**thread name change* Nvidia and AMD moral and immoral business practices

Page 20 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

Mem

Lifer
Apr 23, 2000
21,476
13
81
AMD went with Havok/Intel, that OTHER direct competitor of AMD.

Point is Nvidia is more of a threat to AMD in the graphics gaming world then Intel(everybody knows you don't help your no.1 competition),end of the day you can't blame them,a lot of companies do this,I bet if things were reversed Nvidia would be doing the same thing.

As to AMD going with Havok well thats their call and you have to respect that.
 
Last edited:

Scali

Banned
Dec 3, 2004
2,495
0
0
Point is Nvidia is more of a threat to AMD in the graphics gaming world then Intel(everybody knows you don't help your no.1 competition),end of the day you can't blame them,a lot of companies do this,I bet if things were reversed Nvidia would be doing the same thing.

As to AMD going with Havok well thats their call and you have to respect that.

I think your entire argument fell apart completely.

Also, I don't agree that nVidia is more of a threat.
Currently AMD's weak point for gaming is its CPU line. Even their fastest CPUs aren't that great... In case you didn't notice... a few years ago they promoted their Spider platform, go for an AMD CPU + AMD GPU.
With the introduction of the 5000-series, there was no talk about platforms at all, AMD was completely silent about using AMD CPUs. The reason is simple: Even AMD's fastest CPUs will limit their high-end cards somewhat. They wouldn't look as good in reviews as they would with Intel CPUs.

And Intel is currently busy stealing AMD's thunder with Fusion. Intel may be doing to AMD exactly what AMD was trying to do to nVidia: Take over the entire low-end market.

Unlike nVidia, Intel also has no interest in GPU accelerated physics at all, and as such they've basically held AMD hostage with Havok.
 

Mem

Lifer
Apr 23, 2000
21,476
13
81
I think your entire argument fell apart completely.

Also, I don't agree that nVidia is more of a threat.
Currently AMD's weak point for gaming is its CPU line. Even their fastest CPUs aren't that great... In case you didn't notice... a few years ago they promoted their Spider platform, go for an AMD CPU + AMD GPU.
With the introduction of the 5000-series, there was no talk about platforms at all, AMD was completely silent about using AMD CPUs. The reason is simple: Even AMD's fastest CPUs will limit their high-end cards somewhat. They wouldn't look as good in reviews as they would with Intel CPUs.

And Intel is currently busy stealing AMD's thunder with Fusion. Intel may be doing to AMD exactly what AMD was trying to do to nVidia: Take over the entire low-end market.

Unlike nVidia, Intel also has no interest in GPU accelerated physics at all, and as such they've basically held AMD hostage with Havok.

Sure they are ,gamers don't buy Intel gaming cards but either AMD or Nvidia gaming cards ,thats my point that Nvidia are more of a threat in this area.

As to AMD CPUs which is slightly off topic they are still more then good enough for gaming and good value.

Companies will always look after their own interests first,you can't expect them to jump into bed with Nvidia,same goes for Nvidia they will look after their own interests first.
 
Last edited:

Scali

Banned
Dec 3, 2004
2,495
0
0
Sure they are ,gamers don't buy Intel gaming cards but either AMD or Nvidia gaming cards ,thats my point that Nvidia are more of a threat in this area.

As to AMD CPUs which is slightly off topic they are still more then good enough for gaming and good value.

I think you need to see CPUs and GPUs in perspective.
AMD's CPU division is much larger than their GPU division, and a whole lot more money goes round in the CPU market.
No matter how well their GPU division does, they cannot sustain the CPU division.
And the GPU division is doing well, while the CPU division has been struggling for years.
Hence, Intel is a much bigger thread than nVidia.
 

Mem

Lifer
Apr 23, 2000
21,476
13
81
I think you need to see CPUs and GPUs in perspective.
AMD's CPU division is much larger than their GPU division, and a whole lot more money goes round in the CPU market.
No matter how well their GPU division does, they cannot sustain the CPU division.
And the GPU division is doing well, while the CPU division has been struggling for years.
Hence, Intel is a much bigger thread than nVidia.

It's all a matter of perspective,Nvidia have their competitors so does Intel,AMD etc....the point I was making is AMD and any other company will always look after their own interests first,you can't expect AMD to follow Nvidia's way of thinking or support their products ie PhysX.
 

SirPauly

Diamond Member
Apr 28, 2009
5,187
1
0
The reason why AMD may of gone for HaVok is because Intel allowed it to be ported to OpenCL and AMD did virtually all the work, from my understanding. This is all speculative on my part but do believe there were some conversations with nVidia and AMD -- and probably desired to do the same with PhysX, meaning, for nVidia to allow PhysX to be ported to OpenCL and AMD will do all/most/some of the work. nVidia may of said, no!

There is no way that I can see AMD licensing Cuda -- the above speculation is the only logical explanation to me.
 

Scali

Banned
Dec 3, 2004
2,495
0
0
The reason why AMD may of gone for HaVok is because Intel allowed it to be ported to OpenCL and AMD did virtually all the work, from my understanding. This is all speculative on my part but do believe there were some conversations with nVidia and AMD -- and probably desired to do the same with PhysX, meaning, for nVidia to allow PhysX to be ported to OpenCL and AMD will do all/most/some of the work. nVidia may of said, no!

There is no way that I can see AMD licensing Cuda -- the above speculation is the only logical explanation to me.

Well, I don't see why nVidia wouldn't allow OpenCL, as they support it better than AMD.
Aside from that, I don't see why AMD would even prefer OpenCL over their own Stream.
I doubt that nVidia meant that AMD needed to license Cuda, because that just makes no sense. Cuda is nVidia's GPGPU architecture. Clearly competitors have their own architectures to program for. I don't see why nVidia would even WANT to license their architecture to competitors. It is their strategic advantage (just like how Intel never wanted to license x86 to competitors).
The most logical explanation is that nVidia wanted to have AMD/other GPU vendors develop their own Stream/OpenCL/whatever backend for PhysX.
 

SirPauly

Diamond Member
Apr 28, 2009
5,187
1
0
It's all a matter of perspective,Nvidia have their competitors so does Intel,AMD etc....the point I was making is AMD and any other company will always look after their own interests first,you can't expect AMD to follow Nvidia's way of thinking or support their products ie PhysX.

That's why politics or even grey territories don't bother me as much as others because it is about what is best for the company, and sometimes their tougher decisions may effect the feelings of some consumers, sadly. It's really great to see passionate posters even when their critical to me on all sides because it raises different points-of-view, offers spirited debate and discourse, which many mind-sets differ; it's all good.
 

Scali

Banned
Dec 3, 2004
2,495
0
0
It's all a matter of perspective,Nvidia have their competitors so does Intel,AMD etc....the point I was making is AMD and any other company will always look after their own interests first,you can't expect AMD to follow Nvidia's way of thinking or support their products ie PhysX.

But that logic is flawed... because although AMD may not have followed nVidia, they DID follow Intel. Which was not in their interest either.
And the result? Well, nothing really. AMD has nothing to show for its 'partnership' with Havok/Intel.
Don't you see your own flawed logic? You only look as far as nVidia, and completely ignore that the situation with Intel/Havok is exactly the same... or actually even worse, as unlike nVidia, Intel has no benefit from GPU-accelerated physics at all.
 

SirPauly

Diamond Member
Apr 28, 2009
5,187
1
0
Well, I don't see why nVidia wouldn't allow OpenCL, as they support it better than AMD.
Aside from that, I don't see why AMD would even prefer OpenCL over their own Stream.
I doubt that nVidia meant that AMD needed to license Cuda, because that just makes no sense. Cuda is nVidia's GPGPU architecture. Clearly competitors have their own architectures to program for. I don't see why nVidia would even WANT to license their architecture to competitors. It is their strategic advantage (just like how Intel never wanted to license x86 to competitors).
The most logical explanation is that nVidia wanted to have AMD/other GPU vendors develop their own Stream/OpenCL/whatever backend for PhysX.

So, you're saying that nVidia will allow PhysX without the need to license Cuda for AMD to offer PhysX?
 

SirPauly

Diamond Member
Apr 28, 2009
5,187
1
0
But that logic is flawed... because although AMD may not have followed nVidia, they DID follow Intel. Which was not in their interest either.
And the result? Well, nothing really. AMD has nothing to show for its 'partnership' with Havok/Intel.
Don't you see your own flawed logic? You only look as far as nVidia, and completely ignore that the situation with Intel/Havok is exactly the same... or actually even worse, as unlike nVidia, Intel has no benefit from GPU-accelerated physics at all.

Sure, there was logic, because they ported Havok to OpenCL, and was in their interest.
 

Scali

Banned
Dec 3, 2004
2,495
0
0
Sure, there was logic, because they ported Havok to OpenCL, and was in their interest.

That's not the logic I was talking about (I meant that the claim is that AMD shouldn't side with its competitor (nVidia), while that is EXACTLY what they did (Intel))... but okay, I'll humour you.
Firstly, you seem to assume that nVidia would not allow an OpenCL port... I'm not sure why you think that, since nVidia has been very supportive of OpenCL.

Secondly, did AMD port Havok to OpenCL? I don't think they did. I've never seen an official Havok release with OpenCL support at any rate. Let alone any games supporting OpenCL effects in Havok, or even games with GPU-accelerated Havok effects used to promote AMD products.

So how was this in AMD's interest? It wasn't. They have nothing. Which is why they now moved to Bullet.
 

NIGELG

Senior member
Nov 4, 2009
852
31
91
Sure, there was logic, because they ported Havok to OpenCL, and was in their interest.
Hey Pauly good to see you here.I think you know my stance on the whole physX issue from my posts at Rage3d.I wouLd love to see more advanced features in games including physics.But from what I've seen so far I don't think I an missing out by not having PhysX.

As for the whole morality issue...well let's just say that I just trust AMD cards more based on my own user experience.My perception whether rightly or wrongly of Nvidia is they have an element of sleazyness about them.That ,combined with my horrible 8800gts dying experience has made me not buy their products in the past two years.In contrast my 4850 and 5850 have satisfied me to the extreme..

But I will never hate a company and would always consider buying from both of them...
 

SirPauly

Diamond Member
Apr 28, 2009
5,187
1
0
What is needed is actual clarity instead of speculation. If nVidia desired a port for OpenCL at this time, nVidia would offer a port to OpenCL. I don't see them allowing their brand PhysX be ported to OpenCL before they are willing or don't have a differentiation advantage.
 

Kenmitch

Diamond Member
Oct 10, 1999
8,505
2,250
136
Why did AMD never take nVidia up on this offer? They never even bothered to talk to nVidia to find out of they could negotiate any reasonable license.

How do you know they never talked to nvidia? Maybe the terms and condiditons of the deal were unreasonable? Do you work for nvidia?
 

Scali

Banned
Dec 3, 2004
2,495
0
0
What is needed is actual clarity instead of speculation. If nVidia desired a port for OpenCL at this time, nVidia would offer a port to OpenCL. I don't see them allowing their brand PhysX be ported to OpenCL before they are willing or don't have a differentiation advantage.

I don't think you understand what I said.
Firstly, AMD is not equivalent to OpenCL. In fact, OpenCL was not much more than a draft at the time when nVidia acquired PhysX, and announced that they are open to any other GPU vendors to support it.
Why this focus on OpenCL? It could just as well have been AMD Stream. That doesn't matter.

Secondly, why should nVidia care about 'competition' from OpenCL? They support OpenCL as well as the next vendor... and they have Cuda as their differentiation advantage. They have their own native GPGPU architecture, which they can update and expand whenever they need. OpenCL is always going to be chasing Cuda in PhysX.
The obvious advantage would be that with AMD on board, it's pretty much a given that many more developers will adopt PhysX than what is currently the case. This would give PhysX more leverage. nVidia doesn't have to be the ONLY vendor offering GPU-accelerated physics... in fact, it may be better if all vendors offer it, but nVidia is simply the best one (much like how generally all GPUs support the same DirectX/OpenGL/OpenCL features, and performance/IQ are what make the difference).
In fact, nVidia also supports the Bullet project (with DirectCompute/OpenCL). Erwin has even been more positive about their support than about AMD's, in public. nVidia just doesn't use Bullet as marketing leverage the way AMD does.

Thirdly, as I already said before, I wouldn't be surprised if nVidia already HAD an OpenCL port, or at least has done enough preparation to release on on short notice... should their hand be forced, that is. Until there is any reasonable threat from competing physics middleware, there is absolutely no pressure on nVidia to open up PhysX for all OpenCL-supporting vendors (without them having done any work at all, or even obtaining a license... which would be different from the situation concerning nVidia's original offer to be open to other GPU vendors).

Lastly, why are you ignoring the fact that Havok didn't bring AMD anything either? Why only these nVidia-centric views?
Fact is that partnering with nVidia on PhysX could not possibly be any worse than AMD's partnership with Havok/Intel, since that partnership has delivered exactly 0.
Let us discuss this first, before some random speculation about OpenCL (which is NOT AMD's thing, heck they don't even support it yet officially).
 
Last edited:

Scali

Banned
Dec 3, 2004
2,495
0
0
How do you know they never talked to nvidia? Maybe the terms and condiditons of the deal were unreasonable? Do you work for nvidia?

I know because they publicly discarded the option, then shortly afterwards announced their partnership with Havok/Intel.
We'll never know what the terms and conditions are, because AMD did not even start any negotiations with nVidia, hence no terms and conditions were ever drafted up.
And no, I don't work for nVidia. Why do you ask?

Here you go:
http://www.bit-tech.net/news/hardware/2008/12/11/amd-exec-says-physx-will-die/1
Godfrey Cheng, Director of Technical Marketing in AMD's Graphics Product Group, has said that PhysX will die if it remains a closed and proprietary standard

"There is no plan for closed and proprietary standards like PhysX," said Cheng. "As we have emphasised with our support for OpenCL and DX11, closed and proprietary standards will die.".

(Ofcourse Havok is just as closed and proprietary, but hey)
 
Last edited:

railven

Diamond Member
Mar 25, 2010
6,604
561
126
I know because they publicly discarded the option, then shortly afterwards announced their partnership with Havok/Intel.
We'll never know what the terms and conditions are, because AMD did not even start any negotiations with nVidia, hence no terms and conditions were ever drafted up.
And no, I don't work for nVidia. Why do you ask?

Here you go:
http://www.bit-tech.net/news/hardware/2008/12/11/amd-exec-says-physx-will-die/1


(Ofcourse Havok is just as closed and proprietary, but hey)

Interesting same guy I presume since it's the same article:

It should be noted that title support for GPU accelerated physics simulation is NOT the end game. The end game is having GPU physics as an integral part of game play and not just eye candy. If it is optional eye candy, GPU physics will not gain traction. The titles we have seen today with shattering glass and cloth waving in the wind is not integral to game play and the impact on the game's experience is minimal. We are looking for ways to integrate GPU physics better into game play. Or even things like AI instead of focusing on eye candy / effects physic.


Seems like the opinion many shared in that Mafia II physx thread.
 

Gikaseixas

Platinum Member
Jul 1, 2004
2,836
218
106
In contrast my 4850 and 5850 have satisfied me to the extreme..

But I will never hate a company and would always consider buying from both of them...

so is my 5870. it plays everything i throw at it and that's the number 1 reason when deciding which graphics card to buy...
 

Mem

Lifer
Apr 23, 2000
21,476
13
81
But that logic is flawed... because although AMD may not have followed nVidia, they DID follow Intel. Which was not in their interest either.
And the result? Well, nothing really. AMD has nothing to show for its 'partnership' with Havok/Intel.
Don't you see your own flawed logic? You only look as far as nVidia, and completely ignore that the situation with Intel/Havok is exactly the same... or actually even worse, as unlike nVidia, Intel has no benefit from GPU-accelerated physics at all.

Again you are missing the point,it does not matter what you or I think,AMD did what they thought was right,you can't say its flawed because we dont' know their long term goals or their reasoning behind their decisions.

You sound like you are just upset because AMD did not do what Nvidia wanted them to do.

No, you are the one looking at Nvidia and trying to blame AMD,you forget they are a huge company with their own agenda /interests,what you and I think on this subject does not matter to them or any other company.

Try being neutral and look at it from that angle then look at it from AMD as if you are the director or chairman ,then look at it from Nvidia's point,see how the goalposts shift?
 

Scali

Banned
Dec 3, 2004
2,495
0
0
You mean it wasn't from the majority of the few GPU PhysX accelerated titles that pretty much makes that statement and opinion true?

Not so fast...
Part of that statement is true.
Let's take it line-by-line:
It should be noted that title support for GPU accelerated physics simulation is NOT the end game.

Fair enough, not the end game... but it is a start nonetheless.

The end game is having GPU physics as an integral part of game play and not just eye candy.

Yes, but I suppose nVidia, the PhysX team and every game developer feels pretty much the same way. Bit of a non-statement.

If it is optional eye candy, GPU physics will not gain traction.

This is debatable.
On the one hand we have plenty of examples of games that were successful while riding mainly on their 'eye candy' hype.
On the other hand, we have plenty of examples of games that were successful simply because they had good gameplay... they were basically just good games.
If GPU physics is included in such a game, it will piggy-back on its success and gain traction that way.
Heck, the entire OpenGL API pretty much just piggy-backed on the success of GLQuake, and other major games which adopted the engine (eg Half-Life). It helped push Glide from the number 1 spot of 3D APIs.

The titles we have seen today with shattering glass and cloth waving in the wind is not integral to game play and the impact on the game's experience is minimal.

Yes, but see above... that is not necessarily a bad thing. More eye-candy is pretty much what the entire GPU market revolves around anyway. It's a bit hypocritical. I mean, if you compare the early Voodoo accelerators to today's DX11 cards, they're a truckload of features on the new cards, but hardly any of them have much of an effect on gameplay and experience... They just give you more eyecandy (realtime shadows, HDR effects, sharper texturing, antialiasing etc).
Since pretty much all games make use of these features, and nothing looks like early GLQuake or such anymore, I would say that eyecandy is generally a good thing, and AMD made a lot of money selling it.

We are looking for ways to integrate GPU physics better into game play. Or even things like AI instead of focusing on eye candy / effects physic.

Okay, you said that more than 2 years ago... Where is it?
Seems like you haven't even made a start yet, let alone that you are anywhere near your proposed 'end game'.
 

Scali

Banned
Dec 3, 2004
2,495
0
0
You sound like you are just upset because AMD did not do what Nvidia wanted them to do.

No, I'm just tired of arguing with people who don't seem to understand the basic facts in the whole GPGPU/physics affair, and as such can't make coherent arguments.

Let's get back to what YOU said:
Do you blame them?..Nvidia is AMD's direct competitor ,so you want them to pay Nvidia for the license so they can adopt Nvidia's way of doing things?..get real will you,they are there to compete with Nvidia and take some of the competition's profits away from Nvidia ,not to give them money.

AMD can go their own route and with their own engineers(thats what they get paid for ) and not to help Nvidia.

Now, let's substitute every instance of 'nVidia' with 'Intel':
Do you blame them?..Intel is AMD's direct competitor ,so you want them to pay Intel for the license so they can adopt Intel's way of doing things?..get real will you,they are there to compete with Intel and take some of the competition's profits away from Intel ,not to give them money.

AMD can go their own route and with their own engineers(thats what they get paid for ) and not to help Intel.

That still makes as much sense as nVidia, doesn't it?
Problem here is that reality disagrees with you: AMD did partner up with Intel.
There is your flawed logic.
So my obvious argument here is this:
If, in light of your arguments against teaming up with a competitor, AMD still decided to go with Intel... by that same logic, AMD could (and possibly should) also have gone with nVidia.
Heck, AMD was even fighting Intel in court at that time. At least they weren't entangled with any legal issues with nVidia.

Try being neutral and look at it from that angle then look at it from AMD as if you are the director or chairman ,then look at it from Nvidia's point,see how the goalposts shift?

As I said before, AMD just got cornered. They could go with nVidia, Intel or Sony. They went with Intel and Sony... might as well also gone with nVidia while they were at it. I think it's called "Don't bet all your money on one horse".
Your problem is that you constantly ignore Intel in the story (that's why I say you're being nVidia-centric). Ofcourse that's convenient for your flawed logic, but it doesn't quite match up with reality.
Try being neutral and look at it from that angle then look at it from AMD as if you are the director or chairman ,then look at it from Intel's point,see how the goalposts shift?
 
Last edited:

Mem

Lifer
Apr 23, 2000
21,476
13
81
Scali: Again you are missing the point,AMD can do what they want with their own company,I don't know why you are upset with what they have done,companies will do what they want,PhysX is overated IMHO but thats just my opinion,who cares if they teamed up with Intel or any other company?.....point is they and other companies do what they feel is right.


Forget all the legal fighting crap,one time you had Intel/AMD/Nvidia all involved trying to fight and sue each other in the law courts,so what?..it does not change anything.

Are you upset because AMD did not go with licensing with Nvidia for PhysX?

Remember like me you can only speculate an opinion on reason why,end of the day huge companies like AMD,Nvidia etc..will look after their own agenda and to be honest I can understand that regardless of what company in question.

Btw the reason why I ignored Intel is because of thread title ie it says AMD/Nvidia ,I don't want to branch too much off topic,however Intel is in same boat ie they will do what they feel is right,companies do abandon ideas or change direction etc in the company interests so what is the issue?
 
Last edited: