Jump to content
IGNORED

Its 1993, you're in charge of the Jag, what do you do?


A_Gorilla

Recommended Posts

OK, so I've been thinking more on the non-tech related problems surrounding the Jaguar, namely the financial problems forcing the early release of the Jag. (of course the lack of funding ties directly into the hardware issues, along with questionable management, marketing problems, advertizing, etc; though it was mentioned in another thread that management of Atari corp went down hill after Jack retired and ut Sam in charge and maybe that he wouldn't have given up on computers as easily -but for this discussion let's ignore this)

 

The GPU could have been the main CPU, if only it had a C compiler that allowed code to be executed from main RAM. A direct-mapped I cache (the easiest kind of cache to make) would have unlocked the JagRISC's true CPU potential, but I suspect even this was too hard given all the other constraints. A GPU with a 2KB direct mapped I-cache would completely blow away the 68020. (And the GPU _does_ blow away a 68020 but only on small programs that fit in SRAM.)

 

I've heard before that the Jaguar's release in 1993 was a desperate survival measure to raise the investment capital needed for Atari to continue as a going concern. They sure weren't making any profit on STs or Falcons by 1993, and the lawsuit windfall wasn't until September 1994.

 

Obviously, the Jaguar should have had a strong nation-wide launch, more games, and less buggy hardware. I think Atari did the half-assed NY and SF 1993 launch just to fire up investors. And the further ahead of their competitors they could get, the more likely investors would bite. No well funded company would act like this -- they'd release when the time was right. But Atari didn't have that luxury.

 

So a lot of these discussions sort of assume Atari was like other businesses, with a strong revenue base and plenty of cash on hand, to get it right. They had neither.

 

Which is fine as long as you put, 'Oh yeah, and first my rich uncle puts $100M into Atari... Then they do XYZ.'

 

The obvious solution would be having the Tramiels support Atari with personal funds until the Jaguar was ready for release, or at least help things a bit more, maybe not a perfect launch, but more coordinated than the half assed '93 one.

 

Howver, a very interesting and more ambitious option was suggested by atarian a while back:

OK.on Topic. Since stock speculation was so great at the time Atari should have spun off Jaguar as a different company. Told the public Jag was the next great thing.(Internet Gaming) and done a great big IPO (as many did) (Amazon didn't make money for years). Then used the cash ala Microsoft not only to advertise but to purchase a well known software maker or two. Maybe from 3rd party Sega or a certain company that calls itself Atari now. ;)

 

Can't be worse than it turned out.

[snip]Something along those lines or an independent company since Atari had sullied it's rep a bit. There were many scam IPO's in the 90's. Why not a good IPO like the jag. Jaguar Corp. if you will.Selling it as made by IBM wouldnt hurt either.Remember this is marketing we are talking about. Also the internet marketing aspect would have to be added. As you will remember people invested heavily on speculation that anything internet related was going to be the next great thing. In other word 2 companies. If successful Jag corp now owning a couple good developers and having success in the marketplace buys Atari for $1 and now one company again or just an asset purchase.Atari could be used to market legacy items or develop new budget gaming or whatever.

 

If they'd done that, and it worked, there should have been sufficient funding to really make a proper debut for the jaguar, with bugs fixed (and maybe other improvements like blitter buffering, and get rid of the DSP's 6-cycle delay), the added 020, and maybe even a CD drive. (especially if that would have further increased investor intrest)

 

Have a suitable launch lineup of some dozen games (at least 1/2 heavy hitters like T2K, Wolf 3D, and AvP or Doom, along with some simpler upgraded ports like Raiden, and other exclusives like an improved Cybermorh, and reasonably complete Crecent Galaxy)

 

With that kind of fundign early enough on, maybe flare could have implemented cache for the RISCs and toss out the added host CPU entirely to save cost. In any case, better funding and accelerated development should have facilitated a good RISC compiler, particularly with the bug-free chips. Kskunk mentioned C coding wasn't considered a key feature by flare when designing the Jaguar, but perhaps they chould have shifted rioretys faster or done some other changes with the additional funding available. (later in development, once it became more clear how important C was becoming, particualrly with the 3DO) Maybe the same could have been done with texture mapping support as well, or at least shifted a bit for that once it's importance was realized. (but hardware polygon rasterization might have been better to focus on)

 

If adding the cache and dropping the host wasn't practical, perhaps another option would be possible. Back to the ARM again. The ARM-60 used in the 3DO wa a very small chip 9at 35,000 transistors), much smaller than an 020, and even smaller than the Jag's 68000, it also should have substancially higher performance than an 020, but it lacks a cache, so would have bus contention problems. (but still way better than the 68k, and would still increase jerry to 32-bit width)

Perhaps atari could have licence the ARM-6 core (the ARM 60 being a bare core) and incorporated it into Jerry, maybe adding a 1-2 kB scratchpad to work in as well, or allow it to use the DSP's scratchpad. (even with 2k of SRAM and the ARM core added to jerry, it should still be smaller than TOM).

 

However, ideally you want a chip with cache, and the ARM 600 and slightly cut-down 610 are far larger, and more expensive chips becouse of this, with ~245,000 transistors more from the added cache. Of that 245,000 transistors, 196,608 comprise the SRAm, with the remaider being cache logic, so it should be possible to design a custom chip with the ARM-6 core and coresponding cache logic, but much smaller cache. With a small, 256 byte cache (like in an 020), such a chip should be 95,680 transistors. (about 1/2 that of the 020, and added to Jerry, still more than 50,000 transistors less than Tom, of course, simply adding those onto Jerry isn't simple, it would need to be incorporated while designing the ASIC in the first place, to efficiently accomadate it onto the die)

 

Going by this line of reasoning, you could add 1 kB of cache and still not be as large as Tom, with the ARM portion adding 132,544 transistors. (and as a standalone chip, still a good bit smaller than an 020, and even poduced on 1 nm should use less silicon than the the 600,000 T .5 nm Jerry)

If such a chip was really not practical to include on Jerry, it could either be provided by ARM on the standard 1 nm process, or possibly licenced and produced along with Tom and Jerry (by Toshiba iirc) on the .5 nm process, making it a very small chip indeed. (for the time) Or licence the full 610 and have it produced along with Tom+Jerry on .5 nm. (at which point it would be only 2x the size of a 1 nm ARM-60, and only a little bigger than a 68000 on 1nm)

 

When making such a custom hip, cut down fromt eh 610's cahe, maybe the cache logic could be stripped down as well to a basic direct mapped instruction cache, making the chip even smaller. (though that would probably be rather cumbersome, so only cutting out the cache RAm would seem a lot more practical, maybe 1 kB split into 2 512 byte sections)

Edited by kool kitty89
Link to comment
Share on other sites

OK, so I've been thinking more on the non-tech related problems surrounding the Jaguar, namely the financial problems forcing the early release of the Jag. (of course the lack of funding ties directly into the hardware issues, along with questionable management, marketing problems, advertizing, etc; though it was mentioned in another thread that management of Atari corp went down hill after Jack retired and ut Sam in charge and maybe that he wouldn't have given up on computers as easily -but for this discussion let's ignore this)

 

The GPU could have been the main CPU, if only it had a C compiler that allowed code to be executed from main RAM. A direct-mapped I cache (the easiest kind of cache to make) would have unlocked the JagRISC's true CPU potential, but I suspect even this was too hard given all the other constraints. A GPU with a 2KB direct mapped I-cache would completely blow away the 68020. (And the GPU _does_ blow away a 68020 but only on small programs that fit in SRAM.)

 

I've heard before that the Jaguar's release in 1993 was a desperate survival measure to raise the investment capital needed for Atari to continue as a going concern. They sure weren't making any profit on STs or Falcons by 1993, and the lawsuit windfall wasn't until September 1994.

 

Obviously, the Jaguar should have had a strong nation-wide launch, more games, and less buggy hardware. I think Atari did the half-assed NY and SF 1993 launch just to fire up investors. And the further ahead of their competitors they could get, the more likely investors would bite. No well funded company would act like this -- they'd release when the time was right. But Atari didn't have that luxury.

 

So a lot of these discussions sort of assume Atari was like other businesses, with a strong revenue base and plenty of cash on hand, to get it right. They had neither.

 

Which is fine as long as you put, 'Oh yeah, and first my rich uncle puts $100M into Atari... Then they do XYZ.'

 

The obvious solution would be having the Tramiels support Atari with personal funds until the Jaguar was ready for release, or at least help things a bit more, maybe not a perfect launch, but more coordinated than the half assed '93 one.

 

Howver, a very interesting and more ambitious option was suggested by atarian a while back:

OK.on Topic. Since stock speculation was so great at the time Atari should have spun off Jaguar as a different company. Told the public Jag was the next great thing.(Internet Gaming) and done a great big IPO (as many did) (Amazon didn't make money for years). Then used the cash ala Microsoft not only to advertise but to purchase a well known software maker or two. Maybe from 3rd party Sega or a certain company that calls itself Atari now. ;)

 

Can't be worse than it turned out.

[snip]Something along those lines or an independent company since Atari had sullied it's rep a bit. There were many scam IPO's in the 90's. Why not a good IPO like the jag. Jaguar Corp. if you will.Selling it as made by IBM wouldnt hurt either.Remember this is marketing we are talking about. Also the internet marketing aspect would have to be added. As you will remember people invested heavily on speculation that anything internet related was going to be the next great thing. In other word 2 companies. If successful Jag corp now owning a couple good developers and having success in the marketplace buys Atari for $1 and now one company again or just an asset purchase.Atari could be used to market legacy items or develop new budget gaming or whatever.

 

If they'd done that, and it worked, there should have been sufficient funding to really make a proper debut for the jaguar, with bugs fixed (and maybe other improvements like blitter buffering, and get rid of the DSP's 6-cycle delay), the added 020, and maybe even a CD drive. (especially if that would have further increased investor intrest)

 

Have a suitable launch lineup of some dozen games (at least 1/2 heavy hitters like T2K, Wolf 3D, and AvP or Doom, along with some simpler upgraded ports like Raiden, and other exclusives like an improved Cybermorh, and reasonably complete Crecent Galaxy)

 

With that kind of fundign early enough on, maybe flare could have implemented cache for the RISCs and toss out the added host CPU entirely to save cost. In any case, better funding and accelerated development should have facilitated a good RISC compiler, particularly with the bug-free chips. Kskunk mentioned C coding wasn't considered a key feature by flare when designing the Jaguar, but perhaps they chould have shifted rioretys faster or done some other changes with the additional funding available. (later in development, once it became more clear how important C was becoming, particualrly with the 3DO) Maybe the same could have been done with texture mapping support as well, or at least shifted a bit for that once it's importance was realized. (but hardware polygon rasterization might have been better to focus on)

 

If adding the cache and dropping the host wasn't practical, perhaps another option would be possible. Back to the ARM again. The ARM-60 used in the 3DO wa a very small chip 9at 35,000 transistors), much smaller than an 020, and even smaller than the Jag's 68000, it also should have substancially higher performance than an 020, but it lacks a cache, so would have bus contention problems. (but still way better than the 68k, and would still increase jerry to 32-bit width)

Perhaps atari could have licence the ARM-6 core (the ARM 60 being a bare core) and incorporated it into Jerry, maybe adding a 1-2 kB scratchpad to work in as well, or allow it to use the DSP's scratchpad. (even with 2k of SRAM and the ARM core added to jerry, it should still be smaller than TOM).

 

However, ideally you want a chip with cache, and the ARM 600 and slightly cut-down 610 are far larger, and more expensive chips becouse of this, with ~245,000 transistors more from the added cache. Of that 245,000 transistors, 196,608 comprise the SRAm, with the remaider being cache logic, so it should be possible to design a custom chip with the ARM-6 core and coresponding cache logic, but much smaller cache. With a small, 256 byte cache (like in an 020), such a chip should be 95,680 transistors. (about 1/2 that of the 020, and added to Jerry, still more than 50,000 transistors less than Tom, of course, simply adding those onto Jerry isn't simple, it would need to be incorporated while designing the ASIC in the first place, to efficiently accomadate it onto the die)

 

Going by this line of reasoning, you could add 1 kB of cache and still not be as large as Tom, with the ARM portion adding 132,544 transistors. (and as a standalone chip, still a good bit smaller than an 020, and even poduced on 1 nm should use less silicon than the the 600,000 T .5 nm Jerry)

If such a chip was really not practical to include on Jerry, it could either be provided by ARM on the standard 1 nm process, or possibly licenced and produced along with Tom and Jerry (by Toshiba iirc) on the .5 nm process, making it a very small chip indeed. (for the time) Or licence the full 610 and have it produced along with Tom+Jerry on .5 nm. (at which point it would be only 2x the size of a 1 nm ARM-60, and only a little bigger than a 68000 on 1nm)

 

When making such a custom hip, cut down fromt eh 610's cahe, maybe the cache logic could be stripped down as well to a basic direct mapped instruction cache, making the chip even smaller. (though that would probably be rather cumbersome, so only cutting out the cache RAm would seem a lot more practical, maybe 1 kB split into 2 512 byte sections)

 

The 020 would be many times more effective in cost and implementation than all of this.

Link to comment
Share on other sites

The 020 would be many times more effective in cost and implementation than all of this.

 

OK :D Just another thought I was kicking around. (I guess it's no more wild than Crazyace's suggestion for a "cheap" x86 CPU early in the thread)

 

 

Any thoughts on Atarian's suggestion for funding? (noone other than I responded to him when he originally posted it)

Link to comment
Share on other sites

It's just a shame that the 68K has to be put to sleep in order to make use of the Jags rawest potential. Maybe someone will think of some way to make practical use of the 68K and still leverage the other J-RISC even though the Motorola will probably slow it down. Even the 7800 doesn't go full speed because of the CPU.

Link to comment
Share on other sites

I seem to remember reading that the CPU really wasn't a major limiting factor on the 7800, with MARIA pretty much limited by the RAM bandwidth. (and amount of RAM available) Maybe the misconception comes from statements about the prototype MARIA (MARIA 1, vs MARIA 2 I think) which didn't use the "display list lists" and required too much CPU resourse.

 

I'll have to remember what discussion I read that in.

 

As for the 68k being configured to be more useful in the Jag, they could have added a small chunk of work RAM on a separate bus (like 64-128 kB), but from what I understand, doing that would have been more expensive to switching to the EC020, and with fewer overall benefits. (with the 68k still being a lot less powerful than an 020 and still limiting Jerry to a 16-bit data bus) The added cost coming not from the RAM cost alone, but designing a separate bus and the added board space needed for the RAM. (the simplest set-up would probably be using PSRAM, so you don't need to worry about refresh circuitry or using expensive SRAM) But from that point, you might not be far from adding a second bus the entire system could use. (although if it was just a relatively small block on a 16-bit bus, it would probably be best for just the 68k and Jerry to use -might have been nice for storing audio data -rather than using Jerry's local)

 

I really don't know how modifying things like that would compare cost-wise with using a 16.7 MHz rated 68EC020. (the cheaper model, 35 MHz rated bing the higher end EC020) kskunk mentions that the Jaguar reall was meant to be a one bus design, plus the added locals on Tom and Jerry. Alhough, the proposeded PSRAM block would more or less be a local for the 68k, but external rather than on-chip and using PSRAM rather than SRAM.

 

Using a 68010 would give a small improvement over a plain 68k and should be a lot cheaper than an 020, but still a big bottleneck to the system.

Edited by kool kitty89
Link to comment
Share on other sites

Well the 7800 CPU can halt to allow the Maria do graphics but then it has to be cut back on to control everything else if I remember.

 

Also you have to keep in mind that the "CoJag" arcades used a 68020 or some other chip which allow for fantastic video playback and sprite animation for prototype games such as

for the Atari Jaguar CD should've been arcade perfect, but it's not and it probably have something to do with the 68000 chip. Edited by philipj
Link to comment
Share on other sites

Well the 7800 CPU can halt to allow the Maria do graphics but then it has to be cut back on to control everything else if I remember.

I'm not sure how much of a hindrance that is, maria doesn't need to be continuously accessing the bus durring active display, again I need to find that discussion. I definitely remember that the bandwidth was the critical factor, this was mentioned int the context of some proposing interleaving the bus between CPU and MARIA -someone responding that doing that would only help the CPU if anything, and limit MARIA, the other context was proposed switching to DRAM, which would have been slower. (possibly crippling MARIA, even if it had its own bus -the suggestiion being having its own 16 kB of video RAM like the ColecoVision or Master System, but both of those would probably be using SRAM for that anyway, maybe not though with the A 8-bit and 5200 using DRAM, hmm)

 

Sharing the RAM on a single bus would be a problem with a faster CPU as well, using separate busses would fix this (one suggestion was to split it with one 2 kB SRAM mapped to the CPU and th other to MARIA, rather like in the NES), th eproblem there being that MARIA and the CPU would no longer be able to share the ROM bus (the NES has multiple busses on the cartridge), though maybe they could have made it possible for MARIA to work on its own bus, but also have access to the main bus for ROM without needing many more pins on the cartridge connector and additional ROM chips. (or you could have it set up like above with a much larger block of RAM for MARIA, and have the CPU hand MARIA ROM data to store there instead of accessing ROM)

 

 

Also you have to keep in mind that the "CoJag" arcades used a 68020 or some other chip which allow for fantastic video playback and sprite animation for prototype games such as
for the Atari Jaguar CD should've been arcade perfect, but it's not and it probably have something to do with the 68000 chip.

 

Area 51 was the only CoJag game to use the 25 MHz EC020, the others used a 33 MHz MIPS R3000. I'm not actually sure why Area 51 needed the added processing power, it's almost completely streaming video with 2D enemies and effects generated by hardware?

 

One reason things would need to be cut down in a port to Jag CD regardless of the 68k would be the data rate, the Jag CD has a 2x speed drive with 350 kB/s transfer rate, the arcade games streamed video off hard drives. (so much higher data rate)

Link to comment
Share on other sites

It's just a shame that the 68K has to be put to sleep in order to make use of the Jags rawest potential. Maybe someone will think of some way to make practical use of the 68K and still leverage the other J-RISC even though the Motorola will probably slow it down.

 

See BattleSphere.

 

 

Even the 7800 doesn't go full speed because of the CPU.

 

 

And it is a detriment to the system because of it.

Link to comment
Share on other sites

Area 51 was the only CoJag game to use the 25 MHz EC020, the others used a 33 MHz MIPS R3000. I'm not actually sure why Area 51 needed the added processing power, it's almost completely streaming video with 2D enemies and effects generated by hardware?

 

One reason things would need to be cut down in a port to Jag CD regardless of the 68k would be the data rate, the Jag CD has a 2x speed drive with 350 kB/s transfer rate, the arcade games streamed video off hard drives. (so much higher data rate)

 

 

the host processor would have little to do with streaming and no doubt would be handled by the DSP.

Link to comment
Share on other sites

I wonder if the CoJag had two banks of DRAM rather than 1 ( I've seen sites that state extra ram , but no details ) - If so then blitting may have been more efficient with source in the first bank, and screen in the second due to less pagebreaks.

 

Yep, it does have two, and it does get better blitter performance when the source and destination are in separate banks. This is especially useful for texture mapping, where only one pixel can be moved per blitter transaction.

 

Regarding some other design options:

 

PSRAM was an exotic component in the early 90s (still is pretty unusual today), so it doesn't have the high volumes or dozens of suppliers that lead to low costs. A lot of options that have reasonable costs in low quantity don't get cheaper in high quantity. The really cheap parts are mass market; they get 10 times cheaper when you're making millions. That's why you find FPM DRAM in the Jaguar -- that was the mass market option. Even the 68K was available from many suppliers, so they could negotiate much lower costs at high volumes.

 

(As an aside, there's some history around the 68020 and second-sources. Motorola knew that all the 68K suppliers were hurting their bottom line/ability to gouge customers. So they formed an alliance with Toshiba as the exclusive 68020 second-source. It was alleged that Toshiba got the deal by fixing prices. Motorola's traditional partners took them to court, in vain, trying to get the 020 masks. Some think Motorola's unwillingness to play fair killed off the 68K as a serious contender in the processor wars. Motorola's next big endeavor, PowerPC, was almost ridiculously open with the inclusion of many suppliers, partners, and licensing options. That may be why all three major consoles are using it today...)

 

In hindsight, the Jaguar could have focused on C performance and 3D texture mapping -- that was the winning (technology) formula used by Sony. It's evident that Sony had that focus from the design choices they made. It's also evident that the Jaguar had a different focus: Hand-tuned assembly programming for cutting-edge 2D games and 3D smooth shading.

 

If the Jaguar had a different focus, the technology would be different. Just bolting things onto the side of the Jaguar design doesn't help much from a performance perspective and is detrimental to cost. Tom and Jerry certainly have enough transistors between them to deliver Playstation-like technology, but that's not what they were aiming for.

 

To put it another way, whatever insight could have made Flare think 'we need to add all this to support C programmers', would have made them implement more streamlined solutions in the core of the Jaguar's architecture.

 

And to risk sounding too philosophical, part of enjoying the Jaguar is understanding what it is meant to be. The best games go with the flow, and use the Jaguar just how it was intended.

 

Whether technology determined the winner of that generation is debatable. Atari's entire development and marketing budget for 3 years was smaller than Sony would spend on a single hit game. Even the Tramiel family's life savings was a drop in Sony's tsunami. As it is, they just about broke even on Atari -- which is an impressive feat given that the majority of businesses go bankrupt. If they had risked it all, how long could they have kept any money on the table?

 

I think the Dreamcast is instructive. In my opinion, they did extremely well on the technology side, even beating the Playstation 2 in certain design decisions. But the business side is way more important than the technology.

 

- KS

Edited by kskunk
Link to comment
Share on other sites

And to risk sounding too philosophical, part of enjoying the Jaguar is understanding what it is meant to be. The best games go with the flow, and use the Jaguar just how it was intended.

 

That is the problem. We got too many Genny like ports that did nothing to add color and more levels.

More game like the Morphs, IS's and I-Wars with less 68k code would have worked fine, like the winning

formula of Tempest 2k and more Atari classic updates would have gone a long way in boosting the Jaguar's

market share. I know many Folks including myself were standing on the promises the Tramiels made regarding

this and they did not deliver.

Link to comment
Share on other sites

the host processor would have little to do with streaming and no doubt would be handled by the DSP.

 

So why did Area 51 even need the 020 then? The game looks like a more or less straightforeward 2D overlay with streaming background.

It's just a shame that the 68K has to be put to sleep in order to make use of the Jags rawest potential. Maybe someone will think of some way to make practical use of the 68K and still leverage the other J-RISC even though the Motorola will probably slow it down.

 

See BattleSphere.

 

How does BS manage to efficiently use the 68k without strangling the bus? Does it use it just for interupts?

 

I wonder if the CoJag had two banks of DRAM rather than 1 ( I've seen sites that state extra ram , but no details ) - If so then blitting may have been more efficient with source in the first bank, and screen in the second due to less pagebreaks.

 

Yep, it does have two, and it does get better blitter performance when the source and destination are in separate banks. This is especially useful for texture mapping, where only one pixel can be moved per blitter transaction.

 

Having such an arrangement in a home console would have made it too expensive, right? (even in a hypothetical situation with ample funding, like atarian's suggestion)

 

 

PSRAM was an exotic component in the early 90s (still is pretty unusual today), so it doesn't have the high volumes or dozens of suppliers that lead to low costs. A lot of options that have reasonable costs in low quantity don't get cheaper in high quantity. The really cheap parts are mass market; they get 10 times cheaper when you're making millions. That's why you find FPM DRAM in the Jaguar -- that was the mass market option. Even the 68K was available from many suppliers, so they could negotiate much lower costs at high volumes.

 

Hmm, PSRAM would still be significantly cheaper than SRAM though, wouldn't it? As I understand it Sega used PSRAM for the Genesis's 68k and Z80 work RAM. SNES used actual DRAM for main memory, but that was rather unusual for a console at the time as I understand it. (with some exceptions like the 5200 I beleive) Then again, maybe Sega isn't the best example, they seem to have used rather unconventional hardware choices in some other cases, like using FRAM 1994! (in Sonic 3, also Virtua Racing in Japan) Plus they apparently used SDRAM as the main memory in the 32x (as well as a 1 MB block in the Saturn), which does seem a bit odd given how limiting that 256 kB is. (thy should have been able to put a much larger block of cheaper -and much slower- FPDRAM in there instead for similar cost, they used 80 ns DRAM for the frame buffers)

 

 

(As an aside, there's some history around the 68020 and second-sources. Motorola knew that all the 68K suppliers were hurting their bottom line/ability to gouge customers. So they formed an alliance with Toshiba as the exclusive 68020 second-source. It was alleged that Toshiba got the deal by fixing prices. Motorola's traditional partners took them to court, in vain, trying to get the 020 masks. Some think Motorola's unwillingness to play fair killed off the 68K as a serious contender in the processor wars. Motorola's next big endeavor, PowerPC, was almost ridiculously open with the inclusion of many suppliers, partners, and licensing options. That may be why all three major consoles are using it today...)

 

So it could have been difficult to find a good deal on an 020? (could that at all push toward more of an advantage for the ARM-60? of course, loosing the cache advantage of the 020)

 

In hindsight, the Jaguar could have focused on C performance and 3D texture mapping -- that was the winning (technology) formula used by Sony. It's evident that Sony had that focus from the design choices they made. It's also evident that the Jaguar had a different focus: Hand-tuned assembly programming for cutting-edge 2D games and 3D smooth shading.

 

If the Jaguar had a different focus, the technology would be different. Just bolting things onto the side of the Jaguar design doesn't help much from a performance perspective and is detrimental to cost. Tom and Jerry certainly have enough transistors between them to deliver Playstation-like technology, but that's not what they were aiming for.

 

I don't think the'd have needed completely different hadware to at least be reasonably successful, I know this discussion has gone into major design changes, but the original premise was what could have been done differently in 1993. ;) And I think Sony's big advantage was the enormously deep pockets to churn out advertizing, rope in publishers (and buy Psygnosis), and eat huge losses on console sales, following that with the very freindly programming environment. (something that wasn't carried over to the PS2, which was an even bigger hit) Hell, I'd bet that the PSX could have been similarly successful if they'd put smooth shading first and T-mapping prioritized like in the Jag. (so long as it was still a bit polygon pusher) Who knows what might have happened with 3DO if Panasonic had gone that route. (slashing console prices and charging a lot more for games licensing/royaltees)

 

Anyway, even with the current design, the Jag could have been a lot more freindly with the bugs fixed (maybe a bit of tweaking elsewhwere), with a compiler good enough for quick games, and the real heavy hitters optimized with assembly. (kind of the way the Saturn was, except Sega had other problems) Better management at Atari corp would certainly be a major factor, necessary in any scenario.

 

 

 

And to risk sounding too philosophical, part of enjoying the Jaguar is understanding what it is meant to be. The best games go with the flow, and use the Jaguar just how it was intended.

 

Whether technology determined the winner of that generation is debatable. Atari's entire development and marketing budget for 3 years was smaller than Sony would spend on a single hit game. Even the Tramiel family's life savings was a drop in Sony's tsunami. As it is, they just about broke even on Atari -- which is an impressive feat given that the majority of businesses go bankrupt. If they had risked it all, how long could they have kept any money on the table?

 

I think the Dreamcast is instructive. In my opinion, they did extremely well on the technology side, even beating the Playstation 2 in certain design decisions. But the business side is way more important than the technology.

 

 

Exactly, but I still don't think Atari was completely limited by this, they might have gotten significantly greater success managing some things differently and digging into private funds. (obviously the more they did the bigger the risk, but also the bigger possibility for reward) I'd definitely have them focus on the US and Europe and not put much (if any) effort into Japan (really too hard to break into, especially as a foreign competitor), with Europe the Atari name would have more association with the more recent computers as well, while in th eUS, being more associated with the old 2600 and Arcade games. Having more money outside of personal funding would have been a big help (again, atarian's proposition is interesting here), but even then management might have been questionable, though a lot was directly related to a very tight budget, so maybe that would have made a night and day difference.

Plus, comparing them to Sony isn't exactly fair, Nintendo or Sega (especially) compared quite poorly to Sony even. (with Sega in particular being in bad shape from ~1996 onward, bleeding cache, and with severe cutbacks taking place following this, had Atari been doing fairly well up to that point, they could probably have overtaken the Saturn, especially if they did well in Europe where Sega crashed particularly hard -all the more given it had been their longest held successful market, witht he Master System and MegaDrive being exceptionally popular)

 

As for the Dreamcast, that's a discussion in it's own right (and one I've gotten into on several occasions over at Sega-16), but from the hardware end, it was more or less perfect, very cost effective and programmer friendly, with reasonably comfortable controllers (the VMUs were probably a bit of a waste) one flaw later emerging was th epiracy issue which was a killer in th eeyes of may developers. The hype from Sony hurt it a lot, Sega's major fumbles in the west from the previous generation, and finally the discontinuation of the system itsself. (who knows how things might have goen if the DC had a full lifespan, some blame Peter Moore cutting and selling out to MS, in any case Sega ended up struggling as a 3rd party Software house and getting picked up by Sammy, so the only worse possibility had they followed through with the DC would have been being completely burned out and having the company devided up and absorbed into multiple other companies)

Edited by kool kitty89
Link to comment
Share on other sites

Kool Kitty - what dod Atari have to offer console users in Europe something compelling to get them not to buy a PS, XBox, Dreamcast, or Gamecube?

 

Many folks in the US returned they Jags after finding the game selection sparse and nothing that anyone recognized.

 

Atari was doomed no matter in the console market unless there is another outside of Earth.

Link to comment
Share on other sites

You have to remember that while Atari's reputation was somewhat tarnished in the US it wasn't in Europe. The ST had been a massive success and had outsold the Amiga for many years and the Lynx had proved very popular too with major stores like Game (UK's biggest game store) still carrying the machine when the Jag came out and hell even the 2600 was still going into the early 90's with a very strong presence as a budget machine in places like Toys R Us.

 

UK retailers liked Atari and actually they made a massive mistake not getting the Jag out in numbers over here. They either should have made sure they had enough stock for the Pal release or got the machine out over here first as it could have sold some significant numbers if it had. By the time UK retailers (and I am sure European ones too) got enough units the interest in the machine had waned because of the mega circus over the Playstation and the start of the Anti-Atari Jag hate campaign in the gaming press had started.

 

I can remember in the games store I worked in at the time (I already had my Jag at this point thanks to Telegames) we got a massive delivery for our pre xmas shopping day and all we got in it was 5 Jaguars!!! We had a waiting list of well over 25 people and everytime we got any in we sold them that morning. The company actually got so pissed off that they told us to start pushing the 3D0 and 32X instead as they had such a massive inventory of those!

 

I remember reading somewhere that Atari sold all their European shipments in record time, you would think this might had alerted the Tramiels to something wouldn't you?

Link to comment
Share on other sites

Well the 7800 CPU can halt to allow the Maria do graphics but then it has to be cut back on to control everything else if I remember.

I'm not sure how much of a hindrance that is, maria doesn't need to be continuously accessing the bus durring active display, again I need to find that discussion. I definitely remember that the bandwidth was the critical factor, this was mentioned int the context of some proposing interleaving the bus between CPU and MARIA -someone responding that doing that would only help the CPU if anything, and limit MARIA, the other context was proposed switching to DRAM, which would have been slower. (possibly crippling MARIA, even if it had its own bus -the suggestiion being having its own 16 kB of video RAM like the ColecoVision or Master System, but both of those would probably be using SRAM for that anyway, maybe not though with the A 8-bit and 5200 using DRAM, hmm)

 

Sharing the RAM on a single bus would be a problem with a faster CPU as well, using separate busses would fix this (one suggestion was to split it with one 2 kB SRAM mapped to the CPU and th other to MARIA, rather like in the NES), th eproblem there being that MARIA and the CPU would no longer be able to share the ROM bus (the NES has multiple busses on the cartridge), though maybe they could have made it possible for MARIA to work on its own bus, but also have access to the main bus for ROM without needing many more pins on the cartridge connector and additional ROM chips. (or you could have it set up like above with a much larger block of RAM for MARIA, and have the CPU hand MARIA ROM data to store there instead of accessing ROM)

 

Well the point was to show certain similarities between the 7800 and the Jag concerning "CPU slowing down the system" issues. The good thing about the Jag is that it has more then one brain unlike the 7800, but 68000, when it comes to the Jag doing it's own 3D using it's own dedicated methods, seem to strain under the 68000 chip without proper programming.

Link to comment
Share on other sites

Kool Kitty - what dod Atari have to offer console users in Europe something compelling to get them not to buy a PS, XBox, Dreamcast, or Gamecube?

 

Hmm, well withthe PS, the Jag would be over a year earlier, and a good deal cheaper by the time the PS came out. (and in a hypothetical situation with fixed hardware, improved management, and a properly organized launch would also have a decent back library of games available already)

 

Many folks in the US returned they Jags after finding the game selection sparse and nothing that anyone recognized.

That shouldn't have happened if they'd changed certain things, like bug free, more freindly hardware (and better tools, but elliminating the bugs would greatly facilitate this already). So better support form developers (or rather not having a tone of developers loose interest quickly), get a lot more game sout in a timepy fassion, including first party releases, and having a decen number of titles already available at launch. (say a dozzen, with at least one killer app like T2K, preferably more, like Wolf3D, maybe more revamped classics as well, and maybe have Crecent Galaxy and Cybermorph, but enhanced and more polished, plus Raiden with enhanced graphis -color)

 

You have to remember that while Atari's reputation was somewhat tarnished in the US it wasn't in Europe. The ST had been a massive success and had outsold the Amiga for many years and the Lynx had proved very popular too with major stores like Game (UK's biggest game store) still carrying the machine when the Jag came out and hell even the 2600 was still going into the early 90's with a very strong presence as a budget machine in places like Toys R Us.

 

UK retailers liked Atari and actually they made a massive mistake not getting the Jag out in numbers over here. They either should have made sure they had enough stock for the Pal release or got the machine out over here first as it could have sold some significant numbers if it had. By the time UK retailers (and I am sure European ones too) got enough units the interest in the machine had waned because of the mega circus over the Playstation and the start of the Anti-Atari Jag hate campaign in the gaming press had started.

 

I can remember in the games store I worked in at the time (I already had my Jag at this point thanks to Telegames) we got a massive delivery for our pre xmas shopping day and all we got in it was 5 Jaguars!!! We had a waiting list of well over 25 people and everytime we got any in we sold them that morning. The company actually got so pissed off that they told us to start pushing the 3D0 and 32X instead as they had such a massive inventory of those!

 

I remember reading somewhere that Atari sold all their European shipments in record time, you would think this might had alerted the Tramiels to something wouldn't you?

 

See that's what I was thinkin of, Atari had a pretty strong presence in Europe from their computers, compared to the US where the average persom knew them most for the 2600 and old Arcade games (many fewer probably knew much about the 7800, 5200, or computers, let alone lynx), the stigma surrounding the '83 crash probably not being an issue anymore, but maybe would stil be regarind Atari as "old," something that they'd have to break through for the next-gen games. (but also use to their advantage for the remade classics)

 

I didn't know it was that extreme of a case withthe Jag releasing in Europe, they definitely should have focused more on that. (and viral marketing would probably have worked better in that case too) I'd been speaking mostly in terms of this hypothetical discussion, with a "fixed" Jaguar (and better managment with better launch), knowing Atari Corp's stronger reputation in Europe, but still I had no idea they were selling out like that and had such shortages. They really should have paid more attention to that market.

 

I still think bothering with Japan at all would be kind of a waste, there was basicly no brand recognition, them being foreign wouldn't help either, plus the PSX and Saturn were out much earlier there, so there wouldn't be much of a head start if any. Not to mention the Saturn was popular there, compared to Sega weakining significantly in the US and Europe (particularly dramitic in Europe given Sega's previous popularity).

 

 

 

One other thing I was thinking of would be getting a lot of Atari Games (Warner Interactive's) more recenet arcade titles onto the console, other than cojag games (many requiring the CD) there were others liek T-Mek, or maybe Hard Drivin'/Race Drivin' (older, and a little dated looking, but would be nice to have an arcade perfect, or maybe even enhanced port, and Race Drivin' even came out on the Saturn). Plenty of other Atari Games arcade titles to consider a well, in addition to the classics of course. (which I beleive Atari Games/Warner owned the rights to as well, though getting an argeement for licencing should be rather straightforeward as Warner was a major shareholder in Atari Corp, along with them using the CoJag hardware)

 

The Saturn version was enhanced over the arcade: http://www.youtube.com/watch?v=JGX6rSJ2N5k but perhaps minimize the added textures on the Jaguar (a fair bit is hardly noticable), maybe use gouraud shading on some things too, the scales sprites should be fine, and the added music is a nice touch.

Edited by kool kitty89
Link to comment
Share on other sites

It may also be worth mentioning that this was without ANY advertising at all too. Atari didn't even advertise the Jaguar in the UK press, nearly all the hype and interest generated for the machine was from the fiercely loyal ST magazines (which often covered Lynx stuff too). I especially remember ST Format which was the UK's best selling ST magazine (there were 4 others IIRC) devoting many pages to the cause.

Link to comment
Share on other sites

So why did Area 51 even need the 020 then? The game looks like a more or less straightforeward 2D overlay with streaming background.

 

Like I said the host has litle or nothing to do with streaming(or at least it should'nt.) However the DSP could run the streaming while the

020 ran off the bus most of the time simultaniously. Something the ARM or the 68k can't allow due to lack of cache.

 

How does BS manage to efficiently use the 68k without strangling the bus? Does it use it just for interupts?

 

Ask Scott Legrand who has a PHD or two and several patents pending. It does not use it much at all from what

Doug and Scott tell me. They also told me they could do even better if they started over knowing what they

know now.

Link to comment
Share on other sites

  • 4 weeks later...

The scenarios I use in my head are based in a historical context. For example, as mentioned earlier, Atari didn't have the resources or money that other companies had, so the changes I would have made as CEO would have been completely possible in reality. For example, this scenario is dependent on realizing that the Jaguar could not be redesigned by this point and Atari is stuck with what they had. No 68020 or new code etc. I would drop almost all current lines that were wasting money (unfortunately like the Lynx) in order to save money and be able to push the launch back, and if I had the money, I would help support the Atari with my own funds. After all it is my company!

 

 

In no particular order:

 

1. Change the controller. Do away with the keypad and add three more fire buttons like on the Pro Controller, and possibly ad shoulder buttons too.

 

2. Delay the launch until late summer of 1994 (August) and make Alien vs Predator, Iron Soldier, Tempest 2000, and a much improved Cybermorph the launch titles.

 

3. Spend the extra time gained by pushing back the launch on finishing and polishing games like Cybermorph, Trevor McFur, Checkered Flag, Kasumi Ninja etc. Also, create true development kits and create a team to help developers with tech issues.

 

4. Promote the "New Atari". Schedule press conferences, trade shows, interviews, and stress that the old Atari is gone and a new way of doing business and supporting companies and customers alike is in. This would help the negative perceptions of the past. This would help attract third party developers.

 

5. Go after established game licenses. Atari wasted a lot of money on stupid and ill-advised licenses that were never even published! Go after games like Mortal Kombat II. A special edition of MK2 for the Jaguar with an extra character or two would really sell some systems and show other companies Atari meant business.

 

6. Advertise! $3 million was nothing! Advertise on prime time tv, on Cartoon network and Nickelodeon to get the kids excited, on the 1993 Superbowl, on cereal boxes, anywhere! Spend at least $15 million.

 

All of these things could have been done in reality. I know hind sight is always 20/20, but doing just a few of these things, like the different controller and a decent launch line-up would have helped tremendously.

Link to comment
Share on other sites

The scenarios I use in my head are based in a historical context. For example, as mentioned earlier, Atari didn't have the resources or money that other companies had, so the changes I would have made as CEO would have been completely possible in reality. For example, this scenario is dependent on realizing that the Jaguar could not be redesigned by this point and Atari is stuck with what they had. No 68020 or new code etc. I would drop almost all current lines that were wasting money (unfortunately like the Lynx) in order to save money and be able to push the launch back, and if I had the money, I would help support the Atari with my own funds. After all it is my company!

 

The Tramiels did have a fair amout of personal funds availalbe, butas has been mentioned before, weren't betting the farm on the Jaguar (or Atari's future in general). Really though, they could have envested a moderate amount of funds to allow the crappy 1993 pre-release of the Jaguar (done to drum up intreast from investors), and in that time: work on the hardware a bit more, cleaning up some bugs and such, make better development tools, build more hype toward the release (also dependent on advertizing funds though), and facilitate a considerably better launch lineup. (for Summer 1994) Also, don't bother with a Japanese release, but definitely put more effort into Europe (which had the Atari coputer reputation to build on). (maybe even aim at a simultaneous US EU/UK release)

 

As for side projects, I think they'd already dropped pretty much everything else for the Jag, computer developments were cancelled and remaining inventory sold off (with their computer market being almost dead by that time), and I'm not sure about the Lynx, but I think they may have still been making a modest profit on it. (not very significant though, but not in circumstances where dropping support would be advantageous)

 

And even so, focusing purely on the Jaguar still wouldn't be enough to forego the early release, they needed funding, hence personal funding being th only out in these circumstances.

 

 

1. Change the controller. Do away with the keypad and add three more fire buttons like on the Pro Controller, and possibly ad shoulder buttons too.

 

The pro controller does have shoulder buttons; as to th ekeypad, I'd keep it (even with reduced # of keys) if they were at all interested in pushing a number of PC ports: many requiring numerous action keys. (X-Wing I gave as a previous example) If those kind of games don't matter, than yes, definitely go with th epro controller minus keypad. (which was virtually identical in layout to the Sega aturn controller)

 

As for the rest of your suggestions, I already addresed the time/funding issues (allowing for a later release date), but as to these:

 

4. Promote the "New Atari". Schedule press conferences, trade shows, interviews, and stress that the old Atari is gone and a new way of doing business and supporting companies and customers alike is in. This would help the negative perceptions of the past. This would help attract third party developers.

 

I don't think so, the oldschool charm was part of the attraction, they'd promised updated classics, which could have been an asset, but failed to deliver for the most part. Plus, in Europe, much the public already saw Atari from a different perspective, associated more with their computers than the arcade or VCS. (or 7800 which I beleive was moderately popular) But don't say the "old" atari is dead (technically Atari Inc. was long gone, but that's not the context here), say you'll bring back the best of the old, brought together with the next generation cutting edge gaming experience! (and they could have gotten away with the 64-bit marketing fine if there were enough advanced looking games)

 

 

5. Go after established game licenses. Atari wasted a lot of money on stupid and ill-advised licenses that were never even published! Go after games like Mortal Kombat II. A special edition of MK2 for the Jaguar with an extra character or two would really sell some systems and show other companies Atari meant business.

 

6. Advertise! $3 million was nothing! Advertise on prime time tv, on Cartoon network and Nickelodeon to get the kids excited, on the 1993 Superbowl, on cereal boxes, anywhere! Spend at least $15 million.

 

All of these things could have been done in reality. I know hind sight is always 20/20, but doing just a few of these things, like the different controller and a decent launch line-up would have helped tremendously.

 

I'm sure they did try this, and probably did get some companies onboard with this initially, but lost them due to inherant problems with the hardware, Atari managment, and to Sony's emergence.

 

That's why it's not practical to leave the hardware alone, bugs need fixing, and replacing the 68000 goes a long way as well. In the context of 1993 (assuming the begining of '93), the design is already fairly complete, so a ratical redesign isn't practical, but the nature of the design was to accept any number of CPUs, so replacing the 68000 with an 020, 030, x86 processor, or MIPS CPU would all be relatively simple; the 68EC020 (16.7 MHz model in particular) would probably be the simplest and cheapest route above the 68k. (and would also make the smoothest transtion for any developers already working on the system)

 

Maybe not fix all the bugs, but try to fix as much as possible with the additional time, the 020 switch should add next to nothing to development time, then focus on fixing the MMU bugs, DSP write bugs, and try to double buffer the blitter registers. (also remove the hard coded 6-cycle delay on the DSP, I'm not really sure why they did that in the first place)

Of course still stopping development with sufficient time to get finalized development kits out and give time to start stockpiling consoles.

 

The more they clean up the hardware, the less necessary good tools are; adding the 020 alone goes a long way in that respect. (and with sever bugs fixed, I beleive the existing tools would be a good deal more useful; I think the complier in particular is mainly ruined by the MMU bugs, but I'm not sure on that)

 

With the added wait, maybe costs of faster RAM would come down (not sure on this, and I know DRAM prices overall were rather static in the mid 90s, but I'm not sure how prices for different grades/types changed). If it at all became practical (cost wise), bump it up from 75 ns FPDRAM to 70, 65 ns or maybe even 60 ns DRAM. (allowing system speed to be bumped up to 14.28, 15.38, or 16.67 MHz respectively, the latter being the rated speed of the 68EC020 I mentioned; and both TOM and JERRY running at 2x those speeds -I beleive their design speed was 40 MHz)

Edited by kool kitty89
Link to comment
Share on other sites

On point #4:

 

Many companies simply did not want to do business with Atari and were not willing to give them a chance because of the way Atari had done them wrong in the past. This is what I'm talking about. Promote that the old way of screwing companies is over and show them that Atari could keep promises and be good business partners.

Link to comment
Share on other sites

On point #4:

 

Many companies simply did not want to do business with Atari and were not willing to give them a chance because of the way Atari had done them wrong in the past. This is what I'm talking about. Promote that the old way of screwing companies is over and show them that Atari could keep promises and be good business partners.

 

If you're referring to the 1983/84 crash, any issues with that were long dead, they were probably pretty weel on the way toward resolution in the late 1980s even. (Atari Corp doesn't seem to have had a particularly tought time breaking back into the market with the release of the 2600 Jr. in mid 1985, not in huge quantities but at th esame time they didn't have that much production capacity- according to a statement by Michael Katz at the time, they could have sold a lot more Jrs. in th e1985 holiday season)

 

In any case, I don't think that stigma was any longer an issue with retailers, the problem would be that Atari (either A.Corp or Warner's Atari Games) hadn't really been associated with anyting really significant in the industry in the US since the 7800. (which was only modestly successful) In Europe it was a bit of a different situation with the ST being fialry popular. (not just as a home computer, but video game machine in general, though eventually superceded by the Amiga in this area)

I'm not sure if the Lynx was more popular in Europe (looking at old clips though, I've certainly seen more UK/EU advertizements than US ones). In general they had a more modern following in UK/Europe than in the US.

 

I would however say that Atari Corp shouldn't have been so cheap with some developers and such, they seem to have a bad reputation of driving a hard bargain and not paying on time. (or at least I remember that being expressed several times in discussions here) But that was something ongoing with the Jaguar, not a problem with the "old" image.

Link to comment
Share on other sites

The ST wasn't fairly successful in Europe and the UK it was MASSIVELY successful! It was the best selling 16-bit machine for a very long time and was supported by every major publisher. The Lynx did pretty well over here too, it was carried by all the major retailers and I knew more kids at my school with a Lynx than I did those with a Gamegear. It was also highly regarded by the gaming magazines and always quoted as "THE" Handheld console to own.

Link to comment
Share on other sites

The ST was eventually superceded in popularity by Amiga, right? (both in terms of popularity as a home computer and games being developed/released -with many early games being designed for ST and ported to Amiga irrc) Then PCs finally started pushing ahead in popularity around the early/mid 90s, right?

 

In any case, it seems like Atari really should have pushed more in Europe, even with the buggy hardware and early release. The viral marketing strategy would probably have worked a lot better in Europe as well. (maybe put more effort into getting some European developers on board too, maybe including some that would really push the hardware, like CORE did with the Mega CD)

 

According to this: http://www.atariage.com/forums/topic/132973-comprehensive-atari-jaguar-timeline-1991-2008/page__st__25

 

Atari had announced that the 1993 pre-release would take place in NY, San Fransisco, London, and Paris, but only ended up being released in the 2 US locations...

Edited by kool kitty89
Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...