What could have saved the Jag?

[+GoldLeader]

9 minutes ago, zzip said:

I don't even think this matters that much.   Marketing has always abused tech specs to make hardware sound more impressive than it is.   The average consumer can't even explain what a 32-bit or 64-bit system actually means in real-world terms..  ( It just creates years of debate as to whether it is or isn't 64 bit )

 

But I do think Do The Math" slogan was terrible (games are fun, math is not for most people, why associate math with games?  LOL)  And if you actually do the math, it's hard to come up with 64-bits as your answer.

 

Nintendo: Now Your Playing With Power!

Sega: Sega Does What Nintendon't

Atari: Do The Math!

 

 

All that Power and not much to show for it...

 

21 minutes ago, 42bs said:

There is not one console w/o bugs. This is a lame excuse. There have been killer games on C64 which was weaker and full of bugs. 

 

I assume he's talking about the Jaguar being hard to program for (not to mention Atari's lack of dev support)...And isn't that the accepted wisdom by this point?

[laymanpigeon]

1 hour ago, zzip said:

No the problem is 3DO is not Nintendo or Sega or even Sony.

I never stated such in here nor I have ever though either of three were cause of its demise.

Again I said that price was issue because production wasn't handled by the 3DO company.

Instead production license was issued to various manufactures that then produced it.

 

Some did made changes that deviated from specifications thus created incompatibilities with some games.

1 hour ago, zzip said:

So even less mindshare than Atari going in and with $700+ consoles to boot.

It did not have mindshare at the start because of being new company with first product.

Though person leading the company was previously IIRC CEO of Electronics Arts.

Thus Electronics Arts was major supporters of 3DO company and console.

1 hour ago, zzip said:

Nobody's buying that except for a small band of curiousity-seekers with extra money to burn.

I don't think around 2 million 3DO's would have been produced if it was simply

"a small band of curiousity-seekers with extra money to burn".

Probably same type of crowd that would have bought Neo Geo or Neo Geo CD.

1 hour ago, zzip said:

Again this doesn't fix the real problems-  lack of killer IPs (AvP is a start, but it needed more that that)  and lack of marketing budget to go against competators who are in a much stronger financial position by this point.

You do not have an argument when you assert that I said which I did not that it would have solved problems.

I said it would have been an band-aid that you intentionally ignore because again you have no argument.

SDRAM would have lessened impact of hardware bugs and architectural deficiencies with low latency.

 

"The Blitter could do basic texture mapping of horizontal and vertical spans, but because there wasn't any caching involved, every pixel caused two ram page misses and only used 1/4 of the 64 bit bus. Two 64 bit buffers would have easily tripled texture mapping performance. Unfortunate." - John Carmack

 

"SDRAM performance is dramatically improved over that of FPM or EDO RAM. Because SDRAM is still a type of DRAM, the initial latency is the same, but overall cycle times are much faster than with FPM or EDO. SDRAM timing for a burst access would be 5-1-1-1, meaning that four memory reads would complete in only 8 system bus cycles, compared to 11 cycles for EDO and 14 cycles for FPM. This makes SDRAM almost 20% faster than EDO. Besides being capable of working in fewer cycles, SDRAM is also capable of supporting up to 133MHz (7.5ns) system bus cycling."

Upgrading and Repairing Servers 2006 - Early Server RAM Types: DRAM, EDO DRAM, and SDRAM

 

Most damning deficiency of Jaguar's hardware is in anemic ability to draw and render textured polygons along data not being cached.

Thus SDRAM with its very low latency could allow double render output of textured polygons, whole system would have been faster.

Game lke Skyhammer would not have been 5 frames per second when right in between two building being rendered along floor/road.

 

It is because of latency that Nintendo 64 was effectively chained down due to atrocious RDRAM of few bits data bus and lots of bandwidth.

Edited August 23, 2022 by laymanpigeon

[+CyranoJ]

If an influx of new people won't save the forum from this drivel, then an influx of new games in the 90s wouldn't have saved the jaguar from checkered flag. 

[Frozone212]

2 hours ago, 42bs said:

There is not one console w/o bugs. This is a lame excuse. There have been killer games on C64 which was weaker and full of bugs. 

But

Uses

Great

Software

[+Sauron]

4 hours ago, laymanpigeon said:

You do not have an argument when you assert that I said which I did not that it would have solved problems.

I said it would have been an band-aid that you intentionally ignore because again you have no argument.

SDRAM would have lessened impact of hardware bugs and architectural deficiencies with low latency.

You're making absolutely no sense, nor do you seem to have any grasp of what the gaming market was like back in the early to mid 90s. Atari was facing an uphill battle on every front imaginable. The Jag's hardware bugs were the least of the problems they faced. Here's just a few of the main problems I'd say were far more important than the Jag's hardware:

 

- Lack of marketing budget

- Lack of retailers, thanks to Atari burning bridges with all but a select few

- Lack of AAA developers willing to take a risk on developing a AAA game on unproven hardware from a company that most seemed to loathe

- A gaming press that was largely hostile to Atari, again thanks to Atari's own past actions

- Lack of upper management with experience with launching successful products

- Lack of developer support

 

The Jag being more powerful wouldn't have done anything to fix any of these problems, and saying otherwise is simply pure ignorance.

 

[+cubanismo]

11 hours ago, 42bs said:

Sure, they did:

https://en.wikipedia.org/wiki/R4000

Not to mention one of my childhood computing dreams:

 

https://en.wikipedia.org/wiki/DEC_Alpha

[+cubanismo]

10 hours ago, zzip said:

But I do think Do The Math" slogan was terrible (games are fun, math is not for most people, why associate math with games?  LOL)  And if you actually do the math, it's hard to come up with 64-bits as your answer.

I actually loved this ad campaign as a kid. I *thought* I knew enough about computers to think 64-bit meant things about stuff, and while those that were old enough to own 2600's probably rightfully thought all the in-your-face attitudes were a bit corny, I was just the right age to think the things going on in Jaguar ads and the aggressive aesthetic of the marketing and machine were cool. Then again, I didn't buy a Jaguar until a year or two ago, so maybe it didn't work as well as I thought it did. I ended up buying a Saturn at the time, mostly as a device to play Daytona USA at home.

 

 

[zzip]

17 hours ago, laymanpigeon said:

It did not have mindshare at the start because of being new company with first product.

Though person leading the company was previously IIRC CEO of Electronics Arts.

Thus Electronics Arts was major supporters of 3DO company and console.

Yes Trip Hawkins was behind 3DO.   However the average consumer doesn't follow gaming news closely,  so while they might recognize Electronic Arts, that brand recognition doesn't automatically transfer to '3DO'

 

17 hours ago, laymanpigeon said:

I don't think around 2 million 3DO's would have been produced if it was simply

"a small band of curiousity-seekers with extra money to burn".

2 Million is nothing.   Hit consoles sell tens of millions or even over 100 million.

 

17 hours ago, laymanpigeon said:

You do not have an argument when you assert that I said which I did not that it would have solved problems.

I said it would have been an band-aid that you intentionally ignore because again you have no argument.

SDRAM would have lessened impact of hardware bugs and architectural deficiencies with low latency.

 

"The Blitter could do basic texture mapping of horizontal and vertical spans, but because there wasn't any caching involved, every pixel caused two ram page misses and only used 1/4 of the 64 bit bus. Two 64 bit buffers would have easily tripled texture mapping performance. Unfortunate." - John Carmack

I'm just saying there is no magic hardware fix that would have saved the Jag.   The real 'bug' was in the Atari organization itself.

[zzip]

8 hours ago, cubanismo said:

I actually loved this ad campaign as a kid. I *thought* I knew enough about computers to think 64-bit meant things about stuff, and while those that were old enough to own 2600's probably rightfully thought all the in-your-face attitudes were a bit corny, I was just the right age to think the things going on in Jaguar ads and the aggressive aesthetic of the marketing and machine were cool. Then again, I didn't buy a Jaguar until a year or two ago, so maybe it didn't work as well as I thought it did. I ended up buying a Saturn at the time, mostly as a device to play Daytona USA at home.

Yeah increasing bit size doesn't give as many benefits as people think.   It usually gives access to more memory..   much more than Jag could ever hope to have in the 90s.   As for performance..  if you use 64-bit integers you will have a benefit,  but those are big numbers, 32-bit  is enough in most cases.   I've seen cases where going to 64-bit actually caused a slight performance hit.  

[42bs]

Actually, the size of an bit has been always 1.

[agradeneu]

1 hour ago, zzip said:

Yeah increasing bit size doesn't give as many benefits as people think.   It usually gives access to more memory..   much more than Jag could ever hope to have in the 90s.   As for performance..  if you use 64-bit integers you will have a benefit,  but those are big numbers, 32-bit  is enough in most cases.   I've seen cases where going to 64-bit actually caused a slight performance hit.  

64 bit bus means more and faster data throughput for graphics and sound. I think a legit concept for a video game system that was designed for rendering fast animated graphics.

 

 

[zzip]

11 minutes ago, agradeneu said:

64 bit bus means more and faster data throughput for graphics and sound. I think a legit concept for a video game system that was designed for rendering fast animated graphics.

sure, if you can keep the bus filled.   When you have a bunch of 32 and even 16-bit chips pushing data, I think that's why there are bottlenecks.

[laymanpigeon]

19 hours ago, Sauron said:

You're making absolutely no sense, nor do you seem to have any grasp of what the gaming market was like back in the early to mid 90s.

It is very ironic for you to assert I don't make any sense.

19 hours ago, Sauron said:

The Jag's hardware bugs were the least of the problems they faced.

That is outright false and utter nonsense to assert such while ignoring costs that it incurred for developers and publishers themselves.

Because of hardware bugs more time was required to develop or port a game for Jaguar, thus considerably greater development costs.

Hence more copies of a game have to be sold in order to recoup costs and make a profit to justify its existence on the platform.

 

Another is that because of hardware bugs development is longer and time of window when a game can exploit the market shrinks.

In 1992 the market for games that are full motion video or have such has crashed and burned, FMVs again being relevant in 1995.

It wasn't until 1995 or 1996 when video games that had 3D polygon graphics were solidified as the future when such matured.

 

SDRAM having far lower latency than FPM DRAM would have improved performance of entire hardware that would idle less.

Games that were on Jaguar would have performed better and or very at least could have been released sooner.

Poor performance of games on Jaguar and a drought until near end 1994 is what hit the nail in the coffin.

19 hours ago, Sauron said:

- Lack of marketing budget

- Lack of retailers, thanks to Atari burning bridges with all but a select few

- Lack of AAA developers willing to take a risk on developing a AAA game on unproven hardware from a company that most seemed to loathe

- A gaming press that was largely hostile to Atari, again thanks to Atari's own past actions

- Lack of upper management with experience with launching successful products

- Lack of developer support

Why do I have gut feeling that this is persecution complex on the level of Gamecube or Wii U that some Nintendo fans have?

Electronics Arts has supported Atari ST despite being Atari home computer and far smaller userbase than home consoles.

Existence of 3DO probably cockblocked Atari from Electronic Arts supporting Jaguar since its potential 3DO competitor.

19 hours ago, Sauron said:

The Jag being more powerful wouldn't have done anything to fix any of these problems, and saying otherwise is simply pure ignorance.

At bare minimum Atari should have launched Jaguar in early 1994 as by then more software could have been on launch day.

Or at least there would not have been a drought and impression on Atari and Jaguar would have been better.

Hardware bugs that Jaguar has is major roadblock to developers and publishers in terms of cost and time.

Instead of disaster of rushing it in order to be among first one thus half baked product and lineup.

Implementing SDRAM would have been faster than when critical bugs were fixed by IIRC 1995.

Again, my suggestion is band-aid and would not make Jaguar more powerful by any means.

Unless delay to early 1994 could have been enough time to solve critical hardware bugs.

 

[Punisher5.0]

[JagChris]

22 minutes ago, laymanpigeon said:

Because of hardware bugs more time was required to develop or port a game for Jaguar, thus considerably greater development costs.

Hence more copies of a game have to be sold in order to recoup costs and make a profit to justify its existence on the platform.

 

Another is that because of hardware bugs development is longer and time of window when a game can exploit the market shrinks.

The hardware bugs did make most at Atari think getting a compiler to work for the GPU an impossible task. Though they turned out to be incorrect. 

 

If they thought otherwise they might have attempted one sooner instead if what they termed a way late Hail Mary attempt.

 

Was it the least of their problems?  That's probably debatable.

[+Sauron]

51 minutes ago, laymanpigeon said:

It is very ironic for you to assert I don't make any sense.

There's no irony in it. I'm just stating a fact.

 

54 minutes ago, laymanpigeon said:

That is outright false and utter nonsense to assert such while ignoring costs that it incurred for developers and publishers themselves.

Because of hardware bugs more time was required to develop or port a game for Jaguar, thus considerably greater development costs.

Hence more copies of a game have to be sold in order to recoup costs and make a profit to justify its existence on the platform.

You're drastically overstating the issue here, and I stand by my assertion that the other problems I listed were far bigger issues. 

 

[42bs]

11 hours ago, laymanpigeon said:

That is outright false and utter nonsense to assert such while ignoring costs that it incurred for developers and publishers themselves.

Because of hardware bugs more time was required to develop or port a game for Jaguar, thus considerably greater development costs.

In 30yrs in software business software developers tend to blame others for not reaching their deadlines. Either they blame it on hardware bugs or missing tools.

I see it myself. Something acts weird, so it must be a hardware issue. But often it is simple not anticipating the quirks of the hardware.

If you buy a Beetle it is still a Beetle no matter if they tell you it is a Porsche.

I assume, many developers tried to squeeze something into this Beetle to make it look like a Porsche.

[zzip]

14 hours ago, laymanpigeon said:

Electronics Arts has supported Atari ST despite being Atari home computer and far smaller userbase than home consoles.

Electronic Arts started out publishing on home computers and didn't touch want to touch consoles (until 88/89).  They actually came out and threw their weight behind the Amiga which was unusual for a third-party software publisher to pick sides.  But since the Atari ST started off outselling the Amiga by a significant margin, they had to support that too.  

 

 

[laymanpigeon]

22 hours ago, JagChris said:

The hardware bugs did make most at Atari think getting a compiler to work for the GPU an impossible task. Though they turned out to be incorrect. 

 

If they thought otherwise they might have attempted one sooner instead if what they termed a way late Hail Mary attempt.

 

Was it the least of their problems?  That's probably debatable.

Making a compiler should have been first thing on the list as priority as it adds into developer support be it 1st, 2nd or 3rd party.

At bare minimum it would have made porting an easier process, in best case for some it would have provided better performance.

Those that want to dabble with hardware more directly would probably ignore it, except to be some sort of hardware TLDR.

[JagChris]

One of the programmers was telling us several years ago Atari was basically burning through about ten thousand dollars an hour making them do everything in assembly.

Edited August 26, 2022 by JagChris

[Matt_B]

2 minutes ago, JagChris said:

One of the programmers was telling us several years ago Atari was basically burning through about ten thousand dollars an hour making them do everything in assembly.

That sounds about right.

It's not like Atari seemed to have realized that the console market had moved on from 1986 in many other respects.

 

[JagChris]

1 minute ago, Matt_B said:

That sounds about right.

It's not like Atari seemed to have realized that the console market had moved on from 1986 in many other respects.

 

You think Atari made it to 1986?

 

In 1986 there was Castlevania, Mario World etc.

 

Atari was still stuck.in 2-3 different screen arcade style games. Everyone else was going on grand adventures. 

 

I think scrapyard dog is the only exception.

[agradeneu]

8 hours ago, Matt_B said:

That sounds about right.

It's not like Atari seemed to have realized that the console market had moved on from 1986 in many other respects.

 

NES, SNES and Game Boy games were programmed mostly in assembly too. Otherwise the performance would have been crippled. 

 

https://ersanio.gitbook.io/assembly-for-the-snes/

 

I still believe the programming part was not the weak point of the Jag games, but the overall counterproductive development environment, cost cutting budgets by Atari and general lack of good design and artwork. 

I agree with Sauron that no improvement on the hardware would have fixed the general issues of developing games at Atari and the general lack of development support by 3rd parties. 

The Jaguar itself was powerful enough to eclipse SNES and Genesis for sure, but handling huge data and having the better numbers does not automatically transfer to better games. 

Edited August 26, 2022 by agradeneu

[Chilly Willy]

Nearly all the libraries for the Sega Saturn are assembly. It was that generation of console that started programming games in C, but a considerable number were still all assembly. It was also about this time that compilers were finally generating stable and (reasonably) fast code. The next generation of consoles moved almost entirely to C/C++, and assembly fell by the wayside. Not having a good JRISC compiler didn't help the Jaguar any, but it also wasn't the reason it failed. Too much is put on that belief for the time period the Jaguar occupied. The Jaguar 2 would have been the console to NEED a good compiler. People today say the Jaguar needs a good C compiler because it would make our job doing homebrew easier. That affects many of us, myself included. I work more on the 32X because it has great compiler support. I use the latest gcc, which can be almost as fast as the very best hand-written assembly.

[JagChris]

Ken Rose said they favored C code even before that and he got Sega to switch to C. And Ataris development cycle was too long without it.

 

According to Atari the compiler for the jrisc was a hail Mary attempt because programmers had already moved to C by then and the average programmers were having trouble wrestling with the GPU/dsp in assembly. 

 

It may not have been the complete reason it failed but from accounts it sounds like a part of it.