[Coco] There was a CoCo 3 update to Zaxxon! Before Z-89

[Allen Huffman]

> On Mar 20, 2024, at 2:45 PM, Andrew via Coco <coco at maltedmedia.com> wrote:
> 
> …
> Anyhow - what's strange about this particular revision is that it still looks like it is a 4-color screen? Is it our general "PMODE 4" screen with a different palette (definitely different colors than that mode)? Or is it some 4-color CoCo 3 mode (maybe one of the non-BASIC modes)? Or is it an actual 16-color mode, but only using 4 colors?

It dawned on me that this could have been a way to make 1983 Zaxxon display in color on the CM-8 monitor, similar to games using the PMODE4 screen would patch to use PMODE3 and weird colors. The CoCo 3 would let you remap the colors used in PMODE4, so it could have matched the black/white/red/blue used in the game.

But I think Roger saw “256x192 16 colors” on the disk label, and there is a sprite editor Bjork used that shows the spaceship (four versions — large to small representing different flying heights) clearly showing a palette of 16, though only four colors being used by the ship.

If it truly is using a 16-color screen (4 bits per pixel) then all the graphics would have had to be converted, PLUS it would then be moving twice as much data. The speed it runs makes me think maybe it’s just using a PMODE 4 screen (?) with pallet adjusted.

> All speculation from me; Roger would know best, of course - but still, it doesn't look like 16-colors to me (just red, blue, white, and black)? That doesn't mean it wasn't in a mode to allow additional colors, but without seeing the source code (or a disassembly - but those "ASM" files seem like they are some kind of source code for something, since they are in ASCII format on the disk), all we (except for Roger, maybe) can go on is what the disks "say" (label?)...

Yeah, Roger it trying not to be distracted by the finds during the archiving process, but I see why he had to check this one out.

It’s clear that, from the addition of the logo on the title screen (instead of just “Sega’s Zaxxon” in text) and the enhanced speed (maybe just double speed POKE?) there was more work done than just flipping palettes and changing the PMODE.

> Even so - just the fact that we have something is quite enough, though it would be interesting to know where this code fits in the "timeline"? Did it happen prior to say, 1992? Or after?

What is significant about 1992?

Perhaps when the CC3 came out, he got one and the CM-8, and just wanted to fix it to display in colors. He was no longer with Datasoft by the time the 3 came out (article in Rainbow mentions he left Datasoft to start SRB Software in 1983 - the year Zaxxon came out).

> Or was it before Steve "dropped out" of the community (details of which I'm still not very clear about, but I know it had something to do with speculation on what a CoCo 4 should or could be, his efforts to "create" such a thing, whatever that meant, and his giving up on all of it because of...and that's where it gets a bit fuzzy to me - but maybe I'm completely daft on the whole thing).

What years was he out? I know he was at a PennFest in 97 (I think), and then was back at some ChicagoFests, and was on CoCoTalk quite often. I’d expect, since these were on floppies (and he has mentioned using a hard drive system for development) this might predate all that. But… speculation!

> Or did he dabble with this after that point?
> 
> Or maybe - as mentioned - it was just a precursor to Z-89 (a game I don't think I've played...yet - but I do recall its release and review in the Rainbow)...

I just went back and looked Z-89 over, wondering if this was an evolution, but Z-89 is clearly different… More textures, graphics on the bottom of the playfield, etc. He probably used some of his routines, but Z-89 looks very upgraded.

But maybe that’s part of it… He started by patching Zaxxon for CM-8, then thought “this machine can do more” and then worked on Z-89. With the release of Zaxxon in 83, and Z-89 in (I assume) 89, I’d suspect this was somewhere in between.

Now, Bjork kept date details in his source listing — so if those turn up, we’ll probably know then.

> This is all a bit exciting to me, though I wish it had been something he had done of his own accord, and not by others after his passing...because we can only speculate on what he really would have wanted.

Yeah, I bet there was no way he could have released this without some kind of Datasoft/Sega permission since it was the same code they “owned” just with updates. Else maybe we could have seen it as “Z-87” or something.

> ...and given what I do recall about that prior flame-out with the community - I have this feeling - which is again, just unfounded speculation - that he kinda "disowned us", for whatever reason - but never updated his will - maybe always saying he'd "get 'round to it" - something I suspect we all do. I know I do, as I've already mentioned earlier about my own darn project(s)!

There is alot to the story of the passing we don’t understand yet. He was married, and has a daughter from a previous marriage. Me and my s/o spent a day with both of them, including supper at a KISS-themed restaurant they took us to, back in 2019. The whole world got weird in 2020 so I didn’t think much about not getting the yearly Christmas card from them, but I really should have reached out and kept in touch.

> Regardless, I'm excited to see what comes next (much as I'm waiting for more info to come out about both the "CoCo 3 Prototype Board", and the "GIME decapping" research, though I suspect I'll be holding my breath on those until the day I die myself).

Someone with ASIC design experience has been brought in. Roger sent them my way about getting copies of the decap scans about a month ago, so maybe we’ll know something later this year. I don’t have the 1987 GIME scans yet as the files are so huge Roger started with getting me the 1986 ones first. There is a Bittorrent of them available now for anyone interested.

The more I learn, the more I am impressed with his work. Marty’s Nightmare was done in about 30 days, for example, it seems. Back then, one person could really do a full game — but today, software is done by big companies and huge teams (mostly).

		— A