Great TMS Video Chip Demo

Discuss computers from the Laser series, Time & Fun handhelds, etc.
User avatar
carlsson
Posts: 507
Joined: Fri Jun 13, 2008 7:39 am
Location: Västerås, Sweden

Re: Great TMS Video Chip Demo

Post by carlsson » Fri Mar 23, 2018 2:36 pm

Nice documents, and cool game.

When Karl Guttag writes they've had fully functional VDP's for 2.5 years by April 1981, does he mean the original 9918 without bitmap mode etc? As for the picture quality, would that be due to what happens once you feed its signals to a RF modulator, something TI avoided by bundling the TI-99/4 with a Zenith monitor that if I understand correctly takes the component-ish signals directly without modulation?

It appears to me that all those 3rd party computers and consoles use the A variation whether it is the 9918A or 9928A. In Europe obviously all use the 9929A. Even if they had tried to sell the chips, it doesn't state at which price. If TI charged premium for a chip that wasn't satisfying in quality, no wonder it took some time to take off. As I noted, the TMS9919 PSG sound chip a.k.a. SN76489 tends to very often have been used with the 9918/28/29 VDP, which is why I believe TI bundled those at some point in time and perhaps cut their list prices to make the chips more competitive.

There is this TMS9918A advertisement dated 1980, but it only speaks about the VDP on its own. It isn't clear if this is the advertisement Karl mentions in his notes from 1982-83 as have drawn 6000 inquiries but no action taken from Texas Instruments to those inquiries.
http://spatula-city.org/~im14u2c/vdp-99 ... sement.jpg

There is a document from the Texas Instruments European consumer division dated May 1982 about the problems with the 9929A. Were those issues fixed? I'm not entirely sure when the first PAL machines of this group arrived, e.g. the TI-99/4A but also all the others like Colecovision, Sord M5, SVI-318/328 and of course not to forget Creativision which obviously is dated 1982 on many markets, mainly international and Australian. I think some of the other systems didn't appear until early 1983, which possibly might be due waiting for a fixed 9929A.

Copyright dates on software obviously can be deceiving as either one on purpose puts in an earlier copyright date just in case a conflict would happen, or the copyright date refers to an original version on a different system, not the actual one. We have a bunch of Creativision games copyright 1981 which gives us at most 8 months between Karl's memo in April 1981 and that one or many 3rd party manufacturers - VTech and others - would have begun to develop games for a system using the VDP and games finished in the same year.

In any case, there are lots of different sources and loose threads, but so is real life, it is rarely linear and only one precise truth.
ThomHa
Posts: 4
Joined: Thu Mar 08, 2018 5:22 pm

Re: Great TMS Video Chip Demo

Post by ThomHa » Mon Mar 26, 2018 2:03 pm

Just on the three threads I feel even slightly qualified to pick up:
carlsson wrote:When Karl Guttag writes they've had fully functional VDP's for 2.5 years by April 1981, does he mean the original 9918 without bitmap mode etc?
I think so, because he has also written that work on the original started in 1977 and was completed in less than a year and elsewhere describes his memo on 'super pattern graphics mode' as being "an early 1980s memo".
carlsson wrote:As for the picture quality, would that be due to what happens once you feed its signals to a RF modulator, something TI avoided by bundling the TI-99/4 with a Zenith monitor that if I understand correctly takes the component-ish signals directly without modulation?
I think it's that plus the fact that he's writing in an NTSC country. The 9918 and 9918A produce composite video only and are in-phase with the colour subcarrier. So rather than chroma crawl that the brain can sort of see beyond, you get fixed stationary rainbows at every vertical edge. I think that's what's referred to elsewhere re: experimenting with being 'rainbowless'. Not something we'd see in PAL countries as the colour subcarrier is a different frequency (and the TMS line lengths are the same).
carlsson wrote:In any case, there are lots of different sources and loose threads, but so is real life, it is rarely linear and only one precise truth.
And sadly only one person's document stash to try to follow. I daren't estimate the proportion of things that happen at my company that affect my day-to-day responsibilities but occur entirely without my knowledge, it'd be too depressing.
Last edited by ThomHa on Tue Mar 27, 2018 6:39 pm, edited 1 time in total.
User avatar
carlsson
Posts: 507
Joined: Fri Jun 13, 2008 7:39 am
Location: Västerås, Sweden

Re: Great TMS Video Chip Demo

Post by carlsson » Tue Mar 27, 2018 10:45 am

That the TI VDP was worked on in 1977-78 is interesting as the GI STIC is of similar lineage and it sometimes is claimed one was inspired by the other. Of course we also have Atari, to a lesser extent Commodore and a bunch more hardware developers around the same time. In reality all of them probably came to similar conclusions at the same time, without strictly being inspired by eachother's work.

VDP: Higher resolution, larger number of total sprites (eventually bitmapped mode, though not in the first edition)
STIC: Hardware smooth scrolling, larger number of sprites on a single row

Both have the same number of colours (though different palettes) and both have the somewhat obscure coloured squares mode.
Post Reply