CBMSTEVE.CA - Personal pages of Steve J. Gray |
Changes | Info | Prototypes | Remakes | PET Projects | CBM Projects | Modding | 3D Models | KiCad | Github | Ohio Scientific | Other | My Software | My Collections | Facebooks |
![]() |
Welcome to the Commodore CBM-II pages of Steve J. Gray! This page contains articles, news, technical information and pictures of what some may claim as the sexiest machines Commodore ever released before their ultimate demise...
The CBM-II line was Commodore's followup to their original PET/CBM machines. Enhancements included extra memory, better screen editor, three-voice sound (SID), RS-232 port, cartridge port, expansion port, and a new keyboard with function keys... all in a sexy new case.
Originally the line was to be split into a Business line (B-series) and a Professional/Scientific line (P-series). The P-series was ultimately dropped due to the success of the Commodore 64. A few P-series machines were released to dealers and a few were sold before Commodore decided to cancel the series and recall the dealer units. Luckily I managed to aquire one and you can find some pictures below.
In 1981 Commodore started planning a followup to their aging 6502-based PET line. The PET had 32K, a 6845 video controller, and barely any sound. The VIC-20 had just been released and was doing well. The VIC-20 used a 22 Column colour video chip, the VIC-I. MOS Technologies were developing newer chips; A 40 column colour chip called 'VIC-II', and a sound chip called 'SID'. By this time 32K memory just didn't seem enough so they modified the 6502 chip to support more memory (1 Megabyte) and so the 6509 CPU was born. Around the same time IBM entered the PC market using the 16-bit Intel 8088 chip.
Work started on three lines at the time: The new advanced PET line, a game machine, and another secret project. These were to become the CBM-II series, the Max Machine, and the C64 respectively. We know what happened to the C64... The Max machine was released in Japan but never took off. And the CBM-II line would ultimately split into two lines, the B-series and the P-series. The B-series was, once more, split into both High-Profile and Lo-Profile models.
The CBM-II machines were introduced at the Summer 1982 CES show. There was the P500 Professional Computer for US$995, and the B700 Business Computer for US$2,995. Later, at the Winter CES show the B128 was announced for US$1695. The B700 became the BX256 with built-in 8088 coprocessor board and built-in dual drives. In addition two external drives, the SFD-1001 single and 8250LP dual models, and two printers, the 6400 and MPP-136 were announced with styling/colouring to match.
Due to various factors, such as engineering problems and delays, and the popularity of the C64, Commodore could not produce many CBM-II machines. The P500 was cancelled before going into general production; only pre-production machines exist. The BX machines were never released due to problems with the 8088 board. Eventually, in 1983 some machines were released. Shorly after, in 1985 the remaining were liquidated. By 1987 all machines had been sold and Commodore handed over all information and documentation (see agreement) to the Chicago B User's Group (CBUG). When Commodore went bust some prototype machines, pre-production machines, 8088 boards, and other variations were discovered.
This was not to be the end of the CBM-II. In the labs at Commodore was another machine. This machine started as a merging of the B-series and the P-series, with 6509 CPU and BOTH the VIC-II and CRTC chips for video. The machine was known as the D128 (perhaps "dual" video). However, the machine was not C64 compatible and the C64 was Commodore's greatest success. It was decided that any new machine should be compatible to the C64, and so the machine was re-engineered and became the C128.
Commodore went through many model number changes for the CBM-II line. Products were announced and then changed (see transactor article). The 128/256 naming convention was generally used for the North American markets using the "CBM", "CBMX", "B" or "BX" designations to indicate the case type and the existence of a coprocessor board (X=Extra processor). The 5x0/6x0/7x0 convention was used for the European markets. Some Canadian product sheets also used the European convention. Some models were listed with a "-80" added to the end to indicate 80 column video. The P500 was the only model with 40-column video with colour. In any case the machines are fundamentally the same save for the power supply and NTSC/PAL video setting.
The following are various models and model numbers released. All machines are basically made up of combinations of the following:
CODE | TYPE | CASE | VIDEO | RAM | CO-PRO | DRIVES | MODEL NUMBERS | NOTES |
PL1 | Professional | Low | color | 128K | - | - | 500P, 510, 520, P500, P128, C128-40, PET-II | Very rare. |
BL1 | Business | Low | mono | 128K | - | - | 700B, 610, B500,B128, B128-80 | Most common model. |
BL2 | Business | Low | mono | 256K | - | - | 620, B500, B256, B256-80 | See spec-sheet. |
BL1X | Business | Low | mono | 128K | Yes | - | ? | Did this exist? |
BL2X | Business | Low | mono | 256K | Yes | - | 730BX, 630 | Did this exist? |
BH1 | Business | High | mono | 128K | - | - | 710B, 710, B700,CBM128-80, PET700, PET700/128 (Swedish) | Common. (See below) |
BH1SD | Business | High | mono | 128K | - | YES | 720B | See 700 brochure 2 |
BH2 | Business | High | mono | 256K | - | - | 715B, 720, B700/256, CBM256-80 | Rare. |
BH2DD | Business | High | mono | 256K | - | YES | 740B, 720D | Pre-production or Prototype? |
BH1X | Business | High | mono | 128K | Yes | - | CBMX128-80 | Did this exist? |
BH2X | Business | High | mono | 256K | Yes | - | 730BX, CBMX256-80 | Did this exist? |
BH2XDD | Business | High | mono | 256K | Yes | YES | 750BX, BX700, BX256-80 | Did this exist? See spec-sheet. |
NOTE: There are examples of the B500 with 128K and 256K Ram. Click here for the CBM-II section of Bo Zimmerman's canonical list.
Below are serial#'s gathered from pictures on the net and other sources:
MODEL | HIGHEST SER# | QUANTITY | NOTES |
P500 (US) | P5001554 | 1554 | Owned by UVB. picture |
B500 (US) | B5001739 | 1739 | Owned by Leonard Tramiel url picture. |
B128 | C010929 | 10929 | Info from Ed Shockley. sample label |
CBM128 | B1001344 | 1344 | Owned by Bo Zimmerman. picture |
CBM256 | B1007210 | 7210 | Owned by Dave McMurtrie. picture |
P500 (euro) | WG02014 | 2014 | Ebay.uk auction-164831048652 picture |
B500 (euro) | WG00395 | 395 | sample label |
B700 (euro) | WG00322 | 322 | Owned by Esa Lyttinen. picture |
610 | WG05109 | 5109 | Ebay.de auction-321543419621 picture [oct'2014] |
620 | WG00056 | 56 | Owned by Christian Forstreuter picture |
710 | WG05127 | 5144 | Owned by nun-retro.picture |
720 | WG02599 | 2599 | Ebay.de auction-280763329368 picture |
When added, they total over 37,000 and it's highly unlikely that the highest serial number listed represents the last machine made. I will assume serial#'s start at 1. Most likely Commodore would have made at minimum a few thousand of each model in order to justify a production run. If you have a higher serial# than listed here please email me a picture of the label so that I can update this info. Thank-you!
The machines are very similar, but there are some differences. There are differences between the video of the B-series LP and HP machines. The P-series is based on the VIC-II chip so the machine must run at 1MHz.
FEATURE | B-HIGH PROFILE | B-LOW PROFILE | P-LOW PROFILE |
CPU | 6509 | 6509 | 6509 |
CPU Speed | 2 MHz | 2 MHz | 1 MHz |
Internal Memory | 128K or 256K | 128K or 256K | 128K |
Max Internal Memory | 256K | 256K | 256K |
Max Expansion Memory | 704K | 704K | 768K |
Text Mode | 80x25 lines text | 80x25 lines text | 40x25 lines text |
Character Size (pixels) | 8x14 | 8x8 | 8x8 |
Colours | Monochrome | Monochrome | 16 Colours |
Graphics Mode | None | None | Max 320x200 |
Integral Display | Yes,Green Screen | No | No |
Internal Disk Drives | Optional (IEEE) | No | No |
Sound | 3 voices | 3 voices | 3 voices |
Keyboard | Separate | Integrated | Integrated |
Joystick Ports | None | None | 2 |
IEEE,RS-232,User Ports | Yes | Yes | Yes |
External Video Port | No | Yes-5pin DIN | Yes-8pin DIN |
Cassette Port* | Yes | Yes | Yes |
Cassette Support in ROM | No | No | Yes |
Optional 8088 Board | Yes | Not without mod | No |
Built-in ROM Language | BASIC 4+ | BASIC 4+ | BASIC 4+ |
MLM in ROM | Yes | Yes | No |
Power LED | None | Red LED | Green LED |
* The cassette port is electrically functional but software support was removed in later B-series ROMs. MLM=Machine Language Monitor.
The B and P series share many of the same hardware components, with the main differences being the video chip, clock frequency, and PLA chips.
COMPONENT | FUNCTION | SERIES | DESCRIPTION |
6509 | CPU | BOTH | Main CPU. A variation of the 6502, with 4 extra address lines and opcodes modified to support the additional memory. SEE NOTE BELOW! |
6581 (SID) | SOUND | BOTH | The famous SID chip to generate complex 3 voice sound. |
6545/68845 (CRTC) | VIDEO | B | CRT Controller chip like in the PET, used to generate 80x25 MONO video. |
6567/6569 (VIC-II) | VIDEO | P | Video chip like in the C64, used to generate 40x25 COLOUR video with sprites. NTSC or PAL version. |
6525 (TPI) | I/O | BOTH | Tri-Port Interface. SEE NOTE BELOW! |
6526 (CIA) | I/O | BOTH | Complex Interface Adapter. |
6551 (ACIA) | I/O | BOTH | Asynchronous Communication Interface Adapter. Used for RS-232 port. |
4864 | RAM | BOTH | Dynamic RAM for main memory. 64K x 1 bit. |
2016 | RAM | B | Static RAM for CRTC video, 2Kx8. |
2114 | RAM | P | Static RAM for VIC-II color nibbles. |
2332 | ROM | BOTH | ROM for Character Generator (supports 2364) |
2364 | ROM | BOTH | ROM for KERNAL/BASIC |
82S100 | PLA | BOTH | Programmable Logic Array |
Many of the above components are common and still available from vintage component sellers, or can be obtained from other vintage computers. However, some are not. The 6509 CPU was designed for the CBM-II machines and is not used on any other Commodore machine. It is very hard to find a replacement if defective. The 6525 TPI chip also saw limited use in Commodore products and is hard to replace.
There was some discussion on 6502.ORG for a possible 6509 replacement using a 6502 and some additional support components. Due to the extremely limited supply of 6509 chips this would be a worthy effort. Also, there is the VISUAL 6502 project to reverse engineer, build transistor-level similations, and then FPGA re-creations of various vintage chips. Perhaps one day the 6509 and 6525 can be re-created.
NEWS FLASH! March 3, 2018... Jim Brain has developed a 6509 replacement module that uses a regular 6502 and a CPLD chip! Initial testing shows it works well but some additional programming tweaks are needed to make it more compatible. Additional info can be found here: Fake6509.
These are some of the hardware add-ons that were available back in the day. Most were not produced by Commodore:
DEVICE | COMPANY | NOTES |
8088 Board | Commodore | Board used an AMD 8088 CPU. Works in HP models, but not all LP models due to a hardware issue. See CBUG Escape Article for details. See dedicated 8088 board page. |
8088 Board | CBUG Version | With an NEC V-20 CPU. Same issue as above. |
High-Speed Graphics | Commodore Germany | External Box with video pass-thru |
Proxa 7000 | Ultra Electronic | Board to emulate CBM 8000 series. More Info. |
8250MINI | Ultra Electronic | Board for SFD-1001 to add second drive (in 720D machines) |
IEC interface | Ullrich Von Bassewitz | New Kernal with IEC interface for cassette port (schematic) |
B128 Serial Bus | Anderson Comm Eng | IEC adapter for User Port. Here is the outside and inside of the unit. Here is the hand drawn schematic. Thanks again to Ernie Chorny. |
24K Cartridge | Anderson Comm Eng | Cartridge adds 24K of RAM to Bank 15.
Thanks to Ernie Chorny for lending me the cartridge to test. Here is a hand-drawn schematic and a view of the board. |
24K Cartridge | JCL Software | Cartridge adds 24K of RAM to Bank 15. Thanks to Edward Shockley for the picture. |
B-1024 Board | Anderson Comm Eng | 1MB internal memory expansion. Thanks to Kevin Ottum for the picture. |
Alternate OS Board | Anderson Comm Eng | Board to allow addition of Alternate ROMS or RAMs in BANK15. picture Original Ad |
1MB modification | HSG | 1MB modification of motherboard by German magazine "Heute schon geust?!". Volume 0 Volume 1. Ullrich Von Bassewitz's translation to english with pictures. |
Hi-Res Board | HRT | Hi-Res Technology's Graphics Board. 1024x512 resolution. Ad from CBUG#7 |
Hi-Res Board | Unknown | Unknown B700-series Graphics Board. 640x350 resolution. Thanks to Vossi for the picture. |
Bank 15 RAM Board | Unknown | Unknown B-series Bank15 Memory Board. Adds 6K RAM to BANK 15. Thanks to Vossi for the picture. |
Color Box | Special Illumination Systems | Adapter to connect B-series to RGBI Monitor like the Commodore 1902. Ad from CBUG#3 |
These are some new hardware add-ons created by hobbyists:
DEVICE | DESIGNER | NOTES |
8088 Board Clone | Hobby Project | A clone/improved 8088 board. Replaces the impossible-to-get 6525 chip with a Intel 8255 compatible chip. Created by Ruud Baltissen and Michael Pleban in 2018. More info here. Picture here. Video here. |
8088 RAM Board | Michal Pleban | Updated 8088 CPU board with CBM-II RAM and ROM expansion plus SD storage for PC side. Github here. |
VGA Card | Michal Pleban | Connects to internal expansion slot and provides VGA text mode output. Works with Michal's 8088 board. Github here. |
V9958 Colour Video Board | Vossi | Colour video board for internal expansion slot. Based on Yamaha V9958 VDP chip. Released 2019. Picture. Github. |
Space Chase | Christian Krenner | Special Edition game cartridge. Available to purchase here. picture. Also available as a download at www.spacechase.de. |
256K ROM Cartridge | Steve Gray | ROM Cart using PLCC EEPROM that allows up to 8 sets of 32K ROM (24K Cart + 8K Expansion). Released 2025. picture. webpage. Github. |
RAM/ROM Cartridge | Steve Gray | 32K RAM plus 256K ROM Cart. Same ROM features as above, plus 24K RAM for cartridge space plus two 4K external areas. Released 2025. picture. webpage. Github. |
Below are some of the more notable and interesting hardware addons:
The 8088 board is an internal add-on that allows the B-Series machines to run CP/M-86 and MS-DOS. Originally announced along with the B-Series and planned to be factory installed this board was never completed by Commodore. When Commodore handed all information to CBUG, a few members worked to get them running, and finally did. The CPM and MSDOS disks are specially formatted for the Commodore IEEE drives.
The 8088 board is very rare... only 40 were brought to North America (some more may be in Europe), and of those, only a handfull are known to have survived. See dedicated 8088 board page.
NEWS FLASH!: There is now a modern 8088 board replacement with updated/improved features! See hardware section for details.
The Ultra Electronic Proxa 7000 board is an internal plug-in CBM8000 emulator. It works in the HP machines and plugs into the 6509 socket. It contains a 6502 compatible processor and some peripheral interface chips like in the PET. Some additional cables go from the Proxa board to the IEEE port, keyboard port, and user port. It requires special software to activate it.
Proxa were also Commodore resellers and had a series of 700 series machine with the Proxa 7000 board preinstalled. Known as the 8700 they were available with internal dual drives and memory expansion. See the brochure.
For more details click here.
Mentioned in sales literature, was the Z80 Processor Board. This board would give the CBM-II series the ability to run CP/M software. Other than Commodore marketing material there was no actual evidence that this board ever existed. We know Commodore promotional material could often be, shall we say, "optimistic". But in 2018 actual proof of it's existence was found! We actually now have schematics and a board layout, but sadly no firmware. We don't know if the firmware was ever completed, or at what stage the product was cancelled. It may or may not have even worked...
You can download the Z80 schematics here.
Never released, the Hires Color Grafik (HCG) is the colour version of Commodore's High-speed Graphik (HSG) board for the PET 4000/8000 series. Both boards are based on the Thomson EF936x series of Video Processor chips which can perform 1 million operations per second and have hardware line and text drawing capabilities, perfect for CAD applications. The HCG has 3 bitplanes representing Red, Green and Blue, when combined can produce 8 colours for each pixel. Depending on the chip used this could be either 512 x 256 pixel progressive, or 512 x 512 pixels interlaced. We do not know if the HCG board was completed although it is mentioned in some CBUG newsletters so it was known to be in development back then. Thanks to Bil Herd for providing the schematics, and Mike Naberezny for scanning them!
You can download the schematics here.
The CBM-II machines were most popular in Germany and Sweden and localized versions of the machines were made for these markets.
Here are PDF format schematics for the CBM-II machines.
The firmware of the B-series seems mostly stable after 4 revisions, however the P-series firmware should be considered as BETA. You can find complete source code to the B-Series Kernal and BASIC on disks 67 to 69 of the CBUG library.
Nils Eilers is working on CBM-II firmware source code that can then be assembled to recreate various CBM-II ROMS, and then potentially modify and extend its capabilities. His project can be found on GitHub here.
These are the latest ROM/PLA revisions:
MACHINE CODE |
BASIC-LO | BASIC-HI | KERNAL | CHAR | PLA |
BL1 | 901243-04 | 901242-04 | 901244-04 | 901237-01 | 906114-04 |
BL2 | 901241-04 | 901240-04 | 901244-04 | 901237-01 | 906114-04 |
BH1 | 901243-04 | 901242-04 | 901244-04 | 901232-01 | 906114-05 |
BH2 | 901241-04 | 901240-04 | 901244-04 | 901232-01 | 906114-05 |
PL1 | 901236-02 | 901235-02 | 901234-02 | 901225-01 | 906114-02 906114-03 |
(B)=Business (P)=Professional (L)=Low Profile (H)=High Profile (1)=128K RAM (2)=256K RAM
Announcing the FAST BOOT KERNALS for CBM-II! The machines normally check EVERY byte of RAM on power-up. This can take several seconds on normal B128's and B256's, and 8 seconds on the P500. For machines with expanded memory (ie: 1024K RAM) it takes much longer. I have edited the KERNAL memory test routines to only test a single byte in each page of each bank, so boot-up time is almost instant!
You will need to download the appropriate kernal file(s) and burn to EPROM. If you use a 68764 EPROM (which are hard to find) it will just plug in. If you use a normal 2764 you will need an adapter to install it.
This is a list of some of the commercial software written and sold for the CBM-II series. Most of these will require a B128 with later firmware. If you have any of these packages to donate or can provide a picture please contact me!
TITLE | COMPANY | NOTES |
General Ledger | Commodore (Info Designs) | B010 |
Accounts Receivable | Commodore (Info Designs) | B011 |
Accounts Payable | Commodore (Info Designs) | B012 |
Inventory Control | Commodore (Info Designs) | B013 |
Payroll | Commodore (Info Designs) | B014 |
Order Entry | Commodore (Info Designs) | B015 |
SuperScript II | Commodore (Precision Software) | B030 |
SuperBase | Commodore (Precision Software) | B031 |
Multiplan | Microsoft | - |
MS-DOS 1.25 | Microsoft | Requires 8088 board |
CP/M-86 | Digital Research | Requires 8088 board |
SuperScript III | Precision Software | - |
SuperBase II | Precision Software | - |
SuperOffice | Precision Software | - |
EasyScript | Precision Software | - |
Calc Result | Handic Software | Cartridge + Disk |
Word Result | Handic Software | Cartridge |
700 Assembler/Workshop | JCL Software | Cartridge or Disk |
Inventory | Northwest Music and Computer | - |
BeeLine 2.1 | LemData Software | Telecommunications Program |
Profi-Text | V lkner | Cartridge with Cassette to IEC cable |
Master 1 | Micro Application Software | BASIC Extension. Brochure- Front,Back |
8432 Emulator | Commodore | Emulates the CBM8032 computer with software. Each BANK can be a separate machine. So the B256 could have 4 machines - Hence 8432! |
Diagnostic | Commodore | CBM-II Diagnostics Cartridge. Here is the cartridge image and info for making the cabling. |
PET 700 Komp | Datatronic AB | BASIC Compiler (128K and 256K versions) |
CBUG released many disks for the CBM-II. You can find most of them on Zimmers Commodore Files. These disks are in 8050 format. See a listing with prices.
Here is some of the software I've written for the CBM-II machines. Please understand the software was written in the mid 1980's and was for my own use, but there might be some interesting things if you want to learn. I will be posting more as I sort through my disks...
In 2025 I designed some CBM-II cartridges... one with ROM only and another with RAM+ROM. I wanted something cool to put in ROM and decided to try converting the 8432 Emulator to boot from cartridge, which would make it immediately available and much faster to load than the disk version. I was successful getting the core emulation working, but I will continue to work on it to add more features. Source code and binaries are available on Github here: CBM-II-PET-Emulator/
P500 software is very rare, due to the small number of machines released and lack of support from Commodore. I will try to assemble as much P500 software as I can here. If you have anything to contribute please contact me.
The CBM-II line was a followup to Commodore's PET/CBM line. Commodore continued their tradition of making new machines incompatible to previous models. From a user's point of view the CBM-II machines functions much the same as the older PET/CBM machines, however under the hood the new machines were completely different. While older PET peripherals, such as disk drives and printers, were compatible, other PET/CBM addons were not. Also, while the BASIC language was mostly compatible, any software using PEEK/POKE or accessed the hardware directly would not work. The new KERNAL was expanded, and the new memory model required software to be re-written. Instead of a single 64K address space, the new CPU used memory BANKs, allowing up to 1 megabyte of ram. BASIC programs and variables were split into separate BANKs, and the KERNAL, BASIC, and I/O were now in the "system" bank. Very little RAM was in the system bank, forcing machine language programs to reside in a different bank, which caused access to ROM routines and even the video screen troublesome.
The CBM-II B-series and P-series are very similar architecturally. The main difference between them is the video chip. The B-series uses the 6845 CRT controller just like the older PET models. The P-series has the VIC-II chip as used in the popular C64 (NOTE: The chip was designed separately and both the B and C64 were announced at the same time). Likewise for the SID sound chip. The other main difference is that the B-series is clocked at 2 MHz while the P is only clocked at 1MHz due to the VIC chip.
The CBM-II series is all about memory. The machines use a 6509 processor which is based on the 6502 core. The 6509 adds memory BANKING. Four extra lines are added to the processor giving 16 possible memory BANKs, or 1 megabyte total addressable memory. In the CBM-II series, BANK 15 is called the system BANK. All ROM, IO, and custom chips as well as some RAM, are located here. Banks 0 to 14 can contain RAM. In the P-series BANK 0 is used entirely for video RAM and BANK 1 is for everything else. In the B-series BANK 0 is not used, only BANKS 1 to 4 depending on the model. In order to maximize memory use the B-series had two different BASIC ROMS. Machines with 128K RAM had "BASIC 128" which used BANK 1 for programs, and BANK 2 for variables. Machines with 256K RAM had "BASIC 256" with BANK 1 for programs, BANK 2 for Arrays, BANK 3 for simple variables, and BANK 4 for string variables.
While flexible, in practice this BANK memory scheme is very hard to work with. There is a small amount of memory in BANK 15 which can be used for machine language programs, however if the code is large then the software must reside in one of the other memory banks. This involves setting up very complicated "transfer" routines in order to access ROM routines and custom chips in BANK 15. Perhaps this is a factor for the relatively small software base for the CBM-II line.
The IBM PC was released around the same time and newer "16-bit" processors were being developed. To take advantage of these newer processors the CBM-II line had enhanced expansion cababilities. A second processor could be attached that could "take over" the system and use the 6509 as a slave CPU. Originally Commodore developed an 8088 board that could run a version of MS-DOS and CP/M-86 although due to the complexity of the design these would not be fully working until much later. Plans to include the 8088 board as standard had to be dropped.
The B-series uses the 6845 video chip. This gives the machine an 80 character by 25-line text display much the same as the old PET/CBM 8000 machines. In the low-profile machines when connected to a monochrome monitor the screen looks quite similar due to the 8x8 character matrix. In the High-profile machines, with integrated monitor, an enhanced character set using an 8x14 matrix provides a very crisp video display.
The P-series uses the VIC-II chip giving the P500 the same colour and graphics capabilities as the well-known C64 computer. There is a small amount of memory in BANK 15 for a text screen, however to utilize graphics or sprites you must use BANK 0. Bank 0 is used for the BASIC program and BANK 1 is used for BASIC variables. With 64K of video memory for VIC-II chip we can only guess what might have been possible if the P-series had actually been released in quantity. One interesting note: In the "700 Reference Guide" on page 140, the source code makes reference to the 6566 VIC chip. It appears initial machines were to use static ram (at least for video). The 6566 VIC is the same one used in the Max Machine. Released machines ended up using the 6567/6569 which use dynamic ram.
The SID chip provides 3-voice sound capabilities. The SID was a very advanced sound chip for the time, capable of producing some very complex waveforms. In the P-series this works very similarly to the C64 due to the 1MHz clock. In the B-series, the chip is write-only due to the 2MHz CPU, since the SID is a 1MHz chip.
Finally, rounding out the machine is a standard high-speed serial port, expanded keyboard with 10 function keys and numeric keypad, 4 separate cursor keys, and detached keyboard and integral monitor on the high-profile machines. Much of the rest of the machine is similar to the older PET/CBM machines.
Outwardly, the CBM-II machines are probably the sexiest machines Commodore (or any other company) ever produced. The case is nicely rounded and functional, made of solid plastic in a pleasing tan colour. The keyboard is well designed with a nice feel and good layout. It is the first one to depart from the dark black keys of the PET series and first to have 10 function keys, a "00" key, and the first with 4 separate cursor movement keys. In front of the keyboard the case makes a convenient wrist-rest. In the high-profile models the built-in screen has tilt and swivel for easily adjusting for the best viewing angle. The screen is recessed, reducing light and glare from the side and providing a little privacy from prying eyes in a schoolroom environment.
On the back are all ports. The back panel is metal allowing for quick design changes without worrying about changing the plastic case (unlike the C64). There is no power cable sticking out the side like with the VIC-20/C64 and there is finally a real reset button. This also marks the first Commodore to have a true industry-standard RS-232 serial port accessable on the back.
Inside for the first time again, there is a real switching power supply, not the big bulky transformer of the PET line or brick-like supply of the VIC. The motherboard is complex and packed tightly with plenty of expansion ports. In the B-series the CPU is clocked at 2MHz, making it the fastest of the Commodore 8-bit machines.
At initial announcement there were to be internal drive options for some machines. No production machines actually shipped with internal drives, even though Commodore clearly planned to offer them. The HP machines have a removable bezel and room for two drive mechanisms. Later, Commodore did release the CBM8296D machines which used the B-series case and had internal 8250LP drive units.
Before the CBM-II series, Commodore disk drives were "intelligent" units. Basically they were computers, with their own CPU's, RAM, ROM, and IO chips. The "DOS" is built-in to the drive so that the main computer's resources are not reduced. This made sense in the early days when memory was expensive and the CPU's address space was limited. These disk drives initially connected using standard IEEE-488 protocol. Later a serial version of IEEE, called IEC was developed. These connections were fairly slow. It's clear that initially the internal drives would NOT be IEEE units. This is proven by the fact that no B-series motherboard has an internal IEEE port (like the CBM8296D does).
With memory prices dropping and memory address space expanding, Commodore came up with something called the "Low Cost Disk Units". These drives were promoted as using "DMA" (direct memory access) which implied that they would be very fast, however little was known about the actual implementation itself.
If you look at the B-series documentation (schematics, memory map, spec sheets) you will see that $1000-1FFF is designated as "Disk ROM" and $D900 is "Disk Units". This would suggest that extra ROM, as well as one or more I/O chips would be added to support the drives. It is possible that these "Low Cost Disk Units" might be controlled with the internal 6509 CPU, or could have been the "TCBM" style interface as used in the 1551 drive (C264/TED series machines).
All this was a big mystery until 2018 when the long-forgotten "Low Cost Disk" schematics were found! We now know that these drives were to be VERY simple, consisting of some interfacing logic and a big gate-array. The CBM-II's 6509 CPU would have done the work, and the gate-array would be the interface to the actual mechanisms (ie: there is NO "standard" FDC chip like found in other computers). Unfortunatly we do not have ROM images so there are still some mysteries to be discovered...
You can see the "Low Cost Disk" schematics here! And the original scanned sheets as they were stitched together here!
Some 720D units with internal drives were discovered in the Commodore labs. These drive units are IEEE drives. Very early units have an SFD-1001 board with an additional 8250MINI board made by "ULTRA ELECTRONIC". The story goes that Commodore had trouble making/selling 8250LP units and so the Ultra Electronic filled the gap with the 8250MINI. Commodore eventually bought some from Ultra Electronic, to perhaps study or test. It is possible that Commodore planned to offer both internal single and dual drive machines. If you wanted one drive you would get the SFD-1001 board. If you wanted to add a drive later then the daughterboard could be installed. Other 720D machines have an 8250LP installed with modified mounting hardware, but it is not clear if these were installed by Commodore, Ultra Electronic, or someone else. In a normal 8250LP the "analog" board is mounted to the main board between the two drives. In the 720D the analog board is connected via ribbon cable, and mounted to the side of the drives, as the power supply is at the top back of the unit. In the CBM8296D the board is mounted behind the drives since the power supply is at the side and bottom.
These days, IEEE drives are harder to find and expensive. Serial IEEE (known as IEC) drives are much more common. Almost all new storage solutions, such card-based drives, use the IEC protocol. It would be desirable to be able to connect such devices such as the 1541, 1581, or uIEC to the CBM-II machines. The uIEC in particular will give huge storage capability and easy file transfer from the PC.
Luckily, several solutions exist...
In 2009 I built the cable and tested both of Ullrichs' solutions. It was my intent to adapt and expand the routines and create a new cartridge solution, unfortunately lack of time prevented me from working on it. Luckily Michau took on the task and has done a great job!
There is not a lot of documentation for the CBM-II machines. A User's Guide was included in the box. In North America the first User guide contained info about the "C128-40" (aka P500), however it was removed in the later edition. In Europe there was only one manual (that I am aware of). There was also a Programmer's Reference Guide that was sold by Protecto.
These are spec sheets I picked up while attending a World Of Commodore Show in Toronto (Malton actually). I loved picking up info sheets at all the shows I went to (and I went to every one I believe). Also included are pics from some of the Commodore catalogs, pictures of the user's guide and even the label from the original shipping box from my B128.
Thanks to Facebook and groups like The Commodore International Historical Society new documentation from ex-Commodore employees have started to appear. With permission, I will post any CBM-II related stuff I can.
These are scans of Magazines/Newsletters from CBM-II-related User Groups.
CBUG was the Chicago B User Group. I have some issues, but not a complete set. I will be scanning these as time permits. If you have any missing issues you wish to donate/scan/contribute please contact me!
Mike Naberezny has scanned his issues. They can be found at: http://6502.org/documents/publications/cbug_escape/.
A High-Profile B-series machine was seen in the movie "The Jewel of the Nile" (1985). When Kathleen Turner's character is in the palace snooping around she discovers an operations room with maps and computers in it. There are 4 high profile machines in the room, two on one desk and two in the background. I frame-by-frame advanced the scene and discovered the machine on the right is a PET-SK and the one on the left is a B, based on the keyboards. In between the two machines are two SFD-1001 disk drives. In the background are two more machines that could be either SK or B but I can't tell. View frame from movie (time index 00:32:30)
A CBM-II keyboard was used in the movie Operation: Zeitsturm (2008). Discovered at: Starring the Computer website.
The case design was rumoured to have been designed by Porsche. While true that Commodore went to Porsche initially and a case was designed, it would have been prohibitively expensive to produce. Instead, armed with the original PET design they turned to a Boston design firm. Ira Velinski was the man that ended up designing the case, which later won an international award. The case designs were one of the few computer cases Commodore ever patented.
The following are some machine pictures, both my own and submitted.
These are my machines. I purchased the B128 from Protecto when they were being liquidated. The B500 I bought later from the same guy that had the B Prototype machine below. The P500 was purchased from ebay. The B128 was my main machine for quite a while. I prefered it over my C64. In fact, the machine was used to run the family business, doing invoices and accounts recievable with custom software I wrote for it. I even upgraded the RAM and built a nice little internal speech-synthesizer module for fun. I ran the unit with my SFD-1001 drive.
I recently picked up another B128 from ebay in 2006, a 710 in 2009, and a European P500 in 2010.
P500:The P500 is a rare machine. I got mine on ebay from a guy in California. Estimates are that only 1500 or so of these machines were produced. The European P500 (pal video) is more common, however only a few North American NTSC machines are known to exist now. I now own one of each. The NTSC version is shown below.
Various pictures showing my LP machines together. Ya, I know they look the same but hey it's all about quantity ;-)
I bought this machine from a developer. It is an early B-series prototype. From Edward Shockley's web pages I understand that this might be the only surviving B prototype. It was hand-assembled in a PET case. All chips are socketed and it has a ceramic 6509 CPU, probably from the first production run. Notice there are no PLA chips, so all the addressing logic is done by discrete chips.
I received this unit in working condition but with keyboard problems, making typing impossible. It appears that there were bugs in the firmware roms (EPROMs on adapters) and I was able to burn replacement Kernal/BASIC ROMs to get the unit operational. I was disappointed to find that the machine is missing the small 4K ram in Bank 15 where I normally code my ML routines.
Additional pictures can be found on the B-Series Prototype Page.
B715:The B715 is a machine that was sent to developers before systems became available to the public. It is not known how many were made, but my unit is serial #75. The B715 is an odd machine because it is a high-profile model (HP case) but contains a low-profile motherboard with 256K ram. More info can be found on the B715 page.
CBM128-80:This machine was purchased from a kind gentleman who saw this web page. I always wanted a High-profile machine and this one is in mint condition. I always thought the machines were bigger but they are relatively small and light. It is working but seems to have no sound output.
These pictures have been submitted by other CBM-II owners. Thanks!
I'm looking for pictures of any High-profile machine with internal drives, and any hardware addons. Contact me if you can contribute some pictures!
Here are some CBM-II related YOUTUBE videos that I've found:
The CBM-II machines are very cool to me. I think that the machines had a lot of potential. It seems that they suffered from bad timing and/or lack of foresight. IBM were bringing out the "PC" and 16-bit machines were poised to take over. Plus, with the C64 being so popular Commodore couldn't devote enough resources into producing the CBM-II's. Perhaps they were just too complex to get all the bugs worked out or to program.
I am looking for the following:
I am looking to document the different models of the CBM-II line. Commodore announced many different model numbers, but which ones are "real"? And, how many were made? If you have a picture of the serial number label from your machine please email it to me. I am particularly interested in labels for CBM256-80, 710, 720, 710D, 720D, P500. Does anyone's machine have an actual label that says P128, 505, 510, 520, or C128-40?
I also collect Commodore machines. Check out Steve J. Gray's Commodore Collection page.
I would like to add to my collection of CBM-II and other Commodore equipment... I am looking for the following equipment. If you have anything on the list you would like to donate or sell please contact me!
Last updated: Jun 8/2025, 1:30pm EST
Content by Steve J. Gray. Send comments or feedback to sjgray@rogers.com.