The official GT300/Fermi Thread

[To(V)bo Co(V)bo]

Yeah its pathetic whats going on. Im not a fanboy of either camp. I have a 9800gt, Im sorry 8800gt, whats the difference but I tend to favor Ati's buisiness ethics. At least they improve the industry for all consumers instead of the nvidian hypocrites. Nvidia, Were saving gameing by making pc games "special" and adding Physx. If they truley ment it, they would do whats best for us consumers and improve games for all platforms, even at a loss. Whats the saying "you recieve 10X what you give away" and I believe every word of it. In fact Id have more respect if they improved games at a loss, it isnt a loss if you recieve something in return, like sabotaging ati performance. If nvidia truley wants Physx to take the world by storm, then give it away and set it free, and open it up to everyone who really needs it. Just by doing this alone could be enough to make it adopted and used by everyone. They could make more money making it work better than anyone else, than being the only one with it, and then the money will come to them. Its really simple aint it?

If I had the cure for cancer what would I do?
a: Sell it by the bottle to be a millionaire
b: Give it away to save millions of lives

I would do b, but thats me, But I bet you I would make more than enough cash giving it away and saving lives.

Think if they marketed physx as folding for cures, and how they donated this technology to better help mankind. That would make me feel so fuzzy inside that Id want to buy a nvidia card! cuz were saving lives here and Nvidia can save more when you use their cards.

Instead of donating money to charity, just buy a nvidia gpu for 50$ or 100$ or more and help find the cures! anyone can help, see how much your helping day to day and make a difference. That sounds great if I do say myself!

[Cybercat]

If nvidia truley wants Physx to take the world by storm, then give it away and set it free, and open it up to everyone who really needs it. Just by doing this alone could be enough to make it adopted and used by everyone. Then they could make more money making it work better than anyone else, than being the only one with it, and then the money will come to them. Its really simple aint it?

Check this out:

http://developer.nvidia.com/object/physx_downloads.html

Literally anyone can download the SDK for free. So why hasn't it "taken the world by storm" as you said? It's proprietary. NVIDIA claims that PhysX, along with CUDA, is open to any hardware should one want to support it. The most likely reason holding AMD back from doing this is the unbalanced performance prospects of such an implementation. They probably don't feel like they need to anyway, since more open standards like Bullet physics will eventually arise to make use of OpenCL.

Of course if AMD was so confident in OpenCL, they might be more enthusiastic to get a working driver out for it, but as it is, even NVIDIA has beat them to the punch there. AMD makes great hardware but they really suck in the software/developer support side compared to the competition.

[To(V)bo Co(V)bo]

I was just making a marketing point. It sure sounds a whole lot better the way I spun it, I think so. Like I said donate the technology to the community! but obviosly your happy supporting a propietary standard thats going nowhere fast. Why dont you support something thats beneficial to everyone? There is a difference between letting someone barrow your stuff and just giving it to them to do what hey want! You can only use what they let you, you cant just go and build your own physx ppu and sell it why? because its not free or open!

AMD is already working on getting their opencl certification for the gpu. They have it for the cpu, but the gpu is still waiting for certification from the khronos group. There opencl driver will be released within 2 months!

[Farinorco]

Check this out:

http://developer.nvidia.com/object/physx_downloads.html

Literally anyone can download the SDK for free. So why hasn't it "taken the world by storm" as you said? It's proprietary. NVIDIA claims that PhysX, along with CUDA, is open to any hardware should one want to support it. The most likely reason holding AMD back from doing this is the unbalanced performance prospects of such an implementation. They probably don't feel like they need to anyway, since more open standards like Bullet physics will eventually arise to make use of OpenCL.

Of course if AMD was so confident in OpenCL, they might be more enthusiastic to get a working driver out for it, but as it is, even NVIDIA has beat them to the punch there. AMD makes great hardware but they really suck in the software/developer support side compared to the competition.

IMO, the most likely reason holding AMD back from adopting CUDA is the fact of being propietary itself.

If AMD would have adopted CUDA, it would have been probable that CUDA would have been standardized (given the absence of other standards that are starting to appear now.

Then, being a propietary API, everything would be in the hands of their main (only?) competitor:

What if they want to, from a certain version of the API, not allow the use of the latest versions to the competitors, to keep themselves ahead? (Creative and EAX).

The developement of the API would depend only on the criteria of NVIDIA, and they wouldn't have a word on that, also.

Of course, there's what you say about performance, but IMO it would be one of the less problematic here.

There are tons of reasons why AMD isn't interested in a propietary API from it's main competitor becoming the de facto standard for GPGPU, so better if they don't help their enemy to achieve that.

About OpenCL, I have seen them very enthusiastic. They have send a driver to the Kronos group that is approvation pendant (for compliance), even before (little before, I know) than NVIDIA. And the same that NVIDIA, they have a pre-release driver for developers with which you can work. They are not publicising so much as NVIDIA, but they are using it more. They are working with Havok, Bullet Physics and Pixelux, for example, to implement OpenCL. So I don't know how is NVIDIA beating AMD in OpenCL ground.

[Chumbucket843]

IMO, the most likely reason holding AMD back from adopting CUDA is the fact of being propietary itself.

If AMD would have adopted CUDA, it would have been probable that CUDA would have been standardized (given the absence of other standards that are starting to appear now.

Then, being a propietary API, everything would be in the hands of their main (only?) competitor:

What if they want to, from a certain version of the API, not allow the use of the latest versions to the competitors, to keep themselves ahead? (Creative and EAX).

The developement of the API would depend only on the criteria of NVIDIA, and they wouldn't have a word on that, also.

Of course, there's what you say about performance, but IMO it would be one of the less problematic here.

There are tons of reasons why AMD isn't interested in a propietary API from it's main competitor becoming the de facto standard for GPGPU, so better if they don't help their enemy to achieve that.

About OpenCL, I have seen them very enthusiastic. They have send a driver to the Kronos group that is approvation pendant (for compliance), even before (little before, I know) than NVIDIA. And the same that NVIDIA, they have a pre-release driver for developers with which you can work. They are not publicising so much as NVIDIA, but they are using it more. They are working with Havok, Bullet Physics and Pixelux, for example, to implement OpenCL. So I don't know how is NVIDIA beating AMD in OpenCL ground.

CUDA is a computing architecture. the API for CAL and CUDA are both just C with extensions so they are really both proprietary. the API must have a compiler for both architectures. CUDA is not something you just port to another gpu. there would be really no point because they are the same thing.

[Farinorco]

CUDA is a computing architecture. the API for CAL and CUDA are both just C with extensions so they are really both proprietary. the API must have a compiler for both architectures. CUDA is not something you just port to another gpu. there would be really no point because they are the same thing.

I'm obviously talking about the CUDA API, not about the CUDA architecture. The CUDA architecture is something internal, as it's the ISA, and access is granted through other intermediate layers, so nothing to standardize here. It's their propietary API what they want to see used, nobody uses "architectures" directly...

As has been stated, its not exclusively for gaming. Its for bigger projects.

Deal with it.

If it is also a very good competitor in gaming...thats just a bonus from nvidias perspective.

That's exactly the problem, IMO (and it's extensible to the previous gen too).

NVIDIA is trying to reach a new (emerging) market, what they call the HPC market, with their architecture. The problem is that this new market isn't there yet, so they can't split their R+D and chip manufacturing in 2 different architectures and/or product lines, so they have to make 3D rendering chips (the current market) that are good for the HPC market.

That eats transistors, developement of the architecture, and so. So the resultant chip isn't specialiced to the 3D rendering market, and has difficulties to compete with other products that are (in an efficiency performance/features to cost manner).

I'm starting to think that's the main reason of the weak performance to size ratio of the past generation, and I'm starting to think that we are going to see a repeat in this one.

[Luka_Aveiro]

I'm obviously talking about the CUDA API, not about the CUDA architecture. The CUDA architecture is something internal, as it's the ISA, and access is granted through other intermediate layers, so nothing to standardize here. It's their propietary API what they want to see used, nobody uses "architectures" directly...



That's exactly the problem, IMO (and it's extensible to the previous gen too).

NVIDIA is trying to reach a new (emerging) market, what they call the HPC market, with their architecture. The problem is that this new market isn't there yet, so they can't split their R+D and chip manufacturing in 2 different architectures and/or product lines, so they have to make 3D rendering chips (the current market) that are good for the HPC market.

That eats transistors, developement of the architecture, and so. So the resultant chip isn't specialiced to the 3D rendering market, and has difficulties to compete with other products that are (in an efficiency performance/features to cost manner).

I'm starting to think that's the main reason of the weak performance to size ratio of the past generation, and I'm starting to think that we are going to see a repeat in this one.

That could be a way to describe what's going on with nV lately, using a hybrid solution in order to explore both markets (gaming and gpgpu) with one product.

If they stand in both markets, I bet the gpgpu market will have it's own line of chips and the gaming market will have it's own too. Maybe we are far from that, maybe one or two generations for that to happen... It depends on many factors, but the most relevant of all, imo, is Fermi's success.

[tajoh111]

Yeah its pathetic whats going on. Im not a fanboy of either camp. I have a 9800gt, Im sorry 8800gt, whats the difference but I tend to favor Ati's buisiness ethics. At least they improve the industry for all consumers instead of the nvidian hypocrites. Nvidia, Were saving gameing by making pc games "special" and adding Physx. If they truley ment it, they would do whats best for us consumers and improve games for all platforms, even at a loss. Whats the saying "you recieve 10X what you give away" and I believe every word of it. In fact Id have more respect if they improved games at a loss, it isnt a loss if you recieve something in return, like sabotaging ati performance. If nvidia truley wants Physx to take the world by storm, then give it away and set it free, and open it up to everyone who really needs it. Just by doing this alone could be enough to make it adopted and used by everyone. They could make more money making it work better than anyone else, than being the only one with it, and then the money will come to them. Its really simple aint it?

If I had the cure for cancer what would I do?
a: Sell it by the bottle to be a millionaire
b: Give it away to save millions of lives

I would do b, but thats me, But I bet you I would make more than enough cash giving it away and saving lives.

Think if they marketed physx as folding for cures, and how they donated this technology to better help mankind. That would make me feel so fuzzy inside that Id want to buy a nvidia card! cuz were saving lives here and Nvidia can save more when you use their cards.

Instead of donating money to charity, just buy a nvidia gpu for 50$ or 100$ or more and help find the cures! anyone can help, see how much your helping day to day and make a difference. That sounds great if I do say myself!

This is not how you run a business. What world do you live in, a fantasy world where everything is powered by love. Should they give away computers to third world countries so they can learn, should they give free videocards to people so that they may fold? Your being completely unrealistic.

If they give physX out to anyone and didn't charge licensing for it, they would have just lost money developing the technology and lost money for buying ageia.

From your lovey dovey perspective, what as AMD done to benefit mankind? They run a business like anyone else.

By some of the points you are making, NV is already starting to save lives because their cards are the best at folding for cures and as a result they are one of the biggest contributers(ps3(rsx) too). You most likely know this too, so buy more NV cards. Honestly, I would find it tacky and relatively deceptive if NV main advertisements mentioned folding for diseases everywhere since we all know this was not their main purpose. It would seem shallow and artificial, because all of us that are not naive would know that NV sells videocards mainly and the folding part is just a bonus.

I don't even see how AMD improves the industry for all consumers, besides being present to present competition for the industry they are in. Something NV does as well.

Also how are you giving it away(the cure for cancer), if your still making money off it. The most selfless thing(and the right thing to do in your opinion) would be to not collect any of the money, any money generated from the patent donated to charity and lastly to do this somehow anonymously so you don't get the credit for it.

[Farinorco]

This is not how you run a business. What world do you live in, a fantasy world where everything is powered by love. Should they give away computers to third world countries so they can learn, should they give free videocards to people so that they may fold? Your being completely unrealistic.

If they give physX out to anyone and didn't charge licensing for it, they would have just lost money developing the technology and lost money for buying ageia.

From your lovey dovey perspective, what as AMD done to benefit mankind? They run a business like anyone else.

By some of the points you are making, NV is already starting to save lives because their cards are the best at folding.

I don't even see how AMD improves the industry for all consumers, besides being present to present competition for the industry they are in. Something NV does as well.

Guys, you know that NV does give PhysX out to anyone and don't charge licensing for it (developers or IHVs as AMD...), don't you?

They offered to implement CUDA for free to AMD (and PhysX in the batch), and PhysX it's completely free to license for software developers.

They bought AGEIA to use their PhysX as an attempt of marketing CUDA in the mainstream, gamer territory in it's race to standardize CUDA as the de facto GPGPU standard.

They didn't need money to produce a benefit from their investment, what they need it's CUDA being used.

[To(V)bo Co(V)bo]

This is not how you run a business. What world do you live in, a fantasy world where everything is powered by love. Should they give away computers to third world countries so they can learn, should they give free videocards to people so that they may fold? Your being completely unrealistic.

Also how are you giving it away(the cure for cancer), if your still making money off it. The most selfless thing(and the right thing to do in your opinion) would be to not collect any of the money, any money generated from the patent donated to charity and lastly to do this somehow anonymously so you don't get the credit for it.

Im sure Nvidia wouldnt make any money selling video cards if they had a life changing initiative to fold for cures.

There has been countless people who have left a impact in this world, some good and some bad. The only ones that matter are the good ones, because without them you may have never even had a chance to live. How many people can you name that have done something good for mankind? Its always harder to remember the good people that made a difference than the the ones who never do.

Why dont you read the buisiness practices of this guy : Percy Julian http://inventors.about.com/library/i...lcortisone.htm


Read about this guy too while your at it: Forrest Bird http://inventors.about.com/od/bstart...rrest_Bird.htm

[saaya]

Think if they marketed physx as folding for cures, and how they donated this technology to better help mankind. That would make me feel so fuzzy inside that Id want to buy a nvidia card! cuz were saving lives here and Nvidia can save more when you use their cards.

Instead of donating money to charity, just buy a nvidia gpu for 50$ or 100$ or more and help find the cures! anyone can help, see how much your helping day to day and make a difference. That sounds great if I do say myself!

good point, that actually DOES make a diference... thats the ONLY useful application of gpgpu so far imo...

AMD makes great hardware but they really suck in the software/developer support side compared to the competition.

yepp... they have come a long way and improved their driver support, but its still not great, and still slightly behind nvidia... and their dev support is much worse compared to nvidia afaik...

and farinoco, if cuda wouldnt have been a closed standard, we might have never seen opencl and direct compute... they might have adopted it or made it a v1.1 or 2.0 of cuda... but nvidia was too caught up in promoting themselves and too greedy, trying to lay chains on everybody else...

asking others to trust them in not crippling cuda for them is really quite something after nvidias history with other industry players...