserdisc due to the high definition transfer. One example of this is that it is easy to identify in which shots (except for long shots) Chaney is wearing his mask with painted on eyes and which shots he is using a different mask with cutouts for his eyes. However, sometimes the focus is soft, so you should not expect razor sharp clarity in every frame.
As has been noted by others, the Image 24fps version had some amount of post-processing work done to reduce the scratches and wear on the print. However, no attempt was made on either Image version to remove large hairs that pop in the edges of the frame, large scratches or a broken film perforation fragment which appears as Christine meets the Phantom for the first time. People have stated that the Image 20fps version, while looking more worn, also looks more natural and less processed. Milestone and BFI seem to have made attempts to clean up their video presentation of the major blemishes and scratches. There is no broken film perforations in these versions.
Milestone and BFI substituted 16mm footage from the Hampton print for most of the scene where Christine is in her room in the Phantom's lair. This was due to the nitrate decomposition damage already present on the Eastman Print. Image DVD and Blu-ray used the 35mm footage instead. The 35mm footage here is not as sharp or clear due to the mottling, but still sharper and has better continuity with the 35mm footage preceding and following it.
Image Laserdisc, DVD and Blu-ray have the Man with a Lantern sequence, Milestone does not and BFI has it as an untinted extra. When Photoplay Productions created the video broadcast master used by Channel Four, Photoplay or Channel Four must have deleted it as it would be baffling to the uninformed viewer.
Image also has the cast credits at the end of its Eastman Print, Milestone and BFI do not.
Image Laserdisc, DVD and Blu-ray have no Technicolor footage after the scene on the roof of the opera. The footage is tinted. Milestone and BFI has colorized all the footage of the Bal Masque occuring after the scene. Image Blu-ray recreated the Handschiegl process for the face of the Rat catcher and to visualize the heat in the torture chamber. While I was previously aware of the possibility of some Handschiegl work for the rat catcher's scarf, I was not aware that there was the possibility of the Handschiegl process being used for his face or the torture chamber. Both recreate the process for the Phantom's costume when he is on the roof, but the result is noticeably different.
Milestone's left border is fuzzy, but Image's is sharper. Image's may have a sliver less of information in the video frame as a result. The BFI has a sharp border and a sliver more information than Image. The Technicolor footage of the Bal Masque came from a different source and will be in a true 1.33:1 ratio.
While the tinting is the same color on the Image 24fps and 20fps there are fluctions in the saturation levels between the two prints in some scenes. The scene where we first encounter Florine Papillon, Snitz Edwards' character, is very orange in the 24fps and much more sepia in the 20fps. The BFI Blu-ray tints are often extremely saturated, especially the blue for night scenes and the purple for opera cellar scenes.
There is a frame rate issue inherent in the Eastman Print. The vast majority of the footage was shot in 1925 at silent speed. This should be projected correctly at 20fps. The footage shot in 1929 would have been shot in sound speed. This means that the Opening Titles (where the phantom is skulking in the frame), The Man with a Lantern scene and the shots of Mary Fabian as Carlotta singing should be ideally projected at 24fps (especially for the latter if it is to be synchronized to the existing audio. The shots where Christine is singing just before she is abducted by the Phantom may also be from 1929. The BFI handles this by running these sequences at a true 24 frames per second while running the rest of the film at 20 frames per second.
2. Hampton Prints
Milestone is a straight black and white presentation of a show-at-home print. Previously this version existed on the 1990 Laserdisc and really poor public domain releases. I would advise lowering the contrast a bit when watching it. There are combing artifacts due to the interlacing.
Image apparently added tints to their presentation. Universal would not have gone to the extra expense of tinting 16mm inexpensive show at home prints. Both Milestone and Image are in standard definition.
Despite what the back of the Image Blu-ray case says, they used a 16mm source for the 1925 version. (I am not even sure if motion picture film exists in a "6mm" gauge.) In my opinion, Milestone boasts a slightly sharper image due to the lack of tinting (see below for more detail).
BFI's Hampton Print is in high definition, and the image quality is far superior to either Milestone or Image. It also shows more of the film frame than Milestone. Even the BFI's DVD has superior image quality to anything released before this. Unfortunately, the Man with a Lantern is excised from this version and two of the title credits "freeze" the image weirdly for several seconds, probably to keep the score in sync (the Man with a Lantern sequence takes a few seconds). This also happens on two intertitles during the scenes where Christine is meeting the Phantom for the first time. This was probably done to ease the transition between print changes. (Several prints were used to make up the Hampton version).
D. Audio Options
a. Carl Davis Score
This is the same individual who did the commentary on the Image. His website is http://www.lonchaney.org/. My favorite score for the 1925 version of the film and is tied with the Carter/Killiam score for my all-time favorite score for Phantom.
Unfortunately, Image's scores can be heard only with their respective versions. The scores offered for the 24fps version cannot be heard with the 20fps version and vice versa. This is due to the fact that these scores and commentary were intended for a particular running time. The 24fps version runs 77 minutes long, the 20fps version 91 minutes. The 1925 version of the film runs for 124 minutes.
This score was commissioned specifically for Image. Score is in stereo.
6. BFI Hampton Print
a. Ed Bussey Piano Score
This score was commissioned specifically for BFI. Score is in stereo. Neither the Image or BFI piano scores for the Hampton Print come close to Mirsalis' score, in my opinion, on the Milestone disc.
a. 1925 Theatrical Trailer
Untinted, silent, has few extra seconds compared with Image.
b. 1929 Theatrical Trailer
Has synchronized soundtrack, tinted in sepia.
c. Still Gallery
Los Angeles Premier Version photo Reconstruction, San Francisco Premier Version photo Reconstruction, Posters & Lobby Cards, Promotion, Press Kit, The Phantom Stage, Publicity Portraits, Backstage Stills, Concept Art, Publishing
d. Talking Extracts from 1929 Sound Version
Consists of nine sequences, most of which were not included in the synchronized sountrack option. Sequences are Managers' Office I, Box 12, Managers' Office II, Garden Love Scene, Faust - Trio and Finale, Christine's Dressing Room, Managers' Office III, Through the Mirror, Apollo's Lyre. The length of many of these scenes is longer than their equivalent silent counterparts and are noticeably different. Here is an example :
In the Box 12 excerpt, the managers speak to the maid, enter the box to find the Phantom, exit and talk to the maid again, enter to put the Phantom out, but when they find the box empty, they exit and tell the maid to tell no one. This takes approximately 2 minutes to accomplish at sound speed. In the silent version, the managers only speak to the maid once and are able to finish the scene in 1 minute, 40 seconds at silent speed.
e. Scene Selections - 24
a. Video interview of Carla Laemmle by David J. Skal
Carla Laemmle was the niece of Carl Laemmle Sr., the President of Universal Studios. Laemmle put his relatives on the payroll, and Carla played the part of head ballerina in Phantom. She also plays a passenger in the coach in the opening scene of Universal's 1931 production of Dracula. Her interviewer is David J. Skal, a noted researcher into Universal's horror films of the 30s and 40s. He has written several books on the subject of horror films.
b. Faust opera extract from film Midstream
Contemporary footage of a the 1929 film Midstream which features Charles Gounod's opera Faust, which is so prominently featured in Phantom. It gives you an idea what 1929's sound technology could do when am opera scene needed to be filmed.
c. Audio interview with Charles Van Enger, ASC
This interview was recorded on July 29, 1973. Enger was the principal cameraman on Phantom and the interviewer was Richard Koszarski. He had a low opinion of Director Rupert Julian and respected Lon Chaney.
d. Scene Selections - 29
a. Interview with Composer Gabriel Thibeadeau
4. BFI Blu-ray/DVD Combo
a. 1925 General Release Trailer
No tint, uses 1929 Title Card (with the title merging to the center). Uses 20 frames per second within 24p on Blu-ray. Interestingly, you can see in this extra what the title card looked like without the cropping of the Eastman Print.
b. 1929 Talkie Re-release Trailer
Sepia tinted, uses 24 true frames per second speed. The music comes from the Talkie Re-release discs, but is different than the same trailer on the Milestone.
c. Reel 5 of the 1929 Talkie Re-release
No tint for the footage that comes from the true Reel 5, Sepia tinted for scenes taken from the Eastman Print to complete the reel. There are two places where the frame goes black for a second or two, one of these is where Christine speaks the line "Oh my, I'm afraid." The Phantom's servant can be seen clearly speaking the line "Come through the mirror, have no fear." 24 true frames per second on Blu-ray.
d. Man with a Lantern Sequence
No tint, 24 true frames per second on Blu-ray.
e. Channel 4 Silents restoration programme
This is included as a PDF file on the first DVD. This programme was issued to theatre patrons when the 1996 Brownlow restoration premiered theatrically in London in 1926. It gives the history of the production, the restoration, biographical summaries of Chaney, Julian, Philbin, Kerry and discusses the Davis score.
f. Lon Chaney - A Thousand Faces Documentary
On the second DVD, first broadcast by Turner Classic Movies in 2000, an 86-minute documentary on the life and films of the great actor. PAL Format. As this documentary was produced by Kevin Brownlow in the U.K., the 25fps PAL format is not an inappropriate video standard.
g. Booklet
When the Image Blu-ray disc was first released, it was discovered that the disc was defective. First, all stereo scores ended up in monaural. There was no recall of the discs, so owners of the first release had to have contacted Image to secure a replacement disc. Image has since merged with Acorn Media and replacement discs are no longer available. The Gaylord Carter score was out-of-sync for the first couple of minutes. Finally, the old menu required you to select the version of the film by selecting the score. So to watch the 20fps version, you had to select the Thibeadeau score. You would have to know which score went to each version. Moreover, to change to a different version you had to either eject and reinsert the Blu-ray or chapter select and fast forward to the end of the version. The new menu lets you choose the version first, then the score. If you return to the main menu, you can choose another version. Here are a screenshot of the old and new menus :
The Shepard and Brownlow restorations and the Hampton Print on the Image Blu-ray have different tinting in several sequences. Brownlow barely uses tint in the Phantom's lair on the Milestone and these scenes are completely black and white on the BFI. Shepard uses sepia tinting for these scenes, and I could find no purple tinting at all. Brownlow uses purple for the opera cellars. Image has no tinting for the first intertitle "Sanctuary of the song lovers..." or for the intertitle and the short scene where Raoul reads Christine's note just prior to the Bal Masque. This confirms that the original 35mm Eastman Print is in black and white. Additionally, when Shepard uses sepia tinting, the whole screen, including the borders (since the resolution is 1920x1080 regardless of aspect ratio), is sepia toned on the Image Blu-ray. This was fixed for the Kino Blu-ray. Note that below that the Kino 20fps uses the same tinting choices as the 24fps Kino and Image sources, they all derive from the same source. David Shepard has confirmed that the telecine they worked from was in black and white except for the Technicolor footage. Here are the differences :
Intertitle - "Sanctuary of Song Lovers..." - White (untinted)
Opening - Blue
The Image Blu-ray Hampton Print seems like someone started to tint the film but did not finish. Virtually all the film is in sepia or in a bluish white that barely looks like tint at all. With sepia scenes, the color occasionally flickers to purple or red. There are very few indications that someone made a defensible choice of tinting. Interestingly, the switch to green when Christine pulls the mask off the Phantom's face is the only time where an intentional choice appears to have been made, but one I have never seen in any other print or version of the film. Considering the standard resolution of the film, the tinting makes it slightly harder to distinguish detail.
Appendix - Screen Captures
Milestone Eastman Print
Milestone Hampton Print
Image Blu-ray Eastman Print (20fps Version)
Image Blu-ray Hampton Print
BFI Blu-ray Eastman Print
BFI Hampton Print
Miscellaneous Captures
A. The Los Angeles Preview Version
As its name implies, this was a rough cut of Phantom of the Opera shown to test audiences in Los Angeles on January 7 and 26, 1925. At this point, the direction was totally credited to Rupert Julian. The critical and audience response was largely negative, and led to substantial changes. The ending is the closest to the book, as the Phantom is shown to die of a broken heart after he learns that Christine can never truly love him. An alternate ending where the Persian shoots the Phantom in his lair was shot and may also have been shown in some screenings of this version. Only this version also includes a scene at the Graveyard and Madame Valerius, both of which were in the novel by Gaston Leroux. Also like in the novel, Christine tells Raoul of her unmasking of the Phantom in flashback as they talk on the Opera Roof. At this point, Arthur Edmund Carewe's character is The Persian, just like in Leroux' book, and the Phantom's backstory is consistent with the one given in the book. This version was long enough at 22 reels to have an Intermission.
B. The San Francisco World Premiere Version
This was the second version of Phantom of the Opera shown to the public, in this case at the Curran Theater in San Francisco for a four week engagement from April 26 into May 23, 1925. This version includes newly-shot scenes directed by Edward Sedgewick. It too suffered a critical drubbing and audience apathy, and again the film was sent back for further editing. This version includes a newly-shot substantial subplot with Raoul and a Count Ruboff, played by Ward Crane, competing for Christine's affections, leading up to a duel. Added for mostly comic relief were Chester Conklin as Raoul's valet and Viola Vale as Christine's maid. The subplot occurs outside and separate from the happenings at the Opera and has been criticized as being cliched. The film's ending where the Phantom tries to escape the mob by carriage and is beaten and thrown into the Seine was shot by Sedgewick and first appears in this version. This and all subsequent versions could come in at 10 reels.
C. The New York General Release Version (General Release)
The third time was the charm for the Phantom, this version premiered at the Astor Theater in New York City on September 6, 1925 and shown well into 1926 across the United States. All of the material directed by Sedgewick for the San Francisco version was eliminated except for the new ending. At this stage, The Persian has become Inspector Ledoux of the Secret Police and the Phantom is a madman escaped from Devil's Island. Also, this version has the famously inept shot of Norman Kerry playing with his gloves as the he is eavesdropping on Phantom talking to Christine behind the wall. To the extent that any of the versions from 1925 have survived, this is the version that still exists in some substantial form (see part II-B, below). This version was reported to contain sequences of the Opera stage and stairwell, including the Bal Masque, in Prizma color and Process 2 Technicolor. It also uses the Handschliegl process for at least the Phantom's cape when he spies on Raoul and Christine on the roof of the opera.
D. The Talkie Re-release Version (Talkie Re-release)
Four years after the last footage had been shot for the silent versions of the Phantom of the Opera, Universal decided to make reissue it as a partial talkie. To do this, they reengaged Mary Philbin and Norman Kerry and some of the other original actors alongside some new actors to shoot new sound sequences that would be edited with the silent footage. Carlotta was now played by Mary Fabian, who could sing, and her mother was played by Fay Holderness for the talking scenes with the managers. Edward Martindel replaced John Sainpolis as Phillipe de Chagny except in a few long shots, and These sequences were directed in 1929 by Ernst Laemmle. Lon Chaney was under contract to MGM and Universal could not dub his voice. To get around this problem, they had another actor act as the Phantom's servant for those scenes where the Phantom talks to other people behind a wall or mirror. Once the Phantom as portrayed by Lon Chaney appears in person to Christine, intertitles resume and his servant disappears. The servant was played by Phillips Smalley. This was first released in Dallas and Detroit on January 3, 1930, but is mostly lost (see II-E, below, regarding its survival). It came in 10 reels.
II. What Survives? - Film and Disk
A. The George Eastman House Print (Eastman Print)
The best visual representation of the film in terms of quality survives in a 35mm print struck by Universal at the request of James Card, first curator of the George Eastman House in 1950. This print runs to ten reels and is about 90 minutes at 20fps and 75 minutes when run at 24fps (sound speed). It is not tinted and has no technicolor sequences. It also is apparently cropped on the left side, probably because it was duplicated on machinery intended for sound film. A sound negative in the Academy Ratio has the optical soundtrack running down the left side of the film. Although the original aspect ratio of the film is 1.33:1, due to the blocking on the left side of the image, the resulting Eastman Print has more of a Movietone aspect ratio close to 1.2:1. With sound film, the width of the film stip is the same but the left side of the film strip is taken up by the optical soundtrack. It has some moderate to severe mottling damage during the scene where Christine wakes up in her "bedroom" in the Phantom's lair.
This Print has been erroneously believed to be the Talkie Re-release, but it is not the case. Virtually all the footage in this print was shot in 1924-1925. The opening titles for the Eastman Print are from the Talkie Re-release in 1929. The "Phantom" skulking through the set during the titles is not Lon Chaney. The surviving sound discs for the Talkie Re-release do not synchronize with this footage except during Carlotta's signing from Faust.
The Eastman Print contains a puzzling sequence at the beginning of the film where a man with a lantern is walking in the cellars of the opera and "talks" to the audience. The Phantom's shadow in cape and hat also appears in this scene. The scene lasts for approximately one minute at sound speed are there are no intertitles. This caused film historian Scott MacQueen to believe that this sequence was shot so Universal could release the film internationally as a sound film. The Man with a Lantern could be attached to any foreign print speaking in the local language, and combined with the singing sequences, a foreign distributor could technically release it as a talking film. What he is saying has never been definitively established, but Phillip Riley in the Phantom of the Opera, Magic Image Filmbooks Series, Classic Silents - Volume 1, showed the sequence to a deaf person and a lip reader and one of them opined that the Man with a Lantern was speaking in German. The other person thought he could make out "Gaston Leroux". The other alternatives, albeit without any documentary support, is that this was a U.S. silent version of the Talkie Re-release for theaters that had not converted to sound or it was an internal reference print for Universal itself. No advertisements indicate that the Phantom was being offered for rental as a silent film in 1929-1930.
The actress who played Carlotta in 1925, Virginia Pearson, was redesignated as Carlotta's Mother in the Eastman Print. Carlotta was played by Mary Fabian, using footage shot for the Talkie Re-release. The opening credits give listings for both actresses, but the closing credits are from the General Release of 1925 and only list Virginia Pearson as Carlotta. The shots where Christine sings on the Opera Stage just before she is abducted by the Phantom is also from the Talkie Re-release. The notes and intertitles were also from the Talkie Re-release, although some new ones were necessary for the scenes which had talk but were silent in this print.
B. The John Hampton Show-at-Home Duplications Compilation Print (Hampton Print)
The Eastman Print is a mix of mostly outtakes and second camera takes shot in 1924-1925, B&W alternate takes of color footage and some Talkie Re-release footage. The only other near complete source for the Phantom of the Opera came from the collection of John Hampton, an early silent film preservationist who ran the Old Time Movie theater from 1940 to 1979. He showed only silent films at this theater which is still in operation and now known under new management as The Silent Theatre (and plays sound films as well, but all its silent films are played at sound speed). Back in the 1930s, Universal marketed 16mm reduction prints of some of their silent movies to the mass market. This is how 1923 film of The Hunchback of Notre Dame starring Lon Chaney survived. Hampton was able to collect some of these Show-at-Home prints to compile a mostly complete representation of the 1925 General Release of the film. According to one source, Hampton took the best elements of eight separate prints and spliced the best parts of each to make the print that is available today. Hampton's materials found their way to the Stanton Theater Foundation and are now stored at the UCLA Film and Television Archive.
This version is not at the same level of quality as the Eastman print. Detail is softer, contrast is poorer and scratches abound. It is much longer at about 115 minutes when run at 20 frames per second. No tinting or color sequences are present on the film elements. When compared, The Hunchback of Notre Dame has a better quality picture than Phantom as represented in the Hampton Print because it survives as a true Show-at-Home print and is fewer generations away from the negative. The releases of the Hampton Print currently available use scans of dupes from Hampton's prints. Hampton's original materials, which are stored at the UCLA Film and Television Archive, are superior to what is available but they have not been made generally available. UCLA announced a restoration of Hampton's original materials about fifteen years ago, but no progress has been made.
While the Hampton Print is almost totally comprised of footage from the 1925 General Release, the title sequence is borrowed from the Talkie Re-release until the card "Produced in its entirety at Universal City, California." The Bal Masque footage seen in the print is from the B&W camera, not the Technicolor camera which the 1925 General Release Version used (see below).
There is short sequence with the man with the lantern at the beginning of the Hampton Print, but it only lasts for about 13 seconds in this Print and included a medium shot of him not in the Eastman Print.
C. Trailers for the 1925 and 1929 Releases
The trailers for both the 1925 General Release and 1929 Talkie Re-release versions survive. The trailer for the Talkie Re-release is believed to contain much of the original title sequence for the 1925 film. If there was a sound disk (see section I-E., below) for this trailer then the disk has not survived. Ironically, the trailer for the 1925 General Release on most discs uses the title sequence from the 1929 Talkie Re-release. The surviving trailer film for the 1925 General Release should probably have begun with the Phantom shown over the Opera house with the names of the principal stars. Also, all the footage shown in the Talkie Re-release trailer was shot in 1924-25, no actual footage shot in 1929 was used.
The trailers contain some footage not seen in the Eastman or Hampton prints. For example, the trailers show the Phantom riding up the counterweight of the pulley mechanism that controls the mirror, which gives a more convincing explanation why Christine does not see him at first. It is easy to see why this was cut, however. With that shot, the impact of seeing the Phantom as masked is diminished.
D. Technicolor Bal Masque Sequence
In the 1970s, Film Archivist David Shepard, working for Blackhawk Films, discovered a sequence in two-strip Technicolor of the opening scenes at the Bal Masque. This film runs from the intertitle introducing the Bal Masque to the point when Christine and Raoul go onto the roof. The sequence was shot with Technicolor equipment and is different visually from the same sequence in the Eastman Print, shot using the B&W cameras. Two subsequent sequences, the first being Christine and Raoul descending from the roof and Florine encountering the Red Death have not been located but do exist in the Eastman Print in B&W.
The Eastman Print uses 1929 Talkie Re-release style intertitles and B&W footage. The Hampton Print also uses the B&W footage, but the intertitles have a curtain-style background. The Technicolor footage uses a third set of intertitles, in color with a different typeface from the first two and a different curtain image. Essentially there are three surviving versions of this sequence : Eastman B&W, Hampton Technicolor-to-B&W & Shepard Technicolor.
E. Talkie Re-Release
The 1929 Talkie Re-Release does not survive in its entirety. It had a Vitaphone soundtrack where the sound was stored on vinyl records or discs synchronized to the film projector. The 1929 Talkie Re-Release came on 10 reels and the ten accompanying sound disks survive. Of the film itself, only Reel 5 has been found and is currently stored in the Library of Congress, as are the sound disks. The reel is missing some footage, but Christine and Raoul have aged, especially the former. The Phantom's servant is visible in the newly-discovered Reel 5. Overall, there is about 1 minute and 15 seconds of new footage showing Christine and the Phantom's servant shot in 1929 and about 12 seconds of Raoul in footage shot in 1929 from this reel.
Here are the scenes with no corresponding video on the discs :
Scene 1 - Manager's Office: the Managers and the House Manager discuss why Box 12 has not been rented recently
Scene 2 - Outside and Inside Box 12: the Managers interrogate Madame Giry and investigate the occupant of Box 12
Scene 3 - Manager's Office: the Managers discuss the Phantom's demand to have Christine sign in Faust
Scene 4 - Christine's Dressing Room: Raoul asks Christine about her new-found fortune and her new admirer
Scene 5 - Opera Stage & Box: Christine sings in Faust and Phillipe warns Raoul about the Phantom
Scene 6 - Christine's Dressing Room: Raoul asks Christine again about her admirer and her Master's servant tells Christine that she should be prepared to go soon
Scene 7 - Manager's Office: Carlotta's Mother tells the Managers that her daughter will sign in Faust.
Scene 8 - Christine's Dressing Room: The Master's servant tells Christine to come through the mirror. The film for this scene has been mostly recovered.
Scene 9 - Roof of the Opera: Christine tells Raoul that she has seen the Phantom and they make plans
F. "Scream Scenes"
This is a compilation of clips from various films which was released in 1959. It shows a cut-down version of Christine's encounter with the Phantom in his lair and the unmasking scene. It's importance lies in that it contains three title cards, none of which exist in the Eastman or Hampton prints using the same text, suggesting that these scenes were taken from the Los Angeles or San Francisco versions of the film. It also contains a shot, not found on any other version, of the Phantom warming up on the organ before Christine exits the boudoir. The shots in the boudouir do not suffer from any nitrate decomposition as they do on the Eastman print. The Phantom sequence of "Scream Scenes lasts approximately 4:42.
III. What's Available - Home Video Releases
As the Phantom of the Opera is in the public domain in its raw format, anyone can scan the film and release it on some kind of home video format. There have been cheap VHS and DVD copies found in bargain bins for years. The cheapest releases are usually from an ancient transfer of the Eastman Print, sometimes a 16mm reduction print or derived from the Hampton Print.
All quality releases (with one exception) since the first Laserdisc in 1990 have two pedigrees. The first is from David Shepard, Blackhawk Films and Film Associates. The second is from Kevin Brownlow and Photoplay Productions. Shepard took the Eastman Print and inserted the recovered Technicolor Bal Masque sequence for the B&W footage in the Eastman Print. The work dates from 1990 but has been tinkered with since then.
Brownlow's version dates from 1996 and in addition to substituting the Technicolor footage, makes other changes. First, the Man with a Lantern and the closing cast credits are excised completely from this version. The justification for the first edit was likely that the sequence was superfluous to a silent film and would be puzzling for a modern audience. The second edit was done because the credits mistakenly credit Virginia Pearson as Carlotta when she was re-cast as Carlotta's mother. This version was broadcast to great fanfare on Channel 4 in the U.K. that same year and the general public would not be familiar with the issues surrounding this footage. Second, the Handschiegl effect on the Phantom's cape has been simulated using computer colorization. Third, the Bal Masque sequences after Christine and Raoul have descended from the roof have been computer colorized to approximate the two-color Technicolor look, even though the Technicolor film would have used different shots. Fourth, for the nitrate-decomposition affected scenes set in Christine's intended bedroom in the Phantom's lair, footage from the 16mm Hampton Print was substituted as it does not demonstrate the heavy mottling captured in the Eastman Print.
Shepard's version would eventually follow Brownlow's in part on the Blu-ray by recreating the Handschiegl effect on the Phantom's cape and in the Torture Chamber. The tinting schemes used by both versions have substantial differences, as described below. In their presentations of the Eastman Print, both versions leave out the B&W variation of the Technicolor Bal Masque Sequence.
The main score for the Shepard editions is from French-Canadian composer Gabriel Thibaudeau and composed in 1992, while British composer Carl Davis performed the honors for the Brownlow version in 1996. If one of these composers is identified in the version you are watching, you know from which pedigree your version comes from.
Shepard's version has been officially released on five occasions on an optical disc format, on Laserdisc in 1990, on Laserdisc and DVD in 1997 and on Blu-ray in 2011, all by Image Entertainment, and in 2015 by Kino International. Brownlow's version has been released twice, first by Milestone Film (distributed by Image) in 2003 and by the BFI in 2013. The 1997 DVD and BFI discs are region free, the 2003 DVD is Region 1 and the 2011 and 2015 Blu-rays are Region A. The latest Blu-rays are all still in print or easily available. All releases are single disc unless otherwise stated.
A. Discs Released
1. 1990 Image Laserdisc
This version runs the Eastman Print at 24fps with the Gaylord Carter Organ Score (probably the same recording as on the Image Blu-ray), but also includes the Hampton Print as an extra without a score. It contains the 1925 Trailer. This disc has long since been superseded.
2. 1997 Image Laserdisc and DVD
This version runs the Eastman Print at 20fps with the Thibeadeau score but only includes some scenes from the Hampton Print not included in the Eastman Print. It also includes the 1925 Trailer and an essay. This may have been the first silent film to be released on DVD.
3. 2003 Milestone DVD
This two-disc version runs the Eastman and Hampton Prints at 20fps. The Hampton Print is accompanied by a score from Jon C. Mirsalis. It contains the 1925 and 1929 Trailers, the latter with sound from the Talkie re-release discs, a video interview with Carla Laemmle and an audio archival interview with cameraman Charles Van Enger, still reconstructions of the Los Angeles Preview and San Francisco World Premiere versions, an extract of Faust from the 1929 feature Tiffany, nine dialogue selections from the Talkie re-release discs, the soundtrack from the talkie re-release edited to the Eastman Print, audio commentary by Scott MacQueen. The only major film sequence missing and available at the time of this release was the Man with a Lantern from the Eastman Print. The Eastman Print exhibited here has a problem with ghosting and motion artifacts believed to be the result of an improper PAL to NTSC transfer. While this DVD is long out of print, it contains many extras not found elsewhere.
4. 2009 Reelclassics DVD
This four disc set offers the Eastman Print with the Technicolor Bal Masque sequence and the Eastman Print with the B&W Bal Masque sequence. The score included for the Eastman Print is an organ score from Lee Erwin and a voice for the Man with a Lantern by John Griggs. It also offers the Hampton Print with the Technicolor Bal Masque sequence and a score by Ben Model. The print scanned for the Eastman Print is 16mm reduction from Griggs Moviedrome. Its features include a "widescreen" side by side comparison of the Eastman and Hampton Prints and the Technicolor and B&W Bal Masque sequences. Finally, it is the only release that features the "Scream Scenes" footage. These discs are printed on burned DVDs.
The contents of the DVDs are as follows :
Disc 1 - Eastman Print (from Griggs Moviedrome) with Technicolor Bal Masque sequence, The Light of Faith (short film starring Lon Chaney from 1922), Scream Scenes
Disc 2 - Eastman Print (from Griggs Moviedrome) with B&W Bal Masque sequence, 1925 Trailer
Disc 3 - Hampton Print
Disc 4 - Eastman vs. Hampton Print side-by-side Comparison, Bal Masque Color vs. B&W Comparison with commentary from, Keith Paynter
5. 2011 Image Blu-ray
This Blu-ray disc contains two HD versions of the Eastman Print. One version runs the film at a progressive 24fps at 1080p and shows noticeably less print damage. The other version runs the film at 20fps. Since Blu-ray spec only officially supports progressive frame rates at 24p or 23.976p, the 20fps version uses a 1080/29.97i resolution and telecine pulldown to support the unusual frame rate. A score from the Alloy Orchestra or the Gaylord Carter Organ score can be selected for the 24p version. The Carter score was composed in the 1970s and is synchronized to the film running at 24fps.
The 20fps version on the 2011 Image Blu-ray was interlaced using a pulldown process. The actual frame rate is 30fps, and this is achieved through interlacing. The field rate is here becomes 60fps. So, in between each two pure frames of the video, there is a "dirty frame" consisting of the odd lines of one frame of the film and the even lines of the next frame of the film. You can see this illustrated here :
| Field Number | Frame Displayed |
| 1 | 1o |
| 2 | 1e |
| 3 | 2o |
| 4 | 2e |
| 5 | 3o |
| 6 | 2e |
| 7 | 3o |
| 8 | 3e |
| 9 | 4o |
| 10 | 4e |
| 11 | 5o |
| 12 | 4e |
| 13 | 5o |
| 14 | 5e |
| 15 | 6o |
| 16 | 6e |
| 17 | 7o |
| 18 | 6e |
| 19 | 7o |
| 20 | 7e |
| 21 | 8o |
| 22 | 8e |
| 23 | 9o |
| 24 | 8e |
| 25 | 9o |
| 26 | 9e |
| 27 | 10o |
| 28 | 10e |
| 29 | 11o |
| 30 | 10e |
| 31 | 11o |
| 32 | 11e |
| 33 | 12o |
| 34 | 12e |
| 35 | 13o |
| 36 | 12e |
| 37 | 13o |
| 38 | 13e |
| 39 | 14o |
| 40 | 14e |
| 41 | 15o |
| 42 | 14e |
| 43 | 15o |
| 44 | 15e |
| 45 | 16o |
| 46 | 16e |
| 47 | 17o |
| 48 | 16e |
| 49 | 17o |
| 50 | 17e |
| 51 | 18o |
| 52 | 18e |
| 53 | 19o |
| 54 | 18e |
| 55 | 19o |
| 56 | 19e |
| 57 | 20o |
| 58 | 20e |
| 59 | 21o |
| 60 | 20e |
o stands for odd, e stands for even and the highlighted pair are the dirty frames.
The Thibaudeau score or an audio commentary from Jon C. Mirsalis can be selected for the 20fps version. This Blu-ray contains a tinted version of the Hampton Print, but it is presented in Standard Definition NTSC resolution (720x480) only. It is accompanied by a piano score from Frederick Hodges. The 1925 Trailer is included and an interview with composer Thibeadeau, both also in SD. The Fifth Revised Shooting Script and Souvenir Program are also included as a succession of still images. The contents of this disc were released by Park Circus in the U.K. on a Region B disc.
The first pressing of this disc had a non-intuitive menu where you could only select the versions of the prints you wanted to watch by selecting the score attached to that version. To change to another version you had to eject and reinsert the disc or jump ahead to the end. Also, all stereo scores were output in mono and the Carter score was out of sync for Reel 1. The ballet dancers in the first ballet sequence also appear to be moving in slow motion in both the 24 and 20fps versions. A replacement disc fixes the stereo/mono issue and uses a new, more intuitive and technically impressive menu.
6. 2013 BFI Blu-ray Dual-Format
This package contains one Blu-ray disc and two DVDs. Although this post and blog is written from the perspective of someone in the United States, I am not adverse to including information about a new and substantial release elsewhere in the world. The Blu-ray disc and the two DVDs included are region free, but the DVDs are PAL and may not play in a U.S. player. The Blu-ray and the first DVD have identical content, with the Eastman and Hampton Prints and the extras, the second DVD just has the documentary Lon Chaney : A Thousand Faces.
Both the Eastman and Hampton Prints are run at 20fps, but in both cases the 1080p/24fps mode is used on Blu-ray. This is a dual-format release, so the prints are presented on the Blu-ray and one of the DVDs. On the DVDs, all video is shown at the 50i PAL standard. The film is sped up 4% as a result, but the audio is pitch corrected. This means that one out of every five frames is repeated to make up the missing 4fps. The Hampton Print is in HD, unlike the Image Blu-ray, but is not tinted. The Technicolor Bal Masque sequence in the Eastman Print is windowboxed, as is the sequence where they used a 16mm print for the scene with Christine in the boudoir, presumably to match the Movietone-like aspect ratio of the rest of the Eastman Print. The shots of the Phantom on the roof with the Handscliegl effect applied are cropped and somewhat grainier than the rest of the film.
The BFI Eastman Print is based on the 1996 Brownlow restoration. This restoration was done at standard definition at the PAL format. For those sequences where the 16mm Hampton Print is used to replace the damaged scene with Christine in her bedroom in the Phantom's lair, the three color titles at the Bal Masque, the Handscliegl process on the Phantom's cape and the colorized portion of the Bal Masque, all these sequences have been upconverted from SD materials.
The two trailers are included, and the recently discovered material from Reel 5 of the 1929 Talkie Re-release is included synchronized with its sound disc. The Man with a Lantern from the Eastman Print is included as an extra, the sequence remains excised from the main presentation. All video footage on the Blu-ray is presented in HD, scanned at 2K. The DVD uses the same scans, downscaled from HD into SD PAL. The short sequence of the Man with a Lantern from the Hampton print is deleted in this version. Also, the closing credits card in the Eastman Print for the cast is deleted.
The second DVD contains the documentary Lon Chaney: A Thousand Faces, first broadcast for Turner Classic Movies in 2000. It is directed by Kevin Brownlow and narrated by Kenneth Branagh. A booklet accompanies the discs describing the production and restoration of the film.
7. 2015 Kino-Lorber Blu-ray or DVD
This Blu-ray is something like a re-issue of the now out of print Image Blu-ray, but comes on two Blu-rays. A DVD version is also available. The first Blu-ray has 24fps and 20fps versions of the Eastman Print like the Image Blu-ray, but nothing else. The 20fps version is now using the 24p progressive frame rate supported by the Blu-ray format. The 20fps version also appears to have been noticeably cleaned up compared to the Image Blu-ray and even the 24fps version has minor improvements and corrects tinting errors. Because these versions share a dual-layered disc, their bitrates are higher than the older Image versions.
The 2015 20fps version is displayed progressively at 24p (or 23.976p). In order to do this, every fifth frame is repeated once to make up the extra 4 frames needed by the format. The reason this is done for this and many other silent films is because the Blu-ray format does not officially recognize any other progressive frame rate except for 59.94p at 1280x720.
The 24fps version on the Image and Kino Blu-rays came from Paul Killiam's film collection. Killiam was a comedian in vogue in the 1950s who did much to help preserve silent film. His print was duplicated directly from the Eastman Print and I believe eventually found its way into David Sherpard's Blackhawk Film archive.
The 20fps version on the Kino Blu-ray is a slowed down version of Killiam's print. The old 20fps version on the Image Blu-ray was duplicated from Killiam's copy, making it as a source a generation further away from the Eastman Print. This is one reason why it looks rather rough. Neither version on the Kino disc shows the tint colors in the pillarboxes on the sides of the video, which was a flaw on the Image Blu-ray.
There was a sychronization issue with the Thibedeau score on the Image Blu-ray which was mostly fixed on the Kino Blu-ray. The only issue that was not fixed was the slow motion ballerinas. The ballerina sequence at the beginning of the film was stretch printed on the print Image and Kino used. Stretch printing slow downs the action shown on film by duplicating frames when prints are made. Fortunately this issue lasts only for a minute or two (3:04 to 4:42 on the Kino 20fps version for example) and does not affect any other portion of the film. I think the ballet music used in this scene goes on for too long and spills into the next scene. The Alloy Orchestra's score on the 24fps version is much better timed to the action.
Unfortunately, the Carter score on the 24fps version is totally out of sync for the first 11 minutes of the film. The correct music plays about a minute and a forty-five seconds later than it should. The music for the ballerinas start to play at the end of the next scene with the managers. This sync issue may have occurred on the first version of the Image Blu-ray but did not on the second.
All the extras and the Hampton print are on the second Blu-ray disc, which is also dual layered. The extras are mostly the same as on the Image Blu-ray with three extras.
The Hampton Print is now in 1080i, but it is an upscaled version of the 480i version found on the Image Blu-ray. However, it is better to have the upscaling done by professional software instead of the Blu-ray player or the TV, and the film uses much more of the disc on the Kino than the Image.
Essentially, if the old Image disc is version 1.0, the newer image disc is version 2.0, then the Kino disc is essentially version 3.0.
B. Video Quality
1. Eastman Prints
All discs except for the 1990 Laserdisc have a speed-corrected transfer of the film to 20 frames per second. except where specified as running at 24 frames per second. Sound films use 24 frames per seconds, but silent films generally were projected at 16-22 frames per second. Milestone's Eastman Print has a serious issue with ghosting and jittery movement, probably due to frame rate differences between the PAL video master and its conversion to NTSC video. Image's Blu-ray versions suffer from some unnatural slowdown during the opening ballet sequences and speedup in the first scene in the manager's office, but are otherwise without motion issues (unless they are in the original print).
As befitting Blu-ray, the Eastman Print on either disc looks much sharper than any DVD or Laserdisc due to the high definition transfer. One example of this is that it is easy to identify in which shots (except for long shots) Chaney is wearing his mask with painted on eyes and which shots he is using a different mask with cutouts for his eyes. However, sometimes the focus is soft, so you should not expect razor sharp clarity in every frame.
As has been noted by others, the Image 24fps version had some amount of post-processing work done to reduce the scratches and wear on the print. However, no attempt was made on either Image version to remove large hairs that pop in the edges of the frame, large scratches or a broken film perforation fragment which appears as Christine meets the Phantom for the first time. People have stated that the Image 20fps version, while looking more worn, also looks more natural and less processed. Milestone and BFI seem to have made attempts to clean up their video presentation of the major blemishes and scratches. There is no broken film perforations in these versions.
Milestone and BFI substituted 16mm footage from the Hampton print for most of the scene where Christine is in her room in the Phantom's lair. This was due to the nitrate decomposition damage already present on the Eastman Print. Image DVD and Blu-ray used the 35mm footage instead. The 35mm footage here is not as sharp or clear due to the mottling, but still sharper and has better continuity with the 35mm footage preceding and following it.
Image Laserdisc, DVD and Blu-ray have the Man with a Lantern sequence, Milestone does not and BFI has it as an untinted extra. When Photoplay Productions created the video broadcast master used by Channel Four, Photoplay or Channel Four must have deleted it as it would be baffling to the uninformed viewer.
Image also has the cast credits at the end of its Eastman Print, Milestone and BFI do not.
Image Laserdisc, DVD and Blu-ray have no Technicolor footage after the scene on the roof of the opera. The footage is tinted. Milestone and BFI has colorized all the footage of the Bal Masque occuring after the scene. Image Blu-ray recreated the Handschiegl process for the face of the Rat catcher and to visualize the heat in the torture chamber. While I was previously aware of the possibility of some Handschiegl work for the rat catcher's scarf, I was not aware that there was the possibility of the Handschiegl process being used for his face or the torture chamber. Both recreate the process for the Phantom's costume when he is on the roof, but the result is noticeably different.
Milestone's left border is fuzzy, but Image's is sharper. Image's may have a sliver less of information in the video frame as a result. The BFI has a sharp border and a sliver more information than Image. The Technicolor footage of the Bal Masque came from a different source and will be in a true 1.33:1 ratio.
While the tinting is the same color on the Image 24fps and 20fps there are fluctions in the saturation levels between the two prints in some scenes. The scene where we first encounter Florine Papillon, Snitz Edwards' character, is very orange in the 24fps and much more sepia in the 20fps. The BFI Blu-ray tints are often extremely saturated, especially the blue for night scenes and the purple for opera cellar scenes.
There is a frame rate issue inherent in the Eastman Print. The vast majority of the footage was shot in 1925 at silent speed. This should be projected correctly at 20fps. The footage shot in 1929 would have been shot in sound speed. This means that the Opening Titles (where the phantom is skulking in the frame), The Man with a Lantern scene and the shots of Mary Fabian as Carlotta singing should be ideally projected at 24fps (especially for the latter if it is to be synchronized to the existing audio. The shots where Christine is singing just before she is abducted by the Phantom may also be from 1929. The BFI handles this by running these sequences at a true 24 frames per second while running the rest of the film at 20 frames per second.
2. Hampton Prints
Milestone is a straight black and white presentation of a show-at-home print. Previously this version existed on the 1990 Laserdisc and really poor public domain releases. I would advise lowering the contrast a bit when watching it. There are combing artifacts due to the interlacing.
Image apparently added tints to their presentation. Universal would not have gone to the extra expense of tinting 16mm inexpensive show at home prints. Both Milestone and Image are in standard definition.
Despite what the back of the Image Blu-ray case says, they used a 16mm source for the 1925 version. (I am not even sure if motion picture film exists in a "6mm" gauge.) In my opinion, Milestone boasts a slightly sharper image due to the lack of tinting (see below for more detail).
D. Audio Options
1. Milestone Eastman Print
a. Carl Davis Score
This is an orchestral score performed with the City of Prague Philharmonic in 1996. Davis has had a long history composing scores for films both silent and sound and for television. Score is in stereo.
b. 1929 "Synchronized" Soundtrack
In this version, the producers have tried to match the sound disks for the Talkie Re-release as best they could to the action on-screen. There is music playing throughout the film as composed or compiled in 1929 and there is voice acting roughly at places where it should be. The intertitles are still shown of course. While PowerDVD reports this track as being in stereo, the soundtrack is obviously monaural.
One issue to note is that the film which this track accompanies is being projected at 20fps, but the dialogue and music for the talking version was recorded at 24fps. This means that the music and dialogue is being played back at a 20% slower speed than it should be playing. It sounds a bit creakier than it should.
One issue to note is that the film which this track accompanies is being projected at 20fps, but the dialogue and music for the talking version was recorded at 24fps. This means that the music and dialogue is being played back at a 20% slower speed than it should be playing. It sounds a bit creakier than it should.
c. Scott MacQueen Commentary
MacQueen's commentary shows that he is an authority on the film. Its very entertaining to listen to and possesses a nearly encyclopedic knowledge of the film's production. He tells the story of Phantom from its beginnings in Gaston Leroux's novel through the 1925 and 1929 versions, interrupting to comment when something is on the screen. He was primarily responsible for the 1929 soundtrack.
2. Milestone Hampton Print
a. Jon C. Mirsalis Score
This is the same individual who did the commentary on the Image. His website is http://www.lonchaney.org/. My favorite score for the 1925 version of the film and is tied with the Carter/Killiam score for my all-time favorite score for Phantom.
3. Image Eastman Print/24fps
Unfortunately, Image's scores can be heard only with their respective versions. The scores offered for the 24fps version cannot be heard with the 20fps version and vice versa. This is due to the fact that these scores and commentary were intended for a particular running time. The 24fps version runs 77 minutes long, the 20fps version 91 minutes. The 1925 version of the film runs for 124 minutes.
a. Alloy Orchestra Score
The Alloy Orchestra is a trio of artists who specialize in composing scores for silent films. They are based in Cambridge, Massachusetts. I have listened to some of it, and its a bit too modern for my taste in silent music scores. Their website is http://www.alloyorchestra.com/. They use synthesizer keyboards, musical saw, clarinet, accordion, banjo and what they call "junk percussion". Score is in stereo.
b. Gaylord Carter Organ Score
Composed and performed by Gaylord Carter in 1974 and copyrighted in 1989 by Kino. Carter's career began in the early 1920s, accompanying silent pictures with his theater organ. He was able to find work with sound films, radio and television and performed into the 1990s. In the 70s he scored many silent pictures released by Paul Killiam's Film Classics. Score is in stereo.
The credits for Image indicate that Carter was actually playing at the organ. The version I grew up with from Killiam's Film Classics gave credit to Richard Simonton for Theater Organ. This version was copyrighted in 1971, Was Simonton the organist or the provider of the organ? Image uses a different recording that Carter made for Blackhawk Films instead of the one he made for Killiam.
I was bitterly disappointed when the instrumentation I heard on Image was different than on Killiam's release. I believe Killiam's is vastly superior, uses the organ in a much more "fuller" way and is my personal favorite of all the musical accompaniments for Phantom. The score, except for the first little bit, can be found here : http://www.youtube.com/watch?v=Wp4e8cWp1xQ&list=PLY2WnIaqPW97t53BbiUZe8uPOecqkkPKa
The first bit with the Film Classics opening can be found here : http://www.youtube.com/watch?v=f_xaXGS8j00&list=UUd7WChc4LjvA-QLBkKEWQSA&index=11&feature=plcp
4. Image/Kino Eastman Print/20fps
The credits for Image indicate that Carter was actually playing at the organ. The version I grew up with from Killiam's Film Classics gave credit to Richard Simonton for Theater Organ. This version was copyrighted in 1971, Was Simonton the organist or the provider of the organ? Image uses a different recording that Carter made for Blackhawk Films instead of the one he made for Killiam.
I was bitterly disappointed when the instrumentation I heard on Image was different than on Killiam's release. I believe Killiam's is vastly superior, uses the organ in a much more "fuller" way and is my personal favorite of all the musical accompaniments for Phantom. The score, except for the first little bit, can be found here : http://www.youtube.com/watch?v=Wp4e8cWp1xQ&list=PLY2WnIaqPW97t53BbiUZe8uPOecqkkPKa
The first bit with the Film Classics opening can be found here : http://www.youtube.com/watch?v=f_xaXGS8j00&list=UUd7WChc4LjvA-QLBkKEWQSA&index=11&feature=plcp
4. Image/Kino Eastman Print/20fps
a. Gabriel Thibaudeau Orchestral Score
This score was composed in 1992 with the I Musici de Montreal and included vocals from a female soprano. Personally I am not a fan of it. This is the score also found on the 1997 Image DVD. Score is in stereo.
b. Jon C. Mirsalis Commentary
Mirsalis is an unapologetic admirer of Lon Chaney. The longer his commentary goes on, the more it focuses on Chaney's life and career and less on the film. It is a nice complement to MacQueen's commentary. He almost certainly makes an error when he states that the General Release version premiered in San Francisco in September, 1925. It premiered in New York City. He also makes some unwarranted speculation that the Bal Masque sequence may have been shot in Prizmacolor. There is no substantial doubt that it was shot in two-strip Technicolor.
5. Image/Kino Hampton Print
a. Frederick Hodges Piano Score
This score was commissioned specifically for Image. Score is in stereo.
6. BFI Hampton Print
a. Ed Bussey Piano Score
This score was commissioned specifically for BFI. Score is in stereo. Neither the Image or BFI piano scores for the Hampton Print come close to Mirsalis' score, in my opinion, on the Milestone disc.
E. Extras (Not including Audio Options)
1. Milestone Eastman Print (Disc 1)
a. 1925 Theatrical Trailer
Untinted, silent, has few extra seconds compared with Image.
b. 1929 Theatrical Trailer
Has synchronized soundtrack, tinted in sepia.
c. Still Gallery
Los Angeles Premier Version photo Reconstruction, San Francisco Premier Version photo Reconstruction, Posters & Lobby Cards, Promotion, Press Kit, The Phantom Stage, Publicity Portraits, Backstage Stills, Concept Art, Publishing
d. Talking Extracts from 1929 Sound Version
Consists of nine sequences, most of which were not included in the synchronized sountrack option. Sequences are Managers' Office I, Box 12, Managers' Office II, Garden Love Scene, Faust - Trio and Finale, Christine's Dressing Room, Managers' Office III, Through the Mirror, Apollo's Lyre. The length of many of these scenes is longer than their equivalent silent counterparts and are noticeably different. Here is an example :
In the Box 12 excerpt, the managers speak to the maid, enter the box to find the Phantom, exit and talk to the maid again, enter to put the Phantom out, but when they find the box empty, they exit and tell the maid to tell no one. This takes approximately 2 minutes to accomplish at sound speed. In the silent version, the managers only speak to the maid once and are able to finish the scene in 1 minute, 40 seconds at silent speed.
e. Scene Selections - 24
2. Milestone Hampton Print (Disc 2)
a. Video interview of Carla Laemmle by David J. Skal
Carla Laemmle was the niece of Carl Laemmle Sr., the President of Universal Studios. Laemmle put his relatives on the payroll, and Carla played the part of head ballerina in Phantom. She also plays a passenger in the coach in the opening scene of Universal's 1931 production of Dracula. Her interviewer is David J. Skal, a noted researcher into Universal's horror films of the 30s and 40s. He has written several books on the subject of horror films.
b. Faust opera extract from film Midstream
Contemporary footage of a the 1929 film Midstream which features Charles Gounod's opera Faust, which is so prominently featured in Phantom. It gives you an idea what 1929's sound technology could do when am opera scene needed to be filmed.
c. Audio interview with Charles Van Enger, ASC
This interview was recorded on July 29, 1973. Enger was the principal cameraman on Phantom and the interviewer was Richard Koszarski. He had a low opinion of Director Rupert Julian and respected Lon Chaney.
d. Scene Selections - 29
3. Image Blu-ray
This interview was done for a PBS station and reveals that Thibeadeau enjoys accompanying screenings of silent music with a piano.
b. Still Gallery
Unlike Milestone, the stills on this disc are not given a submenu where they are broken down by categories. Some stills are very sharp, others look like they were scanned from a book.
c. 1925 Theatrical Trailer
Tinted in blue and purple, uses Hodges piano score for music. Missing ending shot of crowd outside Astor Theater.
d. Script
This is the Fifth Revised Shooting Script by Elliot Clawson and Raymond L. Shrock. Each page is shown for 3-4 seconds. The script can also be found in Philip J. Riley's book Phantom of the Opera : Hollywood Archives Series Vol. 1. It is still in print and can be found here : http://www.amazon.com/Phantom-Opera-Hollywood-Archives-Series/dp/1882127331/ref=sr_1_cc_1?s=aps&ie=UTF8&qid=1347579685&sr=1-1-catcorr&keywords=phantom+of+the+opera+riley
e. Theatrical Souvenir Program Reproduction
Presented identically to the script. A Blu-ray's zoom function is useful here.
f. Scene Selections - 12 per version
4. BFI Blu-ray/DVD Combo
a. 1925 General Release Trailer
No tint, uses 1929 Title Card (with the title merging to the center). Uses 20 frames per second within 24p on Blu-ray. Interestingly, you can see in this extra what the title card looked like without the cropping of the Eastman Print.
b. 1929 Talkie Re-release Trailer
Sepia tinted, uses 24 true frames per second speed. The music comes from the Talkie Re-release discs, but is different than the same trailer on the Milestone.
c. Reel 5 of the 1929 Talkie Re-release
No tint for the footage that comes from the true Reel 5, Sepia tinted for scenes taken from the Eastman Print to complete the reel. There are two places where the frame goes black for a second or two, one of these is where Christine speaks the line "Oh my, I'm afraid." The Phantom's servant can be seen clearly speaking the line "Come through the mirror, have no fear." 24 true frames per second on Blu-ray.
d. Man with a Lantern Sequence
No tint, 24 true frames per second on Blu-ray.
e. Channel 4 Silents restoration programme
This is included as a PDF file on the first DVD. This programme was issued to theatre patrons when the 1996 Brownlow restoration premiered theatrically in London in 1926. It gives the history of the production, the restoration, biographical summaries of Chaney, Julian, Philbin, Kerry and discusses the Davis score.
f. Lon Chaney - A Thousand Faces Documentary
On the second DVD, first broadcast by Turner Classic Movies in 2000, an 86-minute documentary on the life and films of the great actor. PAL Format. As this documentary was produced by Kevin Brownlow in the U.K., the 25fps PAL format is not an inappropriate video standard.
g. Booklet
This booklet contains articles from Kevin Brownlow, Patrick Stanbury and Geoff Brown describing their experiences with the Phantom. There is also a description of the features of the discs and production credits for the film and the Chaney documentary. Finally, there is technical details about the transfer.
E. Kino Blu-ray
a. Interview with Composer Gabriel Thibeadeau
See description for the Image disc above, upscaled to HD.
b. 1925 Theatrical Trailer
See description for the Image disc above, upscaled to HD.
c. Script
This is the fifth revised shooting script, although this version is presented as scrolling text set to the Thibeadeau score.
d. Travelogues
Two travelogues by Burton Holmes depicting Paris in 1925, Paris from a Motor and A Trip on the Seine. I am not sure how relevant this is supposed to be for a movie set closer to 1885 than 1925, but movie cameras were still being invented in 1885!
e. Talkie Excerpts with Reel 5
This contains a shortened version of the film in HD running 53 minutes and using a fair amount of the existing audio from the 1929 Talkie Disks. This is superimposed over the Eastman Print footage and the recently found Reel 5 of the 1929 Talkie version. When there is dialogue with no corresponding film, there will be a black screen with explanatory text. This is similar to how it was presented on the Milestone DVD, but the Milestone DVD provided the whole film with the soundtrack, it was not truncated as it is on the Kino Blu-ray. The Milestone DVD has all the talking scenes from the 1929 Talkie Disks available to be heard individually, the Kino Blu-ray does not present all the audio or the talking scenes.
F. Defects in and Changes from the First Pressing of the Image Blu-ray
When the Image Blu-ray disc was first released, it was discovered that the disc was defective. First, all stereo scores ended up in monaural. There was no recall of the discs, so owners of the first release had to have contacted Image to secure a replacement disc. Image has since merged with Acorn Media and replacement discs are no longer available. The Gaylord Carter score was out-of-sync for the first couple of minutes. Finally, the old menu required you to select the version of the film by selecting the score. So to watch the 20fps version, you had to select the Thibeadeau score. You would have to know which score went to each version. Moreover, to change to a different version you had to either eject and reinsert the Blu-ray or chapter select and fast forward to the end of the version. The new menu lets you choose the version first, then the score. If you return to the main menu, you can choose another version. Here are a screenshot of the old and new menus :
 |
Old Menu - You could have obtained an replacement disc if this is what you see
|
 |
New Menu - Much better organized
|
G. Tinting Differences
The Shepard and Brownlow restorations and the Hampton Print on the Image Blu-ray have different tinting in several sequences. Brownlow barely uses tint in the Phantom's lair on the Milestone and these scenes are completely black and white on the BFI. Shepard uses sepia tinting for these scenes, and I could find no purple tinting at all. Brownlow uses purple for the opera cellars. Image has no tinting for the first intertitle "Sanctuary of the song lovers..." or for the intertitle and the short scene where Raoul reads Christine's note just prior to the Bal Masque. This confirms that the original 35mm Eastman Print is in black and white. Additionally, when Shepard uses sepia tinting, the whole screen, including the borders (since the resolution is 1920x1080 regardless of aspect ratio), is sepia toned on the Image Blu-ray. This was fixed for the Kino Blu-ray. Note that below that the Kino 20fps uses the same tinting choices as the 24fps Kino and Image sources, they all derive from the same source. David Shepard has confirmed that the telecine they worked from was in black and white except for the Technicolor footage. Here are the differences :
1. Shepard Eastman
Opening Titles - Green
Man with a Lantern - Red
Intertitle - "Sanctuary of Song Lovers..." - White (untinted)
Exterior of Opera House - Blue
Grand Staircase - Sepia
Cellars w/Florine - Orange (24fps) / Sepia (20fps)
Scene Dock 3rd Cellar - Green
Florine & Simon Bouquet - Sepia
Intertitle - "From hidden places" - White
First Scene with Phantom behind Christine's Mirror - Sepia for Christine, Green for Phantom (including respective intertitles)
Second Scene with Carlotta & Managers - Sepia
Phantom in Chandelier Scene & Intertitle - Green
Second Scene with Phantom behind Christine's Mirror - Sepia for Christine, Green for Phantom (including intertitles)
Christine Meets the Phantom - Green
Christine Descends the Staircase - Sepia (just after the piece of sprocket)
Christine Daae' Disappears - Red
Christine Awakens in Lair - Sepia
Raoul Reads Christine's Note - White
Bal Masque - Technicolor
Rooftop Scene - Blue
Christine and Raoul Encounter Ledoux - Sepia
Raoul Baroche Scene - Blue
Faust Opening - Sepia
Raoul, Ledoux and Phillipe in Cellars -
Torture Chamber - Green
Phantom's Lair - Sepia (Generally)
Phantom swims in the Black Lake - Green
Mob Outside Opera - Blue
Heated Torture Chamber - Red & Yellow
Gunpowder Room - Green
Mob in Cellars - Orange (until they get to the Phantom's door)
Phantom's Carriage Ride - Blue
Finis - Sepia
2. Brownlow Eastman
Opening Credits - Blue
Opera Staircase - Sepia
Ballet Scenes - Red
Manager's Office - Sepia
Opera Basement w/Florine - Purple
Florine & Simon Bouquet - Sepia
Phantom Speaks to Christine - Purple (Phantom's Shots and Intertitles only)
Phantom Removes Chandelier Rope and Intertitle - Purple
Christine & Phantom in Cellars - Purple
Phantom's Lair - Light Green or White (untinted)
Christine Disappears & Christine's Room in Lair- Sepia
Raoul Reads Christine's Note - Sepia
Bal Masque - Technicolor
Rooftop Scene - Blue with Green Intertitles
Bal Masque - Technicolor (Recolorized)
Ledoux at Prefect's Office - Sepia
Raoul's Baroche - Blue
Opera Stage - Sepia
Florine Finds Joseph's Body - Purple
Opera Stage - Sepia
Raoul, Ledoux & Phillipe in Cellars - Purple
Phantom's Lair - Light Green
Torture Chamber - Purple
Torture Chamber Activated - Sepia
Mob Outside Opera - Blue
Gunpowder Room - Purple
Mob in Cellars & Lair - Red
Phantom's Carriage Ride - Blue
3. Image/Kino Blu Hampton
Opening - Blue
Exterior of Opera House - Sepia
Christine Meets the Phantom - Blue
Phantom's Lair - Sepia
Phantom Unmasked - Green
Raoul's Message - Blue
Bal Masque - Sepia
Raoul and Christine Evade Phantom - Blueish White
Return to Bal Masque - Light Red
Ledoux in Prefect's Office - Sepia
Raoul, Ledoux & Phillipe in the Cellars - Blueish White
Until :
Phantom Demands Christine Bring Him Love - Sepia
Raoul & Ledoux Fall into the Torture Chamber - Bluish White
Phantom Returns to His Lair - Sepia
Phantom's Carriage Ride - Blue to Bluish White
Honeymoon at Viroflay - Sepia (Not including intertitle)
4. Killiam Film Classics Eastman
Opening - Blue
Opera Staircase - Purple
Manager's Office - Sepia
Opera Stage - Purple
Box 5 Exterior - Sepia
Box 5 Interior - Purple
Opera Basement - Purple
Christine's Dressing Room - Sepia
Phantom Speaks to Christine - Green (Phantom's Shots and Intertitles only)
Phantom Removes Chandelier Rope and Intertitle - Green
Outside Christine's Dressing Room - Sepia
Christine & Phantom in Cellars & Lair - Green
Christine Disappears - Sepia
Christine's Room in Lair - Green
Raoul Reads Christine's Note - Sepia
Bal Masque & Opera Upper Floors - Purple (Technicolor footage not used)
Rooftop Scene - Blue (no Handschiegl)
Ledoux at Prefect's Office - Sepia
Raoul's Baroche - Blue
Florine Finds Joseph's Body - Purple
Raoul, Ledoux & Phillipe in the Cellars & Torture Chamber- Green
Mob Outside Opera & Chase, Finis and End Credits - Blue
5. Observations
Brownlow is more consistent with its tinting than Shepard. All outside evening scenes are blue and all opera cellar scenes are purple with only one exception for each. Indoor scenes are sepia with the exception of the ballet and the phantom's lair. Beginning credits and end title are blue. The green in the Phantom's Lair scenes is so light as to easily confuse it for not being colorized at all. Apparently the different tints in the rooftop scene may have been original to the print, as it is the only scene where the intertitles are a different color than the action and all the speakers are in the same room (contrast to Christine and the Phantom talking to her behind the mirror when she is in her room).
Brownlow is very consistent in tinting with the Paul Killiam Film Classics version I grew up with. The Killiam Eastman Print has a restoration credit to Paul Malkames. Malkemes was a noted film preservationist who helped preserve dozens of silent films and some early sound films like the 70mm version of The Big Trail, which is notable as being an early sound film shot in widescreen, on large gauge film and John Wayne's first starring role. Malkemes was an expert in silent film cinematography and well-qualified to judge tint. His tinting is very simple, blue for outside the opera, sepia for rooms and offices, purple for the opera stage and cellars (initially) and green for anything having the phantom in it and the cellars after Christine's abduction from the stage.
Shepard's tinting takes a more dynamic approach at times, although there is more sepia than in Brownlow's. In both restorations we are faced with the respective restorer's choices regarding tinting. The major choices are at least fifteen years old for both. However, there are obvious errors in the Shepard versions as two intertitles and one short scene identified in above are not tinted and seemingly should be. I would love to know the decisions behind each restorer's tinting choices. While the fifth revised shooting script contains tinting indications, this does not necessarily mean they were followed through to the General Release. The fifth revised shooting script's tints are very simple, blue for nighttime outdoors, amber for the grand staircase, opera stage and boxes, office and dressing rooms and the Phantom's lair, green for the cellars and a little red in the torture chamber.
The Image Blu-ray Hampton Print seems like someone started to tint the film but did not finish. Virtually all the film is in sepia or in a bluish white that barely looks like tint at all. With sepia scenes, the color occasionally flickers to purple or red. There are very few indications that someone made a defensible choice of tinting. Interestingly, the switch to green when Christine pulls the mask off the Phantom's face is the only time where an intentional choice appears to have been made, but one I have never seen in any other print or version of the film. Considering the standard resolution of the film, the tinting makes it slightly harder to distinguish detail.
H. Conclusion
In an era of high definition televisions and Blu-ray players, Phantom is well-served with two quality, accessible releases that, between them, eliminate any need to keep DVDs, except for the extras on the Milestone and Reelclassics DVDs.
Milestone Eastman Print
Milestone Hampton Print
Image Blu-ray Eastman Print (20fps Version)
Image Blu-ray Hampton Print
BFI Blu-ray Eastman Print
BFI Hampton Print
Miscellaneous Captures
 |
| BFI Blu-ray Talkie Re-release Christine Reel 5 |
 |
| BFI Blu-ray Talkie Re-release Phantom and Servant |
 |
| BFI Blu-ray Talkie Re-release Raoul in Christine's Dessing Room |
 |
| BFI Blu-ray Eastman Print Man with a Lantern |
 |
| Image Blu-ray Eastman Print Man with a Lantern |
 |
| Image Blu-ray Hampton Print Man with a Lantern |
 |
| Milestone DVD Hampton Print Man with a Lantern |
 |
| BFI Hampton - Christine in her Bedroom (Compare with BFI Eastman Print Shot #8) |
 |
| Griggs Moviedrome Release Credits #1 (shown after Chaney/Philbin/Kerry Credit |
 |
| Griggs Moviedrome Release Credits #2 (shown at the end of the film) |
 |
| Killiam Film Classics Release Credit #1 (shown before original main titles) |
 | |
|
 | |
|
Subscribe to:
Posts (Atom)