Stream processors...

[hans007]

I think the way ATI measures it is that each shader unit is seperate and can do 1 instruction at once.

the Nvidia way to do it I believe groups 5 together, and can do "up to 5" depending on the shader code.

So at peak a single shader or nvidia isl iek 5 ati shaders. but the problem is it is not always doing 5 at a time.

 

[Cuular]

Swap ATI and Nvidia in that post and it's a simple way of stating it.

 

[Nda]

Originally posted by: Zap
That's the equivalent to how many rodents running on wheels it takes to power the cards (wheels hooked up to generators). The big difference comes from the way they measure things. ATI uses mice, while NVIDIA uses rats. Obviously rats are bigger and thus can do more work, so 240 rats are kinda-sorta equivalent to 800 mice in performance, with slight variations of course depending on individual rodent strength, age and working conditions.

LMAO, best explanation i've heard!

 

[schneiderguy]

Originally posted by: keysplayr2003

Originally posted by: Azn

Originally posted by: Cuular
Unfortunately in the case of ATI and nvidia there is no shared definition of stream processor.

Each company defines an SP differently.

For ATI they have a processing unit with 5 "streams" in it, which could in the right circumstances do 5 things at once. So in reality they only have 160 real processors(800=5*160), but to make it look better, the marketing people say it's 800.

So if you look at the number of real processing units, not the maximum number of operations that could be going on at once, with the best workload imaginable, it's amazing that the 160 processors in the ATI card can meet or sometimes beat the 240 in the nvidia card.

Since the general public can only understand larger numbers ATI uses the "800 stream processors" as their number instead of 160 multi-stream processors. Technophiles understand the amazing work that 160 real processing units can keep up with 240. The public in general doesn't.

It still does more processing. That's all that matters.

So those other 640 sp's sit by idly and watch the 160 do all the work then? LOL

The average usage rate for most games is around 3.5/5 for ATI's stream processors, so on average there are 240(48)sp sitting idle. There was an ATI presentation powerpoint about it a while back, before rv770 launched IIRC. I think the application with the highest usage they had listed was Bioshock (where ATI does very well in) and it had a usage rate of over 4/5.

 

[AzN]

Originally posted by: cmdrdredd

Originally posted by: Azn

Originally posted by: keysplayr2003

Originally posted by: Azn

Originally posted by: Zap
That's the equivalent to how many rodents running on wheels it takes to power the cards (wheels hooked up to generators). The big difference comes from the way they measure things. ATI uses mice, while NVIDIA uses rats. Obviously rats are bigger and thus can do more work, so 240 rats are kinda-sorta equivalent to 800 mice in performance, with slight variations of course depending on individual rodent strength, age and working conditions.

Good analogy but you are forgetting one thing. SP clock speeds. How fast those mice and rats run to make the wheel spin. ATI 800SP are a little bit more powerful than Nvidia's 240SP at current clocks. In theory at least.

How are you figuring this? Are you talking about shader clock speeds?
In that case:

800sp * 750MHz = 600,000 (theoretical RP (rodent power))
240sp * 1300MHz = 312,000 (theoretical RP (rodent power))

Looks like Nvidia's rodents get a lot more done with a lot less.

If you meant something else, let me know.

Ummm no...

Nvidia GTX 280
3 cycles * 240SP * 1296mhz = 933 GFLOP

ATI 4870
2 cycles * 800SP * 750mhz = 1.2 TFLOP

Rodents or mice does it matter? In the end ATI gets more work done.

And gets lower fps in a majority of tests relating to single gpu performance?

That would be because games aren't SP limited for the most part. There are exceptions however.

 

[AzN]

Originally posted by: schneiderguy

Originally posted by: keysplayr2003

Originally posted by: Azn

Originally posted by: Cuular
Unfortunately in the case of ATI and nvidia there is no shared definition of stream processor.

Each company defines an SP differently.

For ATI they have a processing unit with 5 "streams" in it, which could in the right circumstances do 5 things at once. So in reality they only have 160 real processors(800=5*160), but to make it look better, the marketing people say it's 800.

So if you look at the number of real processing units, not the maximum number of operations that could be going on at once, with the best workload imaginable, it's amazing that the 160 processors in the ATI card can meet or sometimes beat the 240 in the nvidia card.

Since the general public can only understand larger numbers ATI uses the "800 stream processors" as their number instead of 160 multi-stream processors. Technophiles understand the amazing work that 160 real processing units can keep up with 240. The public in general doesn't.

It still does more processing. That's all that matters.

So those other 640 sp's sit by idly and watch the 160 do all the work then? LOL

The average usage rate for most games is around 3.5/5 for ATI's stream processors, so on average there are 240(48)sp sitting idle. There was an ATI presentation powerpoint about it a while back, before rv770 launched IIRC. I think the application with the highest usage they had listed was Bioshock (where ATI does very well in) and it had a usage rate of over 4/5.

I think ATI is optimizing at driver level. No need for games to be written for it to take advantage of all SP.

 

[cmdrdredd]

Originally posted by: Azn
That would be because games aren't SP limited for the most part. There are exceptions however.

Well, all that matters is the end result. If one card can do 10 processes a second but only gets 40fps vs another that can only do 5 processes a second and gets 60fps, well...which card are you gonna get?

Originally posted by: Azn
I think ATI is optimizing at driver level. No need for games to be written for it to take advantage of all SP.

The problem is, how long should we have to wait for the drivers to be fixed for a game? I could have finished the game by the time we get drivers that give the correct performance.

 

[AzN]

Originally posted by: cmdrdredd

Originally posted by: Azn
That would be because games aren't SP limited for the most part. There are exceptions however.

Well, all that matters is the end result. If one card can do 10 processes a second but only gets 40fps vs another that can only do 5 processes a second and gets 60fps, well...which card are you gonna get?

First off this thread is about technical aspects of SP not which card performs better so which to buy. Games today aren't dictated by SP performance anyway. As newer titles get more SP limited RV770 will most likely do better. Bioshock, Assassin Creed, and GRID contribute to this theory as it performs better than GTX280.

4870 perform in line with GTX260 with much less ROP, TMU, Memory bit bus. No need to go there with 50% faster analogy.

Originally posted by: Azn
I think ATI is optimizing at driver level. No need for games to be written for it to take advantage of all SP.

The problem is, how long should we have to wait for the drivers to be fixed for a game? I could have finished the game by the time we get drivers that give the correct performance.

What are you talking about? You must be in entire different realm as I am because that's not what I was even talking about. I was talking about how drivers are incorporated to take advantage of 800SP in games as keysplayr2003 mentioned that RV770 only works with 160SP to take advantage of games. In fact all 800SP are in use within drivers.

As for who has better drivers you should create another thread and stop jacking this thread. Although I think drivers are good with either Nvidia or ATI.

 

[GaiaHunter]

Originally posted by: Azn

Originally posted by: cmdrdredd

Originally posted by: Azn
That would be because games aren't SP limited for the most part. There are exceptions however.

Well, all that matters is the end result. If one card can do 10 processes a second but only gets 40fps vs another that can only do 5 processes a second and gets 60fps, well...which card are you gonna get?

First off this thread is about technical aspects of SP not which card performs better so which to buy. Games today aren't dictated by SP performance anyway. As newer titles get more SP limited RV770 will most likely do better. Bioshock, Assassin Creed, and GRID contribute to this theory as it performs better than GTX280.

4870 perform in line with GTX260 with much less ROP, TMU, Memory bit bus. No need to go there with 50% faster analogy.

This was the main reason I went with a 4850 over a 9800GTX+ (well and the fact it was some euros cheaper at that time helped too).