23 replies to this topic

[Tyrant]

Can anyone tell me (preferably with verifiable sources to back it up) when exactly the Jag started being developed? I know it was mostly (entirely?) developed by Flare, but I'd like to know when they started it, when Atari got interested/involved, and any interesting bits of history that happened at that time.

Google is not being very helpful so far, so I turn to you guys.

Thanks.

[busterm]

1991 is when it was first reported that Atari was working on the Panther the following year they announced they had abandoned the Panther for the Jaguar thats what was in the Game Magz back in the day not sure about any actual press releases

[kskunk]

Jaguar development began in 1990 with the heart of the system, the graphics chip. The design and technical specs of Tom, including all of the Jaguar's graphics capabilities, were set in stone by mid-1991. After that, only small changes were made on the way to mass production in mid-1993.

Once Tom was working, Jerry, the sound and "DSP" chip, was derived from Tom's design. This happened in late 1992. Because Jerry's design is 95% copied from Tom's, you could say that Jerry was designed in 1990-1991 as well.

Tom had two silicon revisions. The first "taped out" in late 1991. That means Jaguar dev kits appeared near the beginning of 1992 with real Tom chips.

At the end of 1992, Tom's second revision and Jerry's first revision taped out. These were the final chips used in Jaguar consoles.

Here is one source where John Mathieson (developer of the Jaguar) explains the process:
http://www.landley.n...s/mathieson.htm

Quote

The concept for Jaguar started in about 1990... We sold the idea of Jaguar to Atari as a concept, and they funded its development. We were looking for a manufacturer with some muscle to be involved in the project from the beginning.

Thus, Atari was involved at day 1 of the design. The business side and funding was ironed out in mid 1990.

From other interviews, we know that Atari was working on Panther as early as '89, and the Panther chip had just been finished when they funded Flare, in 1990, to develop the Jaguar:
http://www.konixmult...&content=martin

Quote

It wasn't called Panther when I joined... I wasn't keen on it, but I designed it and the chip was built. While I was over in California in '89, I actually convinced the bosses at Atari that 3D was the way to go, with the experience we'd gained on Flare one. Atari looked at the Panther and looked at what we were promising, and said can the Panther project.

Another place to look is the actual source code for Tom and Jerry, which has dates indicating when different systems were actually designed:

Blitter - November 22, 1990, basic design complete within 2 months
GPU - January 28, 1991, basic design complete within 2 months
Last major feature added May 16, 1991: Blitter collision detection

Small bug fixes added throughout 1991 and 1992.
Last bug fix added in October 1992: Fix a pipeline hazard in GPU.

Jerry - finished November 12, 1992

Hopefully there's something useful in all these dates!

- KS

Edited by kskunk, Sun Mar 27, 2011 1:35 PM.

[LinkoVitch]

Weirdly I was reading the back stuff recently. Of course no idea how accurate it is, but check the Wikipedia pages for the Atari Jaguar, seemed to have a lot of detail in it. hope its what you are after.

[Tyrant]

KSkunk that's some excellent info, just what I was after. I had seen one of those interviews but not the other one so thank you.

kskunk, on Sun Mar 27, 2011 1:25 PM, said:

Another place to look is the actual source code for Tom and Jerry, which has dates indicating when different systems were actually designed:

That's fascinating... where did you find that? Is it publicly available?

[LinkoVitch]

Lets see if this works...

Enjoy if its of any use

Attached Files

•  netlist.zip   263.82K   62 downloads

[kskunk]

Tyrant, on Sun Mar 27, 2011 4:10 PM, said:

That's fascinating... where did you find that? Is it publicly available?

The Jaguar netlists used to be on the most excellent Atari Museum website. I couldn't find them there anymore, although they still have the Jaguar II netlists up.

Thanks LinkoVitch for putting up a fresh copy!

- KS

Edited by kskunk, Sun Mar 27, 2011 4:27 PM.

[carmel_andrews]

since the jaguar was designed by the same people that also designed the KMS (konix multi-system) I am surprised that they didn't bring over some of the design aspects from the KMS into the jaguar( I.E by making the controller the heart of the gaming experience/system), after all if you think about it, the KMS was the 90's version of the Wii/move/Kinect type affair but with a console built around it) and doing that with something like the jaguar (which in it's time had some interesting hardware) i think more interesting and well thought out games could have been developed/made for the jaguar (at least it would have been a better designed controller then what jaguar ended up with)

[orpheuswaking]

carmel_andrews, on Mon Apr 4, 2011 8:27 AM, said:

since the jaguar was designed by the same people that also designed the KMS (konix multi-system) I am surprised that they didn't bring over some of the design aspects from the KMS into the jaguar( I.E by making the controller the heart of the gaming experience/system), after all if you think about it, the KMS was the 90's version of the Wii/move/Kinect type affair but with a console built around it) and doing that with something like the jaguar (which in it's time had some interesting hardware) i think more interesting and well thought out games could have been developed/made for the jaguar (at least it would have been a better designed controller then what jaguar ended up with)

I suggest you read the entirety of this website http://www.konixmult...dex.php?id=home and get your facts straight before you ask stupid questions.

[Tyrant]

orpheuswaking, on Mon Apr 4, 2011 11:06 AM, said:

I suggest you read the entirety of this website http://www.konixmult...dex.php?id=home and get your facts straight before you ask stupid questions.

That's pretty harsh man, and the question (which wasn't really a question at all, more of an observation) wasn't stupid at all, just based on a misunderstanding that's far too easy to make unless you're familiar with the above mentioned sources (specifically that the guys who designed the Jaguar's chipset also designed the KMS chipset, but didn't design the associated hardware / enclosures of either system).

[mqark]

Wow - I agree, pretty harsh.

It's my website we're talking about here, and while I endeavour to discover as much factual information as I can, there will be mistakes in there - for sure, there's a whole heap of typos and some shamefully bad use of English. But on the whole, it's as factually correct as I can make it. Certainly the interviews are transcribed completely verbatim, but these guys designed these things a LONG time ago and their memories and reasons for doing things aren't as clear as they were back in the day.
So thanks for the link to the my website - but please don't treat it like the ultimate reference, that's what I'm striving for but it'll take a long time to get there.

And while I'm here, I have an opportunity to talk with John Mathieson specifically regarding the Flare One and what lead to that, but i could throw in some Jaguar questions as well as long as the interview doesn't stray too far away from the KMS. Does anyone have any burning questions of a technical nature they need answered?

[Tyrant]

I won't treat it as an ultimate reference hehe, but as one among many. Thanks for posting it, it is useful to me.

As for questions... I guess the only real question I would have is "Why the f*** wasn't the Jaguar chipset given one more revision cycle to fix some of the biggest bugs in it?" Things like the RISC Jump bug, or the blitters A1 clipping failure, things that rather severely limit what the system can do and force programmers to go rather drastically out of their way to work around them.

[mqark]

I'd imagine that it came down to money, Martin said that it was very expensive to keep re-designing the Flare One / Multi-system chips and that some quirks could be lived with or worked around. However that's when it was just Ben, John and Martin with I guess some help from Konix who didn't have any real money to fund development.

Talking with Martin was fascinating as they really were convinced that their concepts were purposely designed to help programmers and to deliver raw power specificaly for games. I believe he meant that for both the Flare One and the Jaguar design.

It was clear to me when I interviewed Martin that he was involved soely with developing the architecture - I pulled out a Jaguar console from my bag for him to sign (Yes, I'm shameless!) and he literally had no clue what it was - he'd done his role long before the case had been manufactured.

[orpheuswaking]

@mqark & @Tyrant

Not going to get into any kind of argument or flame war, but Carmel has a history of asking asinine questions across the AA boards. His "Observations" are made without any basis in fact or even the summary reading of source material. I apologize to both of you if you feel I was being harsh and @mqark, your website is a great resource and for that I thank you.

now back to our regular programming.

[kskunk]

Tyrant, on Tue Apr 5, 2011 3:05 AM, said:

I guess the only real question I would have is "Why the f*** wasn't the Jaguar chipset given one more revision cycle to fix some of the biggest bugs in it?"

Time. And money. Time (to market) is money! Atari had so many time and money constraints their hand was forced:

1) Atari could not have survived another year without launching a product. They had to get investors excited in 1993, which meant announcing an immediate launch. Atari's revenues dropped off a cliff that year, their pending lawsuits were uncertain, their cash was dwindling, and their stock was tanking. Launching a console is VERY expensive. If they had waited another year to polish it, they would be too broke to launch it.

2) They had no time to find bugs. They were almost too late for Christmas 1993. The only way to meet the schedule was to go into mass production before they knew how serious the problems were. At that point, any workaround is 'good enough'. And the bugs did have 'workarounds'...

3) Everybody knew Saturn and Playstation were launching in Japan in 1994. Third party developers were getting tied up in exclusive agreements month by month, and Atari's only hope to get good original titles was to launch well before the other consoles.

People don't give the Jaguar enough credit for its success. Yeah, yeah, it didn't beat the Saturn or Playstation. But it did what it was supposed to: It rescued Atari from certain bankruptcy, saved the jobs of dozens of people for 3 years, and earned the founders and investors millions of dollars.

Not bad, considering the whole thing was designed by 3 people in 3 years for a million bucks. Meanwhile, Sony and Sega employed 100x the people and spent 100x the money on their consoles.

Bugs or no bugs, the Jaguar was amazing. In my opinion, it vastly exceeded expectations, by achieving a good portion of the performance, for a tiny fraction of the cost, years ahead of the other next gen consoles. And people who slam it for not beating the Playstation may as well be slamming the 8 year old star of the little league team for not outscoring Barry Bonds.

The Jaguar is AWESOME for what it is.

- KS

Edited by kskunk, Fri Apr 8, 2011 11:15 PM.

[Tyrant]

kskunk, on Fri Apr 8, 2011 10:57 PM, said:

Bugs or no bugs, the Jaguar was amazing. In my opinion, it vastly exceeded expectations, by achieving a good portion of the performance, for a tiny fraction of the cost, years ahead of the other next gen consoles. And people who slam it for not beating the Playstation may as well be slamming the 8 year old star of the little league team for not outscoring Barry Bonds.

Agreed on all points.

I once saw (although cannot put my hand on right now) an article which said something along the lines of "Atari thought 3DO and Sega were their competitors, when in fact it was Nintendo and Sony"

So yeah, the Jag doesn't beat the playstation, but that's because Sony kind of surprised everyone (and as you said were able to pump billions into development). Also I believe it's around that time when the model shifted from making a profit on the console to making a profit on the games and a loss on the console. Although I'm still amazed that Nintendo have made a profit on every single Wii ever sold, and yet were able to launch it at £180 and still be the best selling system by a mile, but now I'm going off topic a little.

[LinkoVitch]

i would love to see what the jag really can do tho. Look at ridge racer on the ps1, now look at gran turismo on the same system. They are worlds apart... with time and experience what can the jag do I wonder!

re the Wii.. it is hardly cutting edge tech being used in it, everything it uses has been developed before, even the console, which is pretty much a gamecube in a diff box, none of the silicon used is pushing any boundaries and i bet heavily off the shelf stuff.

[Tyrant]

LinkoVitch, on Sat Apr 9, 2011 9:51 AM, said:

re the Wii.. it is hardly cutting edge tech being used in it, everything it uses has been developed before, even the console, which is pretty much a gamecube in a diff box, none of the silicon used is pushing any boundaries and i bet heavily off the shelf stuff.

Oh I know, but I was using it as a counter-example that the razor/blade model isn't always the best choice.

[kool kitty89]

kskunk, on Sun Mar 27, 2011 1:25 PM, said:

Once Tom was working, Jerry, the sound and "DSP" chip, was derived from Tom's design. This happened in late 1992. Because Jerry's design is 95% copied from Tom's, you could say that Jerry was designed in 1990-1991 as well.

95% as in 95% of the RISC core design, right, not the entire ASIC (JERRY being simpler with just the I/O and DACs -and 8k scratchpad- vs the MMU+OPL+blitter and 4k scratchpad)?

Quote

Tom had two silicon revisions. The first "taped out" in late 1991. That means Jaguar dev kits appeared near the beginning of 1992 with real Tom chips.

At the end of 1992, Tom's second revision and Jerry's first revision taped out. These were the final chips used in Jaguar consoles.

All those were .5 micron chips?

I wonder how usable the 1st revision JERRY chips were . . . if they were tolerably usable, maybe Atari could have pushed an even earlier launch with a simple DMA sound+IO ASIC (maybe an FM synth chip too) instead of JERRY, or maybe even directly re-using the sound and controller I/O logic from the Falcon.

Quote

Here is one source where John Mathieson (developer of the Jaguar) explains the process:
http://www.landley.n...s/mathieson.htm

Quote

The concept for Jaguar started in about 1990... We sold the idea of Jaguar to Atari as a concept, and they funded its development. We were looking for a manufacturer with some muscle to be involved in the project from the beginning.

Thus, Atari was involved at day 1 of the design. The business side and funding was ironed out in mid 1990.

From other interviews, we know that Atari was working on Panther as early as '89, and the Panther chip had just been finished when they funded Flare, in 1990, to develop the Jaguar:
http://www.konixmult...&content=martin

Quote

It wasn't called Panther when I joined... I wasn't keen on it, but I designed it and the chip was built. While I was over in California in '89, I actually convinced the bosses at Atari that 3D was the way to go, with the experience we'd gained on Flare one. Atari looked at the Panther and looked at what we were promising, and said can the Panther project.

Yes, Brennan's recognition of the Panther's limitations led to his suggesting that Atari drop the Panther and push for a new, more streamlined, 3D supporting design. (which, in the end, not only added a lot of 3D grunt, but also addressed the major problems that made the Panther so unrealistic)
Albeit, given Brennan's apparent dislike of the Panther concept, I wonder why they didn't go all in with the blitter in the Jaguar for both 2D and 3D performance and do away with the object processor entirely. (so more like the Lynx or Flare 1/Slipstream)

For that matter, I wonder why Atari ever had the Panther in development when the Lynx chipset made the basis for a far more flexible/programmer-friendly/realistic/cost-effective console at the time (granted, it needed a bit of tweaking to be implemented for a proper 4th gen console). Hell, even an STe derived console could have been more practical than the Panther overall (especially if they'd added dual playfield support -basically doubling the SHIFTER- with 2 independent framebuffer scroll layers with separate palette entries as well, perhaps a 16 MHz 68k with wait states).
Then there was Flare's own Slipstream chip that Brennan could have pitched to Atari while working on Panther. (though given Atari already owned the IP to the ST/Lynx hardware, Flare would have needed to make a fairly competitive offer for that to be preferable -except the Slipstream design was definitely more attractive than the Panther design Atari had already looked over the Lynx and ST in favor of; the Slipstream was a ready-made audio+video ASIC that meshed relatively well with programming practices of the time and had some pretty nice features vs the video only Panther that had yet to be implemented in silicon and was about as unfriendly as MARIA had been with the 7800 -of course, the Slipstream ASIC was designed to interface with Z80/x86 based CPUs, so they'd have to select among those options if they were going to use the existing Flare ASICs -a 100 pin version supporting an 8-bit bus for a Z80 or 8088/188 and a 160 pin version supporting 16-bit x86 chips and addin an on-chip floppy controller and supposedly added logic for better bus sharing)

Quote

Another place to look is the actual source code for Tom and Jerry, which has dates indicating when different systems were actually designed:

Blitter - November 22, 1990, basic design complete within 2 months
GPU - January 28, 1991, basic design complete within 2 months
Last major feature added May 16, 1991: Blitter collision detection

Small bug fixes added throughout 1991 and 1992.
Last bug fix added in October 1992: Fix a pipeline hazard in GPU.

Jerry - finished November 12, 1992

Hopefully there's something useful in all these dates!

Hmm I wonder how quickly they completed the redesigned Object Processor. (if they had the OPL+BLITTER completed by early 1991, that could have potentially meant pushing a more limited precursor of the system out by late 1991 -no GPU or JERRY, so they'd have to work with other hardware available much more quickly, be it off the shelf or derived from the ST, Slipstream/Flare 1, or Lynx hardware)

Not having any 4th gen game console really hurt them, especially in the US market (and general revenue/stability). Granted, a 1989 released Lynx/Slipstream (or even STE-derived) console could have been better than waiting until 1991 (even with better hardware than any of the '89 options). Even the Panther could have been better than nothing (mediocre compared to the other options -even the STe based option in terms of practical use and ease of programming).
1991 was definitely the latest point Atari could push, both due to the competition, and due to Atari's financial position ('88/89 was really the healthiest period in that respect, '91 had slipped a bit but marked the last point where they were reasonably stable with the further decline in '92 and dire straits of 1993)
1989 marked the point when the 7800 went from the >1.3 million (US alone) annual sales of '87 and '88 down to under 700k in '89. (and then under 100k in 1990 -not comparing software sales, of course) So '89 was a good time to shift gears as such. (plus their financial position at the time)

Hell, they probably would have been a lot better off launching a home console version of the Lynx in 1989 and waited until 1993 for the handheld version. (at that point, reflective color LCD screeens might have been boarderline acceptable in contrast/quality -of course alongside "deluxe" backlit models -with higher cost, bulk, and much weaker battery life)

Edited by kool kitty89, Mon Apr 18, 2011 2:58 AM.

[kool kitty89]

kskunk, on Fri Apr 8, 2011 10:57 PM, said:

1) Atari could not have survived another year without launching a product. They had to get investors excited in 1993, which meant announcing an immediate launch. Atari's revenues dropped off a cliff that year, their pending lawsuits were uncertain, their cash was dwindling, and their stock was tanking. Launching a console is VERY expensive. If they had waited another year to polish it, they would be too broke to launch it.

Of course, that's assuming they couldn't have have managed to pull through with the Lynx and Computers alone being pushed in 1993. (maybe a Lynx III in 1993 -reflective color LCDs might have allowed a massive jump in battery lift on top of lower cost and bulk, even if it only ended up roughly 6-bit RGB quality output, that probably would have been well worth it -alongside the backlit models, of course, so as to not lose that market sector)

The computers were pretty much dead in the US, but they still had a glimmer of home in 1992 in Europe, especially with Commodore falling apart and PCs still being sluggish in penetrating the market. (the Falcon '040 had some real potential for the mid-range market on top of the low-end '030 -maybe they could have had an middle-ground model with an '030 on a full 32-bit bus and optional fastRAM -like the TT- and optional FPU -I kind of wonder why the original Falcon wasn't using a 16 MHz 68EC020 given the significantly lower cost and relatively close performance -or why they handn't offered lower cost derivatives of the TT prior to the Falcon)
Atari had probably missed their chance to make a place for themselves in the PC clone market in the US, but maybe that could still have been profitable at the time. (haven't seen much on their PC efforts)

The real issue wasn't revenue/funding alone, but CREDIT/investment capital support for Atari. The weak revenue meant weaker credit, of course (short of the Tramiels being their own creditors and loaning private funds to Atari -that alone could have boosted investor confidence though). And that's what the jaguar really helped with, not profitability, but hype driving lines of credit to facilitate deficit spending to support them in the short run. (given the finanacial reports, the jaguar was never profitable, 1994 came close, but not quite -software R&D, advertising, manufacturing, etc costs continually exceeded net revenue from Jag sales, so the best thing the jag did for Atari was hype them up to extend larger lines of credit)

Quote

3) Everybody knew Saturn and Playstation were launching in Japan in 1994. Third party developers were getting tied up in exclusive agreements month by month, and Atari's only hope to get good original titles was to launch well before the other consoles.

Good original titles are great if you can get them, but having lots of decent (or even average/mediocre) multiplatform titles (even if mostly "shovelware") could have been far more important. As it was, they didn't end up with that many "good" original titles at all, and many of the "original" titles were inspired by common mass market games that would have drawn far more interest than what the Jag got. (ie the hot multiplatform titles on every other system but many that the Jag didn't get, or PC ports that other consoles weren't actually getting, etc)

Having original or exclusive titles may seem important on the surface, but if you don't have "shovelware" to cater to the masses, you're pretty much screwed. Yet, a system with few to no compelling exclusives, but lots of good to decent versions of popular multiplatform games (maybe a few that are significantly better than on others or at least cheaper) could be reasonably successful on the mass market.
Of course, that's sort of a chicken an the egg thing: to get really strong mass market support, you need general popularity and influence to get strong 3rd party support in general; out of pocket licensing of 3rd party games can only go so far, but probably would have been a better option than many of Atari's investments in unique games. (and if they got lucky, many they'd end up with a few really good exclusives too, aside from that, the best "exclusives" would be computer ports that noone else was pushing -there was tons of Lucas Arts, Sierra, Epic Megagames/Apogee, etc, etc titles that were only on PC -in Epic's case, you also had a relatively small developer pushing low-cost, yet fairly compelling games -Jazz Jackrabbit and Blake Stone on the Jaguar could have been pretty cool, let alone Duke Nukem 3D -which went multiplatform, of course)

Of course, not having ANY 4th generation game console (let alone a competitive one) was a huge hurdle to overcome for Atari, not just for market position (for consumers, media, and developers), but for Atari's own financial situation. (with even a mediocre success with a 4th gen home console -something on the level of the 7800's sales, more like the panther might have ended up as- that could have meant a lot of heat would have been taken off the Jaguar's release and greatly reduced Atari's problems of brand recognition on the market, let alone if they actually managed to pull off competitive hardware/marketing/3rd party developer negotiations/etc against Sega and Nintendo in the US and/or Europe)
Heh, maybe even Katz would have stayed if he knew something really big was on the horizon (or come back from his vacation to re-join Atari rather than Sega), or maybe go back to Atari Corp after 1990, when he was replaced by Tom Kalinske at Sega. (of course, Katz had also favored the offer for Atari Corp to distribute the MegaDrive in the US -offered back in 1988, before the MD even launched in Japan, but, of course, Dave Rosen and Jack Tramiel couldn't agree on the terms of the partnership -mainly contention over what would happen with Europe- so that fell apart of course, Atari was in a much better position than Sega in the US at the time with the 7800 outselling the SMS by a good margin and the MD was totally untested in Japan -plus, if it hadn't been for the radical shift in management at SoA with Katz and Kalinske, the MD could have ended up another SMS-like snafu in the US regardless of the quality/quantity of 1st/2nd party Japanese software -and some of that JP software was facilitated by SoA management too, and obviously the relationships built with western developers -EA likely would have been sued over unlicensed publishing had it not been for Katz deftly managing a favorable licensing agreement with EA that led to a strong relationship with the 2 companies and made EA a key element of the Genesis's US success)

Quote

People don't give the Jaguar enough credit for its success. Yeah, yeah, it didn't beat the Saturn or Playstation. But it did what it was supposed to: It rescued Atari from certain bankruptcy, saved the jobs of dozens of people for 3 years, and earned the founders and investors millions of dollars.

Well, it managed to build up enough hype to partially mitigate some horrible management mistakes made from '89-92 by Atari, and only in the short run (if it hadn't been for the Sega lawsuit winnings, the jag would have done nothing but dragged them deeper into debt given the financial reports from '93-95 that got posted a few months ago).

In that sense, the Lynx (or even computers) could have been more profitable in '93-96 than the Jaguar, though if they lost the Sega lawsuit due to 1993/94 cashflow problems (and unwilling to loan private funds instead of 3rd party investors), it could have ended up worse overall for them. (though had they opted to make the minimum contribution of private funds, it actually could have been a better investment overall -and no interest being accrued by 3rd party creditors)

Quote

Not bad, considering the whole thing was designed by 3 people in 3 years for a million bucks. Meanwhile, Sony and Sega employed 100x the people and spent 100x the money on their consoles.

Yet Sega ended up with rather sloppy/inefficient (at least cost/performance wise) hardware and had massive internal management issues following 1994 (some in 1994, but mainly starting with the Japanese upper management oddly forcing SoA to push the launch date up to spring of 1995 rather than the planned Fall/Summer date -wrong for some many reasons, and totally exacerbating many of Sega's existing problems).
You could argue Atari's management problems were largely tied to funding (except they started slipping almost as soon as Jack -and Mike Katz for that matter- left the company, so there's obviously more to it than that), but that definitely wasn't the case with Sega prior to late 1996.

Quote

Bugs or no bugs, the Jaguar was amazing. In my opinion, it vastly exceeded expectations, by achieving a good portion of the performance, for a tiny fraction of the cost, years ahead of the other next gen consoles. And people who slam it for not beating the Playstation may as well be slamming the 8 year old star of the little league team for not outscoring Barry Bonds.

From a hardware standpoint, certainly, it was probably the best chipset (in terms of overall cost to performance in a given configuration and manufacturer) of anything on the market in 1993 or even '94 (if you restricted the PSX chipset to a single bus and similar component/manufacturing costs to the Jaguar, it might have even been weaker overall -stripping away Sony's vertical integration and volume production, of course, but the Saturn obviously is many times less cost effective -hell, if TOM+JERRY was configured closer to the Saturn, it probably would have had significantly better performance and still cheaper to manufacture than Sega's hardware). That's both a testament to the Flare engineers, and the general emphasis on such a tight, high performance, low-cost optimized design. (if Sega's management had set their far more substantial R&D resources to work on a project of a similarly aggressive, high performance/low cost design, they might have ended up with something even better -probably for a lot more overhead, but also probably cleaner/less buggy and/or in less time- but that obviously wasn't the case -in many respects the Saturn is a really weird design for the aggressive home game console market

Quote

The Jaguar is AWESOME for what it is.

A shame it wasn't pushed as such from the start. There could have been an emphasis on non polygonal 2D+pseudo 3D with minimal use of polygosn when absolutely necessary to make the best of things. (with a gradual evolution of more and more polished games following that formula)
You had plenty of options for scaled 2D, height maps (voxels or doom/wold3D type ray-casting), flat/gouraud shaded/texture mapped polygons, and beyond (interpolated height maps are awesome).
Even focusing on gouraud shaded optimized models more could have been a big help. (you also have numerous games where they probably should have dropped to lower detail/resolution/screen size to allow a more reasonable framerate -even better as an option rather than default only, be even if you forced AvP to have the same resolution as Doom, it probably would have generally been preferable, you could even have extreme cases like a fully texture mapped polygon renderer in a 160x96 window using 2x2 scaled pixels -maybe interpolated- could have been quite useful -you also had some odd cases on the 3DO like Doom where high detail was forced in spite of low framerate -and dropping the screen size was not an attractive option, dropping to 1/2 horizontal or even horizontal and vertical resolution could have made the game far more playable)

Aside from exclusive games, you already had cases like Commanche to start off with. (Cubermorph with voxel landscapes would have been really neat) So a decent starting point to push voxels as well as more options for scaled 2D. Scaled/rotated stuff, warped perspective (mode 7-like) stuff, Wolf 3D, Blake Stone, etc, etc. (a port of Wing Commander 1 and 2 would have been really nice, let alone some scaled arcade games -or clones of those short of actual licenses, or derivatives of scaling heavy lynx games -then there's also the Atari Games polygon and scaling based games, and they were in a fairly decent working relationship with TWI at the time)

A port of X-Wing would have been awesome for the time among various other PC games the Jag was reasonably to exceptionally well suited for.

[kool kitty89]

kskunk, on Fri Apr 8, 2011 10:57 PM, said:

Not bad, considering the whole thing was designed by 3 people in 3 years for a million bucks. Meanwhile, Sony and Sega employed 100x the people and spent 100x the money on their consoles.

I missed this before, but you say 3 engineers? Who was the 3rd? (I didn't think Ben Cheese had been involved with the jaguar design)

[kskunk]

kool kitty89, on Fri Apr 22, 2011 4:28 PM, said:

I missed this before, but you say 3 engineers? Who was the 3rd? (I didn't think Ben Cheese had been involved with the jaguar design)

Tim Dunn was the "third engineer" at Flare II. He helped with chip bring-up and designed the Jaguar mainboard and all the dev boards, including the Alpine. He helped write the Jaguar chipset documentation (check the credits). As for chips, he designed a small part of Jerry and later designed the Butch custom-chip for the Jaguar CD.

Without Tim, Tom was just a pretty piece of plastic and metal. You need tools and PCBs to find out if it works!

I might still be underestimating how many people worked on Jaguar in the early days. In interviews, John only gives credit to him, Martin, and Tim, but there are references to bugs found by a guy named Steve in the test code. Software developers began helping as soon as the chips were back, from both inside and outside Atari. And of course Richard Miller at Atari played a major management role.

- KS

[Tyrant]

Fascinating, thanks kskunk.

[kool kitty89]

kskunk, on Sat Apr 23, 2011 10:35 PM, said:

kool kitty89, on Fri Apr 22, 2011 4:28 PM, said:

I missed this before, but you say 3 engineers? Who was the 3rd? (I didn't think Ben Cheese had been involved with the jaguar design)

Tim Dunn was the "third engineer" at Flare II. He helped with chip bring-up and designed the Jaguar mainboard and all the dev boards, including the Alpine. He helped write the Jaguar chipset documentation (check the credits). As for chips, he designed a small part of Jerry and later designed the Butch custom-chip for the Jaguar CD.

Without Tim, Tom was just a pretty piece of plastic and metal. You need tools and PCBs to find out if it works!

Hmm, for the Slipstream design, didn't Attention to Detail handle the formal documentation and tool support?

Did Atair ever have a whole team devoted to developing an SDK? (or outsourcing any of that to a 3rd party developer -especially a game developer/publisher given the specific perspective that would benefit from, and that's exactly what Konix did . . . and Sony for that matter with Psygnosis handling their SDK for the PSX -in hindsight at least, it seems like id would have been an excellent option for contracting expanded tool development)