ATI - PS3 is Unrefined

[expletive]

Even if I was making an educated guess...90% of the time I would be right because I had time on my side. Besides...if you read many of threads on this forum, in terms of raw performance, RSX is clearly ahead. overall, the system [PS3] will most-likely have a little more umph than XBOX360...almost all major devs have admitted this (even John Carmack, who notoriously still prefers the PC, Microsoft, and anything with any relation to Bill Gates) .

Do you not agree?

Or are you arguing just for arguments sake?

Besides, technical power has little to do with how well a console does. Quality content is the biggest differentiating factor.

Actually a lot of people on these forums feel Xenos is the superior GPU by a significant margin and raw 'performance' is not always derived by raw 'specs' which is what youre saying here.

Is the x1800xt faster than the 7800gtx across the board? No. So several months difference does NOT always mean chip A is summarily more powerful/faster than Chip B.

Besides, the question was "do these numbers mean that RSX is > than Xenos" and you said "yes but only a little." Which is inaccurate because these numbers mean very little if youre really trying to determine if 'RSX is > Xenos", particularly because they ARE so close.

Whether or not the PS3 as an ENTIRE SYSTEM ends up being faster than the 360 still remains to be seen but if it does, it may or may not be because of RSX and most likely will have VERY little to do with ~255 vs ~240 GFLOPS on the GPUs.

 

[expletive]

Now, I must ask you, "Are you being serious?" Because if you think he's being purely theoretical there and not refering to PS3 and XBOX360...you are guilty of wishful thinking.

Life is full of things which must be inferred and extrapolated...you don't need to be beat over the face with things to know something is true. This type of thing just breeds cynicism.

How could he NOT be theoretical at that point in time, (EDITED out for cohesion) he could only be basing his comments on 2 dev kits, becuase thats what was available then (adn still is now for the PS3).

So youre saying that Carmack can make a definitive statement on the final performance on these two machines based on what he had at the time of his comment (which is basically what i described above)?

 

[j^aws]

Okay, can we agree on 216 GFlops Dave for the sake of argument and satisfy Aarons source?

 

[ROG27]

How could he NOT be theoretical at that point in time, (EDITED out for cohesion) he could only be basing his comments on 2 dev kits, becuase thats what was available then (adn still is now for the PS3).

So youre saying that Carmack can make a definitive statement on the final performance on these two machines based on what he had at the time of his comment (which is basically what i described above)?

Devs pretty much know which one is going to have more under the hood, according not only to the dev kits but the specific specs they are given (which I'm sure is not PR speak that is given to the general public)

Do you think the president of EA would claim that PS3 has a bit more under than 360 if he didn't know...or have far more knowledge than we have?

I'm not claiming a huge difference here...but if you want to put a nail in the coffin for argument's sake.

 

[weaksauce]

Whether or not the PS3 as an ENTIRE SYSTEM ends up being faster than the 360 still remains to be seen but if it does, it may or may not be because of RSX and most likely will have VERY little to do with ~255 vs ~240 GFLOPS on the GPUs.

Hey I know nothing about this :smile: but 255 is for 7800gtx, right? I googled and found this:
http://www.answers.com/topic/playstation-3
# 1.8 TFLOPS floating point performance, 356 GFLOPS programmable

Hey there's some more number there:
# 280 shader operations per cycle
# 154 billion shader operations per second

Seems a bit high, is it together with CPU?

Anyhow, then there's this:
http://www.google.com/search?q=cache:3vUqJvo0YgIJ:pc.watch.impress.co.jp/docs/2005/0701/kaigai_5.pdf

edit:changed link

 

[expletive]

Do you think the president of EA would claim that PS3 has a bit more under than 360 if he didn't know...or have far more knowledge than we have?

I'm not claiming a huge difference here...but if you want to put a nail in the coffin for argument's sake.

Honestly, I think theres a good chance the president of EA has very little idea "whats under the hood".

Read this and tell me if you still think execs are really in touch with the hardware:

http://www.beyond3d.com/forum/showthread.php?t=26853

I dont know what you mean about the 'nail in the coffin' thing so i cant really respond to that.

 

[Titanio]

Hey I know nothing about this :smile: but 255 is for 7800gtx, right?

Clocked at 550Mhz - as the expected RSX - yes.

I googled and found this:
http://www.answers.com/topic/playstation-3
# 1.8 TFLOPS floating point performance, 356 GFLOPS programmable

Hey there's some more number there:
# 280 shader operations per cycle
# 154 billion shader operations per second

Seems a bit high, is it together with CPU?

No. The 356Gflops programmable is probably including flops from the mini-ALUs and FP16 normalisation etc. The 255 figure is counting the 2 main ALUs in the pixel shaders only (and the vertex shaders, of course).

 

[weaksauce]

Clocked at 550Mhz - as the expected RSX - yes.




No. The 356Gflops programmable is probably including flops from the mini-ALUs and FP16 normalisation etc. The 255 figure is counting the 2 main ALUs in the pixel shaders only (and the vertex shaders, of course).

Oh, ok. And how much does xenos do with all that? Or is it not important?

 

[ihamoitc2005]

mini-ALU

Oh, ok. And how much does xenos do with all that? Or is it not important?

What is interesting is mini-ALU and purpose and utilization of such design.

 

[Dave Baumann]

We cannot know what is behind mind of EA, maybe they make guesses maybe no, but maybe a we can know a little of Carmacks mind because he was technical advisor for Microsoft Xbox division.

Carmack is on other technical steering comittees, such as NVIDIA's for example.

 

[aaronspink]

http://www.watch.impress.co.jp/game/docs/20050520/x360_g.htm

They don't go over the shader arrays in any of the foils in that presentation and all the other data except for watch.impress points to 240.

Not being funny, but I'd like a link please? I've been posting that number on this forum for more than a year...

Look at the first page or so of Dave's Xenos article where is denoted things in black as being based directly on information from ATI, and things in blue being based on his deductive reasoning.

Aaron Spink
speaking for myself inc.

 

[aaronspink]

Okay, but 216 and 240 would still be 5D though.

but what would the 24 scalar flops be? And ADD? A MUL? An ADD or a MUL? It really doesn't make sense for it to be anything but a MADD.

Aaron Spink
speaking for myself inc.

 

[Farid]

Thread Pruned

The thread has been pruned from the OT posts.

Keep on topic, and make thoughtful posts, not one-liners questions.

 

[Lysander]

I remember from spanish slides from MS Campus party in Valencia in august 2005 that there was clearly 240 programmable gflops tag on 360 gpu (+ "something" of non-programmable gflops). I can`t find link with pictures anymore.

 

[weaksauce]

I remember from spanish slides from MS Campus party in Valencia in august 2005 that there was clearly 240 programmable gflops tag on 360 gpu (+ "something" of non-programmable gflops). I can`t find link with pictures anymore.

This?

http://news.teamxbox.com/xbox/8862/Xbox-360-QWERTY-Keyboard-Revealed/

 

[Lysander]

This?

http://news.teamxbox.com/xbox/8862/Xbox-360-QWERTY-Keyboard-Revealed/

 

[j^aws]

They don't go over the shader arrays in any of the foils in that presentation and all the other data except for watch.impress points to 240.

As mentioned to Dave earlier, it's derived in the text. And 216 is a non-obvious derivation. Of course 240 is reported elsewhere but the 'leak' still says 216...

Look at the first page or so of Dave's Xenos article where is denoted things in black as being based directly on information from ATI, and things in blue being based on his deductive reasoning.

See my replies to Dave...

but what would the 24 scalar flops be? And ADD? A MUL? An ADD or a MUL? It really doesn't make sense for it to be anything but a MADD.

I'm not disagreeing that it makes sense to have a MADD for scalar. Maybe for vec3+scalar you do but not for vec4+scalar. However, low level details on how the instructions are scheduled across the 3 SIMD engines are locked away unfortunately...

 

[j^aws]

This?

http://news.teamxbox.com/xbox/8862/Xbox-360-QWERTY-Keyboard-Revealed/

Those CPU and GPU numbers still do not agree with the technical leak doc...

E.g the CPU ~ 90 Gflops @ 3.2 Ghz. This would also likely apply to CELLs PPE...

 

[TurnDragoZeroV2G]

216 doesn't seem so strange, since it can be done with a whole number. But the 90 doesn't make much sense. That one seems totally random.

I take it you certainly would mind passing that leak on...

 

[London Geezer]

216 doesn't seem so strange, since it can be done with a whole number. But the 90 doesn't make much sense. That one seems totally random.

I take it you certainly would mind passing that leak on...

You don't have it?

You know, there's a reason people always put it in between quotes... because it's not a leak which is supposed to be secret and not supposed to be seen by anyone, it's a "leak" - everyone has it, not very secret...

Not sure that made sense, but in short, everyone saw it!! Where have u been!