MicroAngelo Clone Graphics Subsystem

Project Photo My first home computer was a homemade S-100 system. It was an eclectic mix of mail-order kit boards, point-to-point wired homemade boards, stuff borrowed from friends, etc., all housed in a sheet metal box that I made in the little metal shop at school. The video card was a VB1B kit that gave me sixteen 64-character lines of text, plus what we called "Radio Shack Graphics", named for the blocky 128 by 48 "graphics" available on the Tandy Model 1 computers available at the time. When I got my first real job, they had an S-100 system there with a graphics card that totally blew me away. Called the MicroAngelo, this board actually had it's own processor (a Z80) to manage the 32K of RAM it used to produce 512 by 480 monochrome images. I was in awe. Fortunately, the manual contained a full schematic. Sixty wire-wrap sockets and several copyright violations later, I was the proud owner of the coolest home S-100 graphics board on the block.

Real MicroAngelo MicroAngelo Clone
A Real MicroAngelo The first MicroAngelo clone

Some time later, the system at work was upgraded to color. This involved hooking 4 MicroAngelo cards together with a "palette" card, which combined the four monochrome signals into RGB color using fast lookup RAMs and D/A converters. Again, I was in awe. But the prospect of building 3 more "clone" boards and reproducing the palette card (with some fairly expensive parts on it) pretty much discouraged me from building a home version.

When the S-100 at work was scrapped for a new PC-based system, I managed to acquire all the MicroAngelo hardware. This was not as exciting to me as you might think; by now I had mothballed the old S-100 box and was concentrating on my Xerox 820 CP/M systems. But it still gnawed at me that here I had all that color video hardware just sitting there idle, especially that palette card with the expensive fast D/A chips. The palette board was designed to accept input from up to 8 monochrome boards. But even with my 4 "real" boards plus the homemade one, that would only give me 32 colors. It didn't seem worth the trouble.

I decided that it was time to re-engineer the MicroAngelo. I replaced the 16 dynamic RAM chips with a single 32K x 8 static part, replaced the four 2K EPROMS with a single 8K part, redesigned the bus interface, and threw out a bunch of unnecessary logic. A friend of mine laid out a PC board, and a friend of his got them made. Of course, when I say "got them made", I still had to drill all the holes, solder in jumpers for vias, and build up all eight of them myself. Rather than forcing host computer to deal with all of these individual bit planes, I wire-wrapped up a video subsystem controller to manage all the gory software details. I even added a RGB to NTSC converter circuit so I could videotape the output.

Improved Clone Block Diagram
The new, improved clone  

By the time it was finished, I had a video subsystem for my CP/M computers that eclipsed the state-of-the-art EGA cards available at the time. With its 512 x 480 resolution and 256 out of 16 million color palette, it was perfect for looking at the GIF pictures that were floating around on the BBS systems. Of course, all my friends didn't bother with the 256 color pictures because they "looked funny" on their 16 color displays. But once they saw a video tape demo from my system, they could hardly stand to look at those dithered 16-color things.

I got a lot of use out of that system over the years: GIFs, Mandelbrots (see the 16x), logic analyzer displays. I still fire it up once in a while, just to see it run, or to make some video tape titles.

Top View Front View Rear View