CompuColor II replica; capturing contents of PROMs

[Andrew Lynch]

Hi, I am posting on this topic on http://www.retrobrewcomputers.org and thought this post would also be relevant here

https://www.retrobrewcomputers.org/forum/index.php?t=msg&th=826&start=0&




Hi

I've been investigating whether a replica of the CompuColor II is feasible. I am primarily focusing on the logic board although there are other components that will have to be addressed too such as the extremely weird floppy drive and the unique keyboard.

Most parts on the logic board (discrete components, 74LSxxx, MSI ICs) seem to be available or pin-for-pin equivalents. One part that appears to be unobtainium is the 17.9712 MHz crystal oscillator (HC-18?) which is the sole system clock used for both the CPU, UART, and CRTC. I believe it can be substituted with a programmable TTL oscillator fed into the 8224 clock generator unless someone can find an actual HC-18/HC-49 style 17.9712 MHz crystal.

The next major hurdle will require someone with an actual hardware CompuColor II willing to remove the multiple PROMs and dump them. Whether you support my attempts or not, if there is ever to be a CompuColor II replica (by anyone), the contents of these PROMs must be captured and recorded somewhere. However, technical the hurdle for this task is extreme because it requires someone with access to actual CompuColor II hardware (preferably working but not required) *and* the willingness to remove components from the logic board *and* the ability to dump their contents so they can be replicated.

My plan is to replace the bipolar PROMs with GALs which will require some minor design changes but will be much more supportable since bipolar PROMs require specialized programmers like DataIO 29B, etc. Common, low-cost programmers like TL866II generally don't support bipolar PROMs since they require multiple unusual voltages and pre-date even PALs (~1978ish)

In addition, there are a couple of 2708 EPROMs for the character generator. They can be replaced with 2716 type EPROMs (or 28C16 EEPROMs) relatively easily (I think). Also need the contents of pair of 2708s (or 9208 mask ROMs) to replicate the character generator.

Here is a table of the PROMs that must be dumped for a replica:

Part ID, Type, Description

UF3 82S129 System Decoder
UE2 82S129 Address Decoder & Timer
UA1 82S123 I/O PROM
UB3 82S123 ROM/PROM Decoder

UG8 82S123 CRT Timing
UF8 82S129 CPU and Horizontal Decoder
UG9 82S129 Scan Decoder
UG5 82S131 Color PROM

UF6 2708 Character Generator
UF7 2708 Character Generator

ROM BASIC 0000-1FFF
FCS & CRT 2000-3FFF




Merovingian: Choice is an illusion created between those with power and those without.

[Mark T]

You probably already found this site, but thought it might be worth a look for anyone else following your project.

http://www.compucolor.org/docs.html

[T Gerbic]

There are quite a few of us with UniSites and other Data I/O programmers that could burn a bunch of PALs if that was required. It is not the end of the world. Could group buy a hundred or more PALs and spread them around.