NVIDIA GF100 & Friends speculation

[Razor1]

Very interesting. Can you show us some real numbers?

Aaaand that benchmarks, performance analysis and other thoughts you're all posting here... Damn who do you think is interested?

Take it easy, it was just green goblins presentation for their best frineds. All cherry picked benchies.

And what about that video leaks... Do anybody here think it was real leak? ROFL... It's all just a game. Nvidias game with us, normal people, tech websites readers, potentioal buyers. "Can you read us [to my cam] that FPS numbers again?"

It was all just part of a plan, marketing. Every company is trying to sell their stuff, and this is preparation phase. Prepare people, let them speculate, then make a big boom release, then sell like hotcakes. Good marketing, nothing special.

Booooring... :smile:

The video leaks, someone found them on PCPrespective's website , while they were preparing thier article.

No one in the press community has the cards in hand yet, they will soon though, and the benchmarks they have shown so far, is actually not thier "Best Advantages" in game situations.

 

[Razor1]

http://www.hardware.fr/articles/770-23/dossier-amd-radeon-hd-5870-5850.html

you are absolutely right... this definitely puts the “up to 2x over GT200 @8xAA high res” in a different light now, does it? ;-)

Cypress is already almost twice as fast as GT200 - on average

X8 AA on the G100 only has a ~10% penelty from going from x4 to x8 AA, also x4 AA has less performance drop then on the gt200 chips. If we work backwards from those numbers with 3 times the shader performance from aribtrary reviews, what do we get? Pointless isn't it?

This is exactly why I stated before these arguements are useless!

 

[compres]

Then again, if you look back in the thread to about 24 hours ago, according to some, we were supposed to have all questions answered by now. I wonder what happened?

-Charlie

I have to say I am very happy with the fantastic geometry performance and this could probably make me consider this chip for my next machine (2h2010). But I have to agree with charlie, none of my question were answered at all...

 

[compres]

Do you know that GF100 will be sold in professional and HPC markets where the margins are sky high and AMD has next to 0% market share, despite having great consumer products?

GF100 will make a lot of money for nv as r&d for mainstream consumer market has been paid off already, and *profits* in quadro and tesla markets are worth a LOT.

There is a big chunk of this thread talking about profits in the HPC and professional market. General consensus is that they are in fact not significant compared to the consumer market. Then again worth a LOT can be interpreted as anything.

 

[Kakkoii]

EDC is not ECC.

Indeed it is not. But it is similar, less useful/functional.

 

[Florin]

There is a big chunk of this thread talking about profits in the HPC and professional market. General consensus is that they are in fact not significant compared to the consumer market. Then again worth a LOT can be interpreted as anything.

Hmm, according to JHH Quadro accounts for 2/3rds of Nvidia profit.

 

[nagus]

X8 AA on the G100 only has a ~10% penelty from going from x4 to x8 AA, also x4 AA has less performance drop then on the gt200 chips. If we work backwards from those numbers with 3 times the shader performance from aribtrary reviews, what do we get? Pointless isn't it?

This is exactly why I stated before these arguements are useless!

You can spin this the way you like, but that doesn't debilitate the fact that nvidia claims "up to 2x performance over GT200" (in marketing slides!) where Cypress is already 80% faster on average.

 

[Sontin]

One thing to keep in mind, NV is only quoting a small portion of Unigine/Heaven for their numbers, and those are by far the most tessellation heavy portions of the benchmark. They are specifically not quoting the whole benchmark numbers. Think there is a reason for that? They don't even show numbers for heavy tessellation WITH other things going on. Hmm......

"Other Things"? Physics, KI? They showed their tessellation performance with a real tessellation benchmark. It's a litte bit funny because AMD showed their tessellation ability in a tessellation tool, too. Why don't they show performance numbers in games?

To me it appears to be a great GPGPU architecture with some graphics tacked on. I can't for the life of me figure out why a 2:1 SPP FP ratio is good for gaming, nor ECC capabilities, but then again, does that cost them anything other than die area, yield, and power?

So, Tessellation is now a GPGPU feature?
Do you know, that AMD has the same DP numbers like nVidia?

 

[aaronspink]

Do you know that GF100 will be sold in professional and HPC markets where the margins are sky high and AMD has next to 0% market share, despite having great consumer products?

and do you know that the professional and HPC markets are incredibly small, and at least one of those markets will likely, NOT by the version that Nvidia wants them to buy? For the HPC market, please read up on the history of it, a large number of the systems sold are sold cheaper than the general market. There is very little advantage to buying a Tesla when a Geforce can run the same software and is a small fraction of the cost.

And oh, on a volume basis, ATI effectively killed one of the largest volume parts of the professional market with eyefinity in the 5xxx series. What used to take an up-priced speciality board can now be done with a standard OTS part. So really, the only part of the professional market left will be the 3DCC and CAD/CAM portions of the market.

GF100 will make a lot of money for nv as r&d for mainstream consumer market has been paid off already, and *profits* in quadro and tesla markets are worth a LOT.

GF100 currently has no revenue as such it has not paid off any R&D. It is likely they can at least earn back the R&D costs but I think people are putting WAY to much stock into the quadro/tesla markets.

 

[aaronspink]

No one in the press community has the cards in hand yet, they will soon though, and the benchmarks they have shown so far, is actually not thier "Best Advantages" in game situations.

Forgive me, but I find it hard to believe that any company, let alone nvidia, wouldn't put their best foot forward for a marketing/PR event.

As far as they will soon, we've been hearing a "will soon" from people like you for almost a YEAR now. Another couple of months and Nvidia will be in SUN Micro territory!

 

[Sontin]

GF100 currently has no revenue as such it has not paid off any R&D. It is likely they can at least earn back the R&D costs but I think people are putting WAY to much stock into the quadro/tesla markets.

Their Quadro business is nearly a profit business. They have margins of 60-80% which is twice of the geforce business. In the last quarter they made with 1/4 revenue of the geforce business the same profit. And this without a growth in their quadro segment.

 

[Razor1]

Forgive me, but I find it hard to believe that any company, let alone nvidia, wouldn't put their best foot forward for a marketing/PR event.

As far as they will soon, we've been hearing a "will soon" from people like you for almost a YEAR now. Another couple of months and Nvidia will be in SUN Micro territory!

trust me

 

[Razor1]

You can spin this the way you like, but that doesn't debilitate the fact that nvidia claims "up to 2x performance over GT200" (in marketing slides!) where Cypress is already 80% faster on average.

that's not spin its stupidity to speculate on numbers from different sources, its not like you are collating data and doing anything empirical, its just BS at the end.

 

[Tchock]

By the way, why was there no mention of video processing capabilities? Unchanged from GT21x presumably?

What is left to improve? Most likely no bitstreaming of certain formats though.

 

[nagus]

that's not spin its stupidity to speculate on numbers from different sources, its not like you are collating data and doing anything empirical, its just BS at the end.

So whats BS then? hardware.fr Reviews or the slides from nvidia? Care to explain? I guess not.

 

[aaronspink]

Their Quadro business is nearly a profit business. They have margins of 60-80% which is twice of the geforce business. In the last quarter they made with 1/4 revenue of the geforce business the same profit. And this without a growth in their quadro segment.

Quadro is only a profit business because it doesn't have to pay the R&D. As goes their mainstream profits, so goes their Quadro profits. And of course it is without growth, as it isn't a growth market. If anything it is a mature market that will SHRINK over time. A large volume portion of the Quadro/professional market isn't actually high end 3DCC/CAD/CAM but multiple display cards which is effectively a DEAD market at this point with what ATI has done with eyefinity.

 

[Razor1]

So whats BS then? hardware.fr Reviews or the slides from nvidia? Care to explain? I guess not.

look up the definition of empirical, then you know what BS is when you guys just pull numbers from all over the place.

 

[jimmyjames123]

A large volume portion of the Quadro/professional market isn't actually high end 3DCC/CAD/CAM but multiple display cards which is effectively a DEAD market at this point with what ATI has done with eyefinity.

A large portion of NVIDIA's Quadro/Professional market is dead because of eyefinity? LOL, that is a ludicrous statement. Eyefinity is actually marketed towards high end gamers.

 

[Razor1]

Quadro is only a profit business because it doesn't have to pay the R&D. As goes their mainstream profits, so goes their Quadro profits. And of course it is without growth, as it isn't a growth market. If anything it is a mature market that will SHRINK over time. A large volume portion of the Quadro/professional market isn't actually high end 3DCC/CAD/CAM but multiple display cards which is effectively a DEAD market at this point with what ATI has done with eyefinity.

What? what does eyefinity have to do with the quadro/ proffesional market?

 

[Florin]

Quadro is only a profit business because it doesn't have to pay the R&D. As goes their mainstream profits, so goes their Quadro profits. And of course it is without growth, as it isn't a growth market. If anything it is a mature market that will SHRINK over time. A large volume portion of the Quadro/professional market isn't actually high end 3DCC/CAD/CAM but multiple display cards which is effectively a DEAD market at this point with what ATI has done with eyefinity.

I agree that it will shrink but it can't be proclaimed dead yet for quite some time. ATI has very little mindshare in this highly conservative market. Even Matrox has managed to this day to subsist on their slice of the profits there with zero product innovation.