"Why won't ATI Support Cuda and PhysX?"

[Gikaseixas]

Originally posted by: Keysplayr
Wreckage
I think ATI cards just don't have the power to run PhysX (opinion)

ShawnD1
It's only a matter of time before you're banned for trolling. (attack)

DaveSimmons
Now that's a silly troll. (attack)

SlowSpyder
Troll. (attack)

Sorry guys, but out of these four posts, who do you think stands a better chance of getting a vacation? You're no better than you believe Wreckage is if you pull this crap. I'll kindly ask you to knock it off.

@Scali: Nice posts there. It does explain a lot about the differences in ATI and Nvidia hardware and their respective GPGPU performance.

so he thinks that ATI cards lack power (it's his opinion, not a fact and his opinion is ATI sucks no matter what), he is also the same person who derails most of the ATI's threads so think a little harder man. With what intentions did he say it? A guy with his forum background (clearly anti-ATI). i'm not surprised people here are starting to loose patience with this kind of negativity towards JUST ONE BRAND.

You should say something more constructive, something more in line with what Scali said, not just come in here telling them to knock it off. Oh wait, you are "focused" in defending Nvidia.

 

[Cookie Monster]

Im sorry keys, but that post made by wreckage was 100% flame bait. Opinion or not.

If someone credible said the same thing with some sort of logical explanation as to why he/she thinks so is acceptable but this is clearly not. We all like to have constructive discussions, share our knowledge and experience, crack jokes once in awhile but Wreckage has done nothing but bring fire and brimstone to the video forums. Ok this is gone OT enough, but maybe its time for someone in the higher ups to look at this issue or else it will be a never ending cycle of flame bait posts -> results flame war -> ends up with thread lock -> everyone very angry.

Back to topic. Isnt it pointless when comparing the peak theoretical FLOPs of the CPU to the GPU? the latter cant reach those numbers in the majority of the tasks where as a CPU can. However that being said, I think its safe to assume that the GPU does speed up physics by quite a margin compared to the CPU.

 

[Denithor]

Originally posted by: Scali
All this combined means that ATi cards indeed have some limitations in GPGPU compared to nVidia. This is also apparent in Folding@home for example.
Read this thread for example:
http://foldingforum.org/viewto...p?f=51&t=10442&start=0
It includes comments of people like mhouston, who work for AMD on the Folding@Home client. Basically they're saying that they calculate certain values multiple times because on ATi hardware this is faster than using the shared memory (LDS - Local Data Storage).

**We need that mouth-hanging-open smiley icon.**

Anyway, wowsers, thanks for finding & posting that tidbit, Scali. I've wondered for a year now why ATi cards sucked so much ass at crunching with the GPU2 client. I compiled an early list of card ppd results for the DC forum here on the forums so people would have an easy place to pick out a ppd/$ winner. The results, needless to say, did not favor ATi.

These days a GTX 260/216 can produce 7k ppd on many of the WUs. Under the same work load the 4890 can only manage around 4k ppd. But we never had a good explanation for why. I'm going to copy your post into the DC forum, feel free to come post in there if you have anything else to add.

 

[Keysplayr]

Originally posted by: Gikaseixas

Originally posted by: Keysplayr
Wreckage
I think ATI cards just don't have the power to run PhysX (opinion)

ShawnD1
It's only a matter of time before you're banned for trolling. (attack)

DaveSimmons
Now that's a silly troll. (attack)

SlowSpyder
Troll. (attack)

Sorry guys, but out of these four posts, who do you think stands a better chance of getting a vacation? You're no better than you believe Wreckage is if you pull this crap. I'll kindly ask you to knock it off.

@Scali: Nice posts there. It does explain a lot about the differences in ATI and Nvidia hardware and their respective GPGPU performance.

so he thinks that ATI cards lack power (it's his opinion, not a fact and his opinion is ATI sucks no matter what), he is also the same person who derails most of the ATI's threads so think a little harder man. With what intentions did he say it? A guy with his forum background (clearly anti-ATI). i'm not surprised people here are starting to loose patience with this kind of negativity towards JUST ONE BRAND.

You should say something more constructive, something more in line with what Scali said, not just come in here telling them to knock it off. Oh wait, you are "focused" in defending Nvidia.

Excuse me? You're endorsing what they said then? Hey, if you can't do it, why not just say so. I have repeatedly stated that you could attack a persons ARGUMENT, not attack the poster for his/her opinions. What isn't clear about this? It's actually in the sticky at the top of this forum. Does nobody read it or heed it?

I think it's time for a cleanup. And if you'd bother to notice, I'm defending a poster, not Nvidia.

 

[Keysplayr]

Originally posted by: Cookie Monster
Im sorry keys, but that post made by wreckage was 100% flame bait. Opinion or not.

If someone credible said the same thing with some sort of logical explanation as to why he/she thinks so is acceptable but this is clearly not. We all like to have constructive discussions, share our knowledge and experience, crack jokes once in awhile but Wreckage has done nothing but bring fire and brimstone to the video forums. Ok this is gone OT enough, but maybe its time for someone in the higher ups to look at this issue or else it will be a never ending cycle of flame bait posts -> results flame war -> ends up with thread lock -> everyone very angry.

Back to topic. Isnt it pointless when comparing the peak theoretical FLOPs of the CPU to the GPU? the latter cant reach those numbers in the majority of the tasks where as a CPU can. However that being said, I think its safe to assume that the GPU does speed up physics by quite a margin compared to the CPU.

You can call it flamebait, or whatever you wish. But when it comes down to it, the poster posted an opinion that he believes. Because of WHO the poster is, is how the message is perceived. Instead of attempting to disprove the posters theory/opinion, he is attacked.
You know this isn't cool, so please don't try to defend it. No matter who the poster is or what you think of them for having opinions.

 

[Scali]

Originally posted by: Keysplayr
@Scali: Nice posts there. It does explain a lot about the differences in ATI and Nvidia hardware and their respective GPGPU performance.

So you're basically saying, for GPGPU purposes, a 4800 series GPU would only have 160 sp's that could read and write to that 128bytes of cache each, while Nvidia's GT200 series has 192/216/240 sp's that could all read and write to 512bytes of cache each.

So if ATI doubled their shaders to say 1600 in the "R870" and increased the cache size to 512bytes per block, they would be much more competitive having 320 read/write sp's.

Am I getting this right?

Well, as I understood it (I haven't actually used a Radeon 4000 for GPGPU myself), it's a difference in concept.
On the Radeon, each thread gets a private bit of data, which it can write to, which other threads can see. So in a way you 'broadcast' that data to the other threads.
That has two limitations:
1) You can't divide the storage among the threads the way you see fit, it's always evenly distributed, so even if you only need a few bytes for one thread, the rest of the memory cannot be used by another thread.
2) You can't have multiple threads modify data easily. They could read from one thread, then 'broadcast' it back from their own shared space. But that means the other threads need to monitor the data in multiple places.

On nVidia it's more like a piece of superfast local memory. All threads in a block can read and write in the same memory space as they see fit.
That's why on nVidia you can just calc these molecule distances once and store them in the shared memory for all threads to use, while on ATi it's faster to have everyone calculate them themselves, rather than to shared them through the LDS.

 

[tommo123]

from what i read before, isn't it a case of ATi cards having a fair bit of power - flop wise, at least from looking at max theoretical posibilities of ATi cards - but it's a PITA to be able to use it? kind of like writing for the 360 vs the PS3s cell?

 

[Scali]

Originally posted by: Cookie Monster
Back to topic. Isnt it pointless when comparing the peak theoretical FLOPs of the CPU to the GPU? the latter cant reach those numbers in the majority of the tasks where as a CPU can. However that being said, I think its safe to assume that the GPU does speed up physics by quite a margin compared to the CPU.

Certainly, FLOPS only go so far. Even between GPUs we see the same thing. On paper, the HD4890 has the highest TFLOP count. In practice the GTX285 is usually the faster GPU.

You also have to look at it from an application-to-application basis.
Obviously graphics are a good case for GPUs, that's what they were designed to do. They will get to put their incredible floating point power to good use, and CPUs don't even get close.

In the case of physics, GPUs may not be as efficient as with graphics, but still manage to stay well ahead of CPUs easily. It's a combination of the huge advantage in theoretical processing power and the fact that physics map to the current GPGPU architecture quite well.

 

[Keysplayr]

Originally posted by: tommo123
from what i read before, isn't it a case of ATi cards having a fair bit of power - flop wise, at least from looking at max theoretical posibilities of ATi cards - but it's a PITA to be able to use it? kind of like writing for the 360 vs the PS3s cell?

There were quite extensive discussions on this. The theoretical power is there in ATI 4xxx series, but it would appear nobody want's to code for maximum potential.

In theory, ATI 4xxx is a power house with 800 sp's and mega FLOPs.
In practice, not so much.

 

[SlowSpyder]

Originally posted by: Keysplayr
Wreckage
I think ATI cards just don't have the power to run PhysX (opinion)

ShawnD1
It's only a matter of time before you're banned for trolling. (attack)

DaveSimmons
Now that's a silly troll. (attack)

SlowSpyder
Troll. (attack)

Sorry guys, but out of these four posts, who do you think stands a better chance of getting a vacation? You're no better than you believe Wreckage is if you pull this crap. I'll kindly ask you to knock it off.

@Scali: Nice posts there. It does explain a lot about the differences in ATI and Nvidia hardware and their respective GPGPU performance.

So you're basically saying, for GPGPU purposes, a 4800 series GPU would only have 160 sp's that could read and write to that 128bytes of cache each, while Nvidia's GT200 series has 192/216/240 sp's that could all read and write to 512bytes of cache each.

So if ATI doubled their shaders to say 1600 in the "R870" and increased the cache size to 512bytes per block, they would be much more competitive having 320 read/write sp's.

Am I getting this right?

Had he said that AMD GPU's lack the functionality/power to run Physx then explained a coherant reason why, that'd be one thing. But here he is just doing what he typically does... try to bait and derail anything that doesn't blow sunshine up Nvidia's keister. How about if I went into a "Why Nvidia chose not to implement DX10.1" thread and just stated that their cards don't have the power to run DX10.1 and left it at that?

I think Nvidia needs AMD to help get Physx the traction it needs to make sure it has staying power, so it's not just a page on some website that looks back at the beginnings of hardware physics, personally. With both on board I would think that Physx would really take off, as it is now I think it has a 50/50 chance of making it longer term.

 

[Keysplayr]

Originally posted by: SlowSpyder

Originally posted by: Keysplayr
Wreckage
I think ATI cards just don't have the power to run PhysX (opinion)

ShawnD1
It's only a matter of time before you're banned for trolling. (attack)

DaveSimmons
Now that's a silly troll. (attack)

SlowSpyder
Troll. (attack)

Sorry guys, but out of these four posts, who do you think stands a better chance of getting a vacation? You're no better than you believe Wreckage is if you pull this crap. I'll kindly ask you to knock it off.

@Scali: Nice posts there. It does explain a lot about the differences in ATI and Nvidia hardware and their respective GPGPU performance.

So you're basically saying, for GPGPU purposes, a 4800 series GPU would only have 160 sp's that could read and write to that 128bytes of cache each, while Nvidia's GT200 series has 192/216/240 sp's that could all read and write to 512bytes of cache each.

So if ATI doubled their shaders to say 1600 in the "R870" and increased the cache size to 512bytes per block, they would be much more competitive having 320 read/write sp's.

Am I getting this right?

Had he said that AMD GPU's lack the functionality/power to run Physx then explained a coherant reason why, that'd be one thing. But here he is just doing what he typically does... try to bait and derail anything that doesn't blow sunshine up Nvidia's keister. How about if I went into a "Why Nvidia chose not to implement DX10.1" thread and just stated that their cards don't have the power to run DX10.1 and left it at that?

I think Nvidia needs AMD to help get Physx the traction it needs to make sure it has staying power, so it's not just a page on some website that looks back at the beginnings of hardware physics, personally. With both on board I would think that Physx would really take off, as it is now I think it has a 50/50 chance of making it longer term.

You can absolutely do that. In fact, I encourage it. That's what these forums are for dude. Discussions about Graphics hardware. People should then try to debate on whether your statements are true or unfounded, just like what should have happened here without attacking the poster. So, what are you waiting for? Go ahead. :thumbsup:

 

[Creig]

Originally posted by: Keysplayr
You can call it flamebait, or whatever you wish. But when it comes down to it, the poster posted an opinion that he believes. Because of WHO the poster is, is how the message is perceived. Instead of attempting to disprove the posters theory/opinion, he is attacked.
You know this isn't cool, so please don't try to defend it. No matter who the poster is or what you think of them for having opinions.

Just because somebody puts the words "I think" in front of their statement, that doesn't automatically exclude it from being deliberate flame bait.

Taking his posting history into account puts his statement in a very clear light.

 

[SlowSpyder]

Originally posted by: Keysplayr

Originally posted by: SlowSpyder

Originally posted by: Keysplayr
Wreckage
I think ATI cards just don't have the power to run PhysX (opinion)

ShawnD1
It's only a matter of time before you're banned for trolling. (attack)

DaveSimmons
Now that's a silly troll. (attack)

SlowSpyder
Troll. (attack)

Sorry guys, but out of these four posts, who do you think stands a better chance of getting a vacation? You're no better than you believe Wreckage is if you pull this crap. I'll kindly ask you to knock it off.

@Scali: Nice posts there. It does explain a lot about the differences in ATI and Nvidia hardware and their respective GPGPU performance.

So you're basically saying, for GPGPU purposes, a 4800 series GPU would only have 160 sp's that could read and write to that 128bytes of cache each, while Nvidia's GT200 series has 192/216/240 sp's that could all read and write to 512bytes of cache each.

So if ATI doubled their shaders to say 1600 in the "R870" and increased the cache size to 512bytes per block, they would be much more competitive having 320 read/write sp's.

Am I getting this right?

Had he said that AMD GPU's lack the functionality/power to run Physx then explained a coherant reason why, that'd be one thing. But here he is just doing what he typically does... try to bait and derail anything that doesn't blow sunshine up Nvidia's keister. How about if I went into a "Why Nvidia chose not to implement DX10.1" thread and just stated that their cards don't have the power to run DX10.1 and left it at that?

I think Nvidia needs AMD to help get Physx the traction it needs to make sure it has staying power, so it's not just a page on some website that looks back at the beginnings of hardware physics, personally. With both on board I would think that Physx would really take off, as it is now I think it has a 50/50 chance of making it longer term.

You can absolutely do that. In fact, I encourage it. That's what these forums are for dude. Discussions about Graphics hardware. People should then try to debate on whether your statements are true or unfounded, just like should have happened here. So, what are you waiting for? Go ahead.

Well, why would I post that when I don't believe it? If I did post something like that I would think that you should back it up, but I guess not? By this logic I can go into an the CPU section and post in a thread something like, "Intel processors are garbage." and leave it at that. You'd go for that as the mod?

 

[Scali]

Originally posted by: SlowSpyder
Well, why would I post that when I don't believe it? If I did post something like that I would think that you should back it up, but I guess not? By this logic I can go into an the CPU section and post in a thread something like, "Intel processors are garbage." and leave it at that. You'd go for that as the mod?

Well, I don't think he's actually encouraging that part, but fact remains that such posts sometimes DO happen.
I think his focus was mainly on how people should respond to it when such a post is encountered. Which would be to challenge the poster into arguing why he thinks processors are garbage. An interesting, meaningful and perhaps even educational discussion might ensue.
Or he has no reasons, so it just remains quiet, and that's the end of it.

 

[Keysplayr]

Originally posted by: SlowSpyder

Originally posted by: Keysplayr

Originally posted by: SlowSpyder

Originally posted by: Keysplayr
Wreckage
I think ATI cards just don't have the power to run PhysX (opinion)

ShawnD1
It's only a matter of time before you're banned for trolling. (attack)

DaveSimmons
Now that's a silly troll. (attack)

SlowSpyder
Troll. (attack)

Sorry guys, but out of these four posts, who do you think stands a better chance of getting a vacation? You're no better than you believe Wreckage is if you pull this crap. I'll kindly ask you to knock it off.

@Scali: Nice posts there. It does explain a lot about the differences in ATI and Nvidia hardware and their respective GPGPU performance.

So you're basically saying, for GPGPU purposes, a 4800 series GPU would only have 160 sp's that could read and write to that 128bytes of cache each, while Nvidia's GT200 series has 192/216/240 sp's that could all read and write to 512bytes of cache each.

So if ATI doubled their shaders to say 1600 in the "R870" and increased the cache size to 512bytes per block, they would be much more competitive having 320 read/write sp's.

Am I getting this right?

Had he said that AMD GPU's lack the functionality/power to run Physx then explained a coherant reason why, that'd be one thing. But here he is just doing what he typically does... try to bait and derail anything that doesn't blow sunshine up Nvidia's keister. How about if I went into a "Why Nvidia chose not to implement DX10.1" thread and just stated that their cards don't have the power to run DX10.1 and left it at that?

I think Nvidia needs AMD to help get Physx the traction it needs to make sure it has staying power, so it's not just a page on some website that looks back at the beginnings of hardware physics, personally. With both on board I would think that Physx would really take off, as it is now I think it has a 50/50 chance of making it longer term.

You can absolutely do that. In fact, I encourage it. That's what these forums are for dude. Discussions about Graphics hardware. People should then try to debate on whether your statements are true or unfounded, just like should have happened here. So, what are you waiting for? Go ahead.

Well, why would I post that when I don't believe it? If I did post something like that I would think that you should back it up, but I guess not? By this logic I can go into an the CPU section and post in a thread something like, "Intel processors are garbage." and leave it at that. You'd go for that as the mod?

You gave the example, I thought you believed it when you said it. And by the way, Wreckage didn't post, "ATI GPU's are garbage." did he? You're embellishing on your own.
How many more examples are you going to give that you don't believe in? Obviously Wreckage believes ATI hardware cannot run PhysX. I don't agree with this at all. I will say that I do not think ATI hardware can run PhysX anywhere near as well as Nvidia hardware, which I think would be more close to factual that what Wreckage posted.

My point is, as Scali has correctly mentioned, is if you believe a posters statement to be false, even if it's just an opionion, shoot him down with your argument about the subject matter, don't shoot the poster for his opinion. Do you understand what I mean?
You would accomplish a lot more than just to shout "troll". It's easier than arguing a point, but it has no usefullness whatsoever.

 

[Gikaseixas]

Originally posted by: Keysplayr

Originally posted by: Gikaseixas

Originally posted by: Keysplayr
Wreckage
I think ATI cards just don't have the power to run PhysX (opinion)

ShawnD1
It's only a matter of time before you're banned for trolling. (attack)

DaveSimmons
Now that's a silly troll. (attack)

SlowSpyder
Troll. (attack)

Sorry guys, but out of these four posts, who do you think stands a better chance of getting a vacation? You're no better than you believe Wreckage is if you pull this crap. I'll kindly ask you to knock it off.

@Scali: Nice posts there. It does explain a lot about the differences in ATI and Nvidia hardware and their respective GPGPU performance.

so he thinks that ATI cards lack power (it's his opinion, not a fact and his opinion is ATI sucks no matter what), he is also the same person who derails most of the ATI's threads so think a little harder man. With what intentions did he say it? A guy with his forum background (clearly anti-ATI). i'm not surprised people here are starting to loose patience with this kind of negativity towards JUST ONE BRAND.

You should say something more constructive, something more in line with what Scali said, not just come in here telling them to knock it off. Oh wait, you are "focused" in defending Nvidia.

Excuse me? You're endorsing what they said then? Hey, if you can't do it, why not just say so. I have repeatedly stated that you could attack a persons ARGUMENT, not attack the poster for his/her opinions. What isn't clear about this? It's actually in the sticky at the top of this forum. Does nobody read it or heed it?

I think it's time for a cleanup. And if you'd bother to notice, I'm defending a poster, not Nvidia.

keep reading Keys, you'll see that i'm not the only one with this point of view. I'm not endorsing anybody, i'm just saying that i understand them as to why they're saying this.

Yes it is time for a clean up, starting with people who cannot contribute to healthy discussions.

 

[ShawnD1]

from what i read before, isn't it a case of ATi cards having a fair bit of power - flop wise, at least from looking at max theoretical posibilities of ATi cards - but it's a PITA to be able to use it? kind of like writing for the 360 vs the PS3s cell?

Maybe, but isn't flops one of those nearly useless measurements when trying to compare different architectures?

 

[Wreckage]

Originally posted by: Scali

All this combined means that ATi cards indeed have some limitations in GPGPU compared to nVidia. This is also apparent in Folding@home for example.
Read this thread for example:
http://foldingforum.org/viewto...p?f=51&t=10442&start=0
It includes comments of people like mhouston, who work for AMD on the Folding@Home client. Basically they're saying that they calculate certain values multiple times because on ATi hardware this is faster than using the shared memory (LDS - Local Data Storage).

Exactly, also if you notice the ATI video transcoder (AVIVO) relies heavily on the CPU.

So I stand by what I said about ATI cards lacking the power to run PhysX.

Until we actually see PhysX running on an ATI GPU and can benchmark it. I don't see much evidence to the contrary.

 

[Keysplayr]

This is becoming quite rediculous Shawn. Did Wreckage say anything like that?
Did anybody? If you equate what he said with what you just said, there are apparently some issues that need to be worked through. I just can't believe you took his comment about ATI in such a personal way

 

[Keysplayr]

Originally posted by: Wreckage

Originally posted by: Scali

All this combined means that ATi cards indeed have some limitations in GPGPU compared to nVidia. This is also apparent in Folding@home for example.
Read this thread for example:
http://foldingforum.org/viewto...p?f=51&t=10442&start=0
It includes comments of people like mhouston, who work for AMD on the Folding@Home client. Basically they're saying that they calculate certain values multiple times because on ATi hardware this is faster than using the shared memory (LDS - Local Data Storage).

Exactly, also if you notice the ATI video transcoder (AVIVO) relies heavily on the CPU.

So I stand by what I said about ATI cards lacking the power to run PhysX.

Until we actually see PhysX running on an ATI GPU and can benchmark it. I don't see much evidence to the contrary.

If a 3xxx series could be made to run physx on the GPU, I have little trouble believing that a 4xxx series GPU can run it too. I'm pretty certain it could, just likely not as well as an Nvidia GPU.

 

[SSChevy2001]

Originally posted by: Wreckage
Exactly, also if you notice the ATI video transcoder (AVIVO) relies heavily on the CPU.

So I stand by what I said about ATI cards lacking the power to run PhysX.

Until we actually see PhysX running on an ATI GPU and can benchmark it. I don't see much evidence to the contrary.

Actually I've seen just the opposite. In a case where a slower CPU is used ATi Stream actually runs faster than CUDA while using less CPU usage.

http://www.legitreviews.com/article/978/2/

The results were surprising to say the least. It seems that with NVIDIA CUDA hardware acceleration enabled the CPU load was still at 100% load during the transcoding of the video. When using the ATI card with Stream technology the CPU load for the same video was just 73%. Since the ATI Radeon HD 4770 was faster in the benchmarks it goes to show more work is being off loaded and done on the GPU than the CPU. It seems that on the NVIDIA solution more of the work is being dumped to the processor and the dual-core AMD Athlon X2 4850e isn't that quick.

So I disagree with you ATi card is more than able to run PhysX, if it had Nvidia complete support.

 

[SlowSpyder]

Originally posted by: SSChevy2001

Originally posted by: Wreckage
Exactly, also if you notice the ATI video transcoder (AVIVO) relies heavily on the CPU.

So I stand by what I said about ATI cards lacking the power to run PhysX.

Until we actually see PhysX running on an ATI GPU and can benchmark it. I don't see much evidence to the contrary.

Actually I've seen just the opposite. In a case where a slower CPU is used ATi Stream actually runs faster than CUDA while using less CPU usage.

http://www.legitreviews.com/article/978/2/

The results were surprising to say the least. It seems that with NVIDIA CUDA hardware acceleration enabled the CPU load was still at 100% load during the transcoding of the video. When using the ATI card with Stream technology the CPU load for the same video was just 73%. Since the ATI Radeon HD 4770 was faster in the benchmarks it goes to show more work is being off loaded and done on the GPU than the CPU. It seems that on the NVIDIA solution more of the work is being dumped to the processor and the dual-core AMD Athlon X2 4850e isn't that quick.

So I disagree with you ATi card is more than able PhysX, if it had Nvidia complete support.

It appears that Nvidia GPU's are not powerful enough to accerlerate transcoding of video.*


*Trying to create a debate on true statements vs. unfounded statements.

 

[Keysplayr]

You get an "Atta-boy!!"
Start another thread though.

 

[Wreckage]

Originally posted by: SSChevy2001

So I disagree with you ATi card is more than able PhysX, if it had Nvidia complete support.

You totally ignored AVIVO, I noticed.

http://www.pcper.com/article.p...=647&type=expert&pid=3

You can clearly see that while the ATI software is faster, it does require quite a bit more CPU power than the NVIDIA application.

 

[Atechie]

Originally posted by: Wreckage

Originally posted by: SSChevy2001

So I disagree with you ATi card is more than able PhysX, if it had Nvidia complete support.

You totally ignored AVIVO, I noticed.

http://www.pcper.com/article.p...=647&type=expert&pid=3

You can clearly see that while the ATI software is faster, it does require quite a bit more CPU power than the NVIDIA application.

As far as I remember AVIVO also have crap quality and lovely encoding errors.