CBM-II 8088 Co-Processor Board

Introduction

The CBM-II series of computers was designed from the beginning to support additional co-processor boards. Although several different boards were planned based on popular CPU's at the time, in the end only the 8088 board was started. Commodore planned to release a Computer called the CBMX256-80 or BX720 (depending on country) as it's top-of-the-line CBM-II machine. This machine was to have 256K of RAM, dual disk drives and dual processors (6509 and 8088). Commodore started developing the 8088 board, but never finished it before the CBM-II line was cancelled.

Alternate Operating Systems

Commodore planned to have MS-DOS and CP/M-86 operating systems available. MS-DOS and the IBM PC had just been released and were about to take over the world. CP/M was an older operating system but had a large selection of software and applications already available. Supporting both would allow the CBM-II's to potentially run a huge selection of software.

Development

Commodore continued to develop the board. They managed to get up to Revision F of the 8088 board. By this time however the Commodore 64 became wildly popular, while the future of the CBM-II line, although advanced, was in doubt. The CBM-II line even with the 8088 board was not fully IBM-PC compatible, and a lot of "PC Clones" were being released. Ultimately they decided to cancel the CBM-II line and instead release their own line of PC compatible machines. With the cancellation Commodore passed the documentation and notes to the Chicago B User Group (CBUG) who continued to work to get the 8088 boards working. Eventually they were able to get MS-DOS and CP/M-86 working. CBUG was also working to get the board working with the lower-powered NEC V20 processor.

Unfortunately the 8088 board was never completely finished. There is a bug in the PCB that prevents a numeric co-processor from working. The board is not optimized (requiring ribbon cables rather than connecting directly to the motherboard) and fits only in the HP (high profile) systems. Getting them working in the LP (low progile) systems is possible but difficult. The LP machines also have a design flaw preventing the 8088 boards from working unless a resistor is installed to terminate the data lines properly. Also, the LP machines may not have a power supply strong enough to reliably run the 8088 board with an 8088 cpu, although the NEC V-20 version might have been possible.

Personal

Since I learned of these boards I wanted to get my hands on one. Back in 2007 I finally managed to get one from ebay. The seller was the son of the owner of NW Music. In any case, I initially tried to get it working and failed. I had offers of help from several people but never could find the time to look at it. I did verify that my HP machine had the correct PLA and the board was easy enough to install, so I assumed there was some hardware issue on the board. After a while the 8088 board was stored and forgotten. Several years later (2012 actually) I managed to purchase another one, but my life was pretty busy and it too got stored away for some future time that I could try it. Well, enter november 2014, I managed to buy a CBM-II B715 machine and it got me interested again in the 8088 boards. I tested both boards in a 710 machine I was using as my BX720D Project. Unfortunately both boards did not work. My friend Mike N suggested I try one in another HP machine so I dug out my CBM128-80, which was the machine I initially tested my first 8088 board on. Low and behold it worked! So did the second one!

I do not know why the 8088 boards work on one machine and not the other. Both seem to have the same power supply and PLA. Both boards seem similar component-wise, although the machine they work in has real masked ROMs while the other has EPROMS on adapters. I guess some additional diagnostics are needed.

Pictures