Monday, September 28th 2009

NVIDIA Rolls Out its First OpenCL GPU Drivers

In the tussle between its own CUDA GPGPU standard and the OpenCL, NVIDIA is making the right moves by offering its first drivers for OpenCL GPGPU acceleration. The drivers are available for Windows, Linux, and Mac. Any CUDA-capable NVIDIA GPU will be able to use these. The drivers can be downloaded from here, which provide compliance with OpenCL 1.0.

NVIDIA has also released the OpenCL Visual Profiler software that helps developers improve their code by recognizing possible bottlenecks and room for improvements. At a higher level, it profiles actual hardware signals, kernel efficiency, and instruction issue rate; memory latencies; auto-analysis to point of serialization problems, among other things. More information on this can be found here.

Source: TechConnect Magazine
Add your own comment

21 Comments on NVIDIA Rolls Out its First OpenCL GPU Drivers

#1
Breathless
waaa?? Does this add any deliciousness to my OS or games or is this just for devs? :wtf:
Posted on Reply
#2
soldier242
i thought that ATi had OpenCL in their last drivers as well ...
Posted on Reply
#3
Andy77
by: soldier242
i thought that ATi had OpenCL in their last drivers as well ...
"its first"
Posted on Reply
#4
mdm-adph
Come on ATI -- get the fucking ball moving. I'm tired of people with Nvidia cards jumping ahead of me in BOINC stats. :P
Posted on Reply
#5
Roph
It's nice and all being able to encode H.264 with my ATI card, but the encoder is so poor. It is orders of magnitude worse than x264, and is much much worse than even apple's h.264 implementation. I guess if speed is all you care about, it's ok.
Posted on Reply
#6
imperialreign
by: mdm-adph
Come on ATI -- get the fucking ball moving. I'm tired of people with Nvidia cards jumping ahead of me in BOINC stats. :P
ATI/AMD have been working on it: http://www.hpcwire.com/topic/processors/AMD-Submits-OpenCL-for-GPU-to-Review-by-Standards-Body-60000452.html

They already have beta CPU drivers available, and numerous articles and blog postings available on AMD's website.

It's looking like AMD/ATI might be getting their pawns lined up for the official 9.10 CAT release, to corrolate with the first official CAT drivers for the HD5000 series. ATI is pretty good about dropping the hammer hard, we saw how they played out the HD4000 series releases . . . I guess only time will tell, eh?


Anyhow, I'd much rather see an open standard like this become more prevalent - good to see nVidia pushing it forward as well, instead of hiding behind PhsyX . . . now the question is . . . what will Intel do with Larrabee?
Posted on Reply
#7
devguy
IMHO opinion, Nvidia needs to get PhysX running on OpenCL. Then, AMD would be much more willing to accept it (not necessarily saying they would). While PhysX is Nvidia's little baby right now, DirectX 11 physics is likely to take it over. However, DirectX 11 physics is limited to Windows, whereas OpenCL is not. Thus, the inevitable death of PhysX could be postponed quite a bit if both AMD and Nvidia are supportive of it.

Of course, any of that is unlikely to happen, with Nvidia's latest move of excluding PhysX on any machine that detects non-Nvidia hardware (what a dick thing to do). I mean, if someone asks for help in an ATI/Nvidia situation, Nvidia is fully entitled to tell that person they do not support such a configuration and leave it at that. But blocking it all together says something about Nvidia's attitude.
Posted on Reply
#8
qwerty_lesh
by: devguy
i said a bunch of stuff
Its a typical move from Nvidia, sure it says something, but that something has always been part of their practices, look at sli for example. Don't get me wrong either, i'm no amd fanboy but fact is fact, which is what surprises me about nvidia pushing development of an open standard like in this article, its not like them at all :eek:
Posted on Reply
#9
MrMilli
by: devguy
IMHO opinion, Nvidia needs to get PhysX running on OpenCL. Then, AMD would be much more willing to accept it (not necessarily saying they would). While PhysX is Nvidia's little baby right now, DirectX 11 physics is likely to take it over. However, DirectX 11 physics is limited to Windows, whereas OpenCL is not. Thus, the inevitable death of PhysX could be postponed quite a bit if both AMD and Nvidia are supportive of it.

Of course, any of that is unlikely to happen, with Nvidia's latest move of excluding PhysX on any machine that detects non-Nvidia hardware (what a dick thing to do). I mean, if someone asks for help in an ATI/Nvidia situation, Nvidia is fully entitled to tell that person they do not support such a configuration and leave it at that. But blocking it all together says something about Nvidia's attitude.
DirectX 11 doesn't have built-in physics! Direct Compute != physics engine.
Direct Computer is MS's alternative to OpenCL, Stream & CUDA.
Posted on Reply
#10
devguy
by: MrMilli
DirectX 11 doesn't have built-in physics! Direct Compute != physics engine.
Direct Compute is MS's alternative to OpenCL, Stream & CUDA.
Right, I meant programming a physics API on on DirectX Compute.
Posted on Reply
#11
tkpenalty
iirc unlike AMD's recent offerings none of their GPUs com equipped with error correction algos, making it rather useless for many tasks. Doesnt seem like AMD and most of TPU can read between the lines as to what nvidia's doing. Nvidia is just misguiding shareholders to invest in them instead of AMD.
Posted on Reply
#12
Easy Rhino
Linux Advocate
by: tkpenalty
iirc unlike AMD's recent offerings none of their GPUs com equipped with error correction algos, making it rather useless for many tasks. Doesnt seem like AMD and most of TPU can read between the lines as to what nvidia's doing. Nvidia is just misguiding shareholders to invest in them instead of AMD.
sort of like when AMD put out the rumor that NVIDIA's new 40nm chips were experiencing epic fail rates...
Posted on Reply
#13
imperialreign
by: tkpenalty
iirc unlike AMD's recent offerings none of their GPUs com equipped with error correction algos, making it rather useless for many tasks. Doesnt seem like AMD and most of TPU can read between the lines as to what nvidia's doing. Nvidia is just misguiding shareholders to invest in them instead of AMD.
That's been such an on-going story for the last 10-years now . . . I don't even pay much attention to it anymore :p


by: Easy Rhino
sort of like when AMD put out the rumor that NVIDIA's new 40nm chips were experiencing epic fail rates...
ain't it funny how this industry rolls? :D
Posted on Reply
#14
Easy Rhino
Linux Advocate
by: imperialreign
That's been such an on-going story for the last 10-years now . . . I don't even pay much attention to it anymore :p





ain't it funny how this industry rolls? :D
exactly which is why i dont think we should take sides. there is so much going on behind the scenes between these two companies becuase they are worth billions of dollars and thousands of millionaire investors have stake in the companies. who are we? consumers! we drives sale prices. we shouldnt take sides arbitrarily, we should instead buy the the product that best suits our need, regardless of brand or manufacturer!
Posted on Reply
#15
Mussels
Moderprator
by: devguy
Right, I meant programming a physics API on on DirectX Compute.
or, leave it to havok.
Posted on Reply
#16
ZoneDymo
by: Easy Rhino
sort of like when AMD put out the rumor that NVIDIA's new 40nm chips were experiencing epic fail rates...
umm a lot more sources report that.
Posted on Reply
#17
mdm-adph
by: Easy Rhino
sort of like when AMD put out the rumor that NVIDIA's new 40nm chips were experiencing epic fail rates...
Well, why else haven't you seen a lot of them around? Nvidia's the bigger company with more money -- one would think that they would've been the first to heavily market 40nm chips...
Posted on Reply
#18
Easy Rhino
Linux Advocate
by: ZoneDymo
umm a lot more sources report that.
yea the sources got the info from AMD
Posted on Reply
#19
Easy Rhino
Linux Advocate
by: mdm-adph
Well, why else haven't you seen a lot of them around? Nvidia's the bigger company with more money -- one would think that they would've been the first to heavily market 40nm chips...
you can create all the conspiracy theories you want, but nvidia has already said that rumor is bull.
Posted on Reply
#20
mdm-adph
by: Easy Rhino
you can create all the conspiracy theories you want, but nvidia has already said that rumor is bull.
...as if they would say anything else. That's hardly conclusive evidence. :wtf:
Posted on Reply
#21
Benetanegia
by: mdm-adph
...as if they would say anything else. That's hardly conclusive evidence. :wtf:
Man not trying to fight so take this with humour please, but you always look like involved in a conspiracy theory, if there's something you can doubt about, you do, doubting is your job. :toast:

On-topic, I thought they had released them 2 months ago or so. It could be they were beta and these are official?
Posted on Reply
Add your own comment