wait till we are able to get ati cards to work on this project.
the: ATi Radeon HD 4870 X2 has 1600 Shader Processors
the 4870 has 800
even my 3870 has 320
the nVidia GeForce GTX 280 only has 240

i believe thats what does the calculations is the shaders

when we can use ati cards ... my guess is workunits will be done in less than 5 minutes ...

Then why do nVidia cards always beat the ati cards?
Oh wait...
Shader clock = core clock (ati)
Shader clock = 2 x core clock (nVidia)

---

Do you Good Search for Seti@Home? http://www.goodsearch.com/?charityid=888957
Or Good Shop? http://www.goodshop.com/?charityid=888957

Then why do nVidia cards always beat the ati cards?
Oh wait...
Shader clock = core clock (ati)
Shader clock = 2 x core clock (nVidia)

Always? Read the latest hardware reviews. The ATi Radeon 4870 & 4870 X2 are wiping the floor with nVidia. This round definitely belongs to ATi.

---

Or at least in the review's I've read...

Also looking at Folding@home's comparison... Cuda is about 2-3x faster than CAL (At the current optimization level which is admittedly in CUDA's favour)

240*1500=360000
800*750=600000

So fair enough ATI should be faster.. But so far isn't... (in terms of distributed computing)

I guess a better comparison could be the amd phenom (@ 2.0Ghz) to a core2duo (@ 4.0Ghz) Sure the phenom has more cores but the core2duo has more speed... Both could come out on top for speed depending how the program is built.

---

Do you Good Search for Seti@Home? http://www.goodsearch.com/?charityid=888957
Or Good Shop? http://www.goodshop.com/?charityid=888957

It's not a 1:1 ratio

Shader Processors != Stream Processors

It'll still be interesting to see how they stack up in a science app side by side.

---

Or at least in the review's I've read...

Also looking at Folding@home's comparison... Cuda is about 2-3x faster than CAL (At the current optimization level which is admittedly in CUDA's favour)

240*1500=360000
800*750=600000

So fair enough ATI should be faster.. But so far isn't... (in terms of distributed computing)

I guess a better comparison could be the amd phenom (@ 2.0Ghz) to a core2duo (@ 4.0Ghz) Sure the phenom has more cores but the core2duo has more speed... Both could come out on top for speed depending how the program is built.

Since ATi hasn't released an official SDK for science apps yet, I think its premature to conclude that nVidia is faster for DC. We both know what can happen with a good programmer and a decent compiler.

Really considering how buggy the app is faster just means it will return invalid results quicker.

Also I'm pretty sure Nvidia is quicker because there isn't an ATI app and it's not like people can port the Nvidia app to ATI until it is stable.

---

I think it'll all boil down to several things.

Mostly the Drivers and the actual Shader Processors ability to process. If the 1600 Shader Processors have a limited instruction set, it might take 10 Shader Processors to compute the same amount of data as a single nVidia Stream Processor.

I'm not saying I have any basis for these estimates, but I've yet to see good benchmarks comparing the two.

---

Really considering how buggy the app is faster just means it will return invalid results quicker.

Also I'm pretty sure Nvidia is quicker because there isn't an ATI app and it's not like people can port the Nvidia app to ATI until it is stable.

A lack of a working application doesn't mean that nVidia is/will be faster given proper coding for an ATi chip.

I think it'll all boil down to several things.

Mostly the Drivers and the actual Shader Processors ability to process. If the 1600 Shader Processors have a limited instruction set, it might take 10 Shader Processors to compute the same amount of data as a single nVidia Stream Processor.

I'm not saying I have any basis for these estimates, but I've yet to see good benchmarks comparing the two.

Or the shader processors may have better programmability and be able to do the work in 1 shader processor as nVidia does in 10 Stream Processors. We don't know until there's a working sample that has been properly coded.

All I wanna know is how the heck do I get all 100+ shaders to run 100+ WUs on my Nvidia?? I set the max processor usage to 100 but the slots are only coming up as 4. Im not even sure how many this 8800 has asfar as cores but the thing does one WU in 1.5 mins with one .04 cores in use.. LOL!!
Any ideas or info guys and gals?

---

Not Ready Reading BRAIN. Abort/Retry/Fail?

All I wanna know is how the heck do I get all 100+ shaders to run 100+ WUs on my Nvidia?? I set the max processor usage to 100 but the slots are only coming up as 4. Im not even sure how many this 8800 has asfar as cores but the thing does one WU in 1.5 mins with one .04 cores in use.. LOL!!
Any ideas or info guys and gals?

You cant run more than 1 wu per vid card. The app is written to use all available shaders. In other words it already using all cores...

---

Do you Good Search for Seti@Home? http://www.goodsearch.com/?charityid=888957
Or Good Shop? http://www.goodshop.com/?charityid=888957

hummm... ok.. Thanks :)
well it is faster but if I could use all 4 cores of the quad to do APs than it'd be better. Is there a way to change the number of slots? I have only 0-3.

---

Not Ready Reading BRAIN. Abort/Retry/Fail?

as far as to which is faster, they will switch back and forth as the new versions are released. ATI will release a card that is faster than the Nvidia latest. Then Nvidia will release another in a couple months that is faster than that. Same with Intel and AMD...

And there is going to be the ATI is faster on project A while NVidia is faster on project B to come...

Kinda moot really right now as ATI is not capable of doing BOINC ... yet ...

---

as far as to which is faster, they will switch back and forth as the new versions are released. ATI will release a card that is faster than the Nvidia latest. Then Nvidia will release another in a couple months that is faster than that. Same with Intel and AMD...

Normally I'd agree with you, but the fact is nVidia has been beating the pants off ATi for the last two years. This is a bright moment for ATi, a chance to show that they're still in the game, and a chance to offer some real competition to nVidia (as with Intel vs. AMD).

Any discussions of one being faster than the other for crunching is just speculation at this point, because, as you stated, ATi isn't in the game - yet. Hopefully they'll get onboard with this open source OpenCL and BOINC will/should support it, as should nVidia. For ATi to join nVidia's CUDA program would require ATi to pay royalties to nVidia (again, similar to Intel vs. AMD) and could keep ATi from being able to suggest new features in future specifications or releases of CUDA, so I'd rather see a neutral consortium like OpenCL keep things fair and balanced.

as far as to which is faster, they will switch back and forth as the new versions are released. ATI will release a card that is faster than the Nvidia latest. Then Nvidia will release another in a couple months that is faster than that. Same with Intel and AMD...

And there is going to be the ATI is faster on project A while NVidia is faster on project B to come...

Kinda moot really right now as ATI is not capable of doing BOINC ... yet ...

The thing that may help ATI in the long run is the fact that their GPUs will do double precision FP (NVidia is single precision only).

---

Boinc Button Abuser In Training >My Shrubbers<

The thing that may help ATI in the long run is the fact that their GPUs will do double precision FP (NVidia is single precision only).

If there is a call for it, Nvidia will add it soon enough.

---

Really considering how buggy the app is faster just means it will return invalid results quicker.

Also I'm pretty sure Nvidia is quicker because there isn't an ATI app and it's not like people can port the Nvidia app to ATI until it is stable.

A lack of a working application doesn't mean that nVidia is/will be faster given proper coding for an ATi chip.

It does imply this discussion is pointless... Even more so because CUDA doesn't run on ATI it will not be a apples to apples comparison.

---

It does imply this discussion is pointless... Even more so because CUDA doesn't run on ATI it will not be a apples to apples comparison.

Well, in this case the "apples" are the tasks and if one card runs them faster than the other, that is a valid comparison. But, you are correct that it is somewhat pointless in that there is no application for the ATI type cards at the current time and so ...

---

Well the reason I say that is all of the discussions on how because the optimized app was different between Mac and PC, for example, it wasn't
"fair" as some of the threads were saying at the time. Really I don't see how it makes that big of a difference either way.

---