If fermi flops what happens to Nvidia

[SunnyD]

Life goes on. They come back and spin their next design out. They have enough in the bank that they can royally fuck up about 2 or 3 generations before they have to worry.

 

[jvroig]

It's too much to quote, so I'll just mention the references.

It takes two to tango, so what we have now (as I understand it, and I certainly could be wrong and/or you most certainly can have a very different opinion and we're still good) is because of the dance between nVidia and ATi. To me, things seemed to go as any outside observer expects it. nVidia, being the giant in the playground, will play relatively safer, and will focus on maximizing profits and being outrageously innovative will play second fiddle, and the focus on marketing becomes more obvious (it's always that way, even in most other fields. Take SAP. Far different market, same symptom: makes world-renowned enterprise software that some love, some hate. Commonality? They've got a ton of sales people, and they reek of marketing.)

ATi, on the other hand, is the small player (more so back then, now not quite as much but still not a giant the equal of nVidia, but seems to be making the right moves and can get there), so their focus is to get recognition, to win new customers, to grab marketshare. All these things aren't for "safe" moves, so they've got to one-up nVidia - first to deploy DX version-something, introduce tessellation or whatever feature, anything to make them stand out, to make them seem like "not just another player". In their dance, nVidia is leading, doing the relatively safer moves in general and acting like a giant company (which they are), while ATi does the dance of underdog.

ATi did what they had to, back then and now, and nVidia simply acted not too far from how one would expect given their standing (and you can't fault them; for all our disappointments with them due to any Fermi concerns we may have, they are pretty cash rich and their sales figures are still good, and no doubt their stockholders are currently very happy, maybe even happier than those of ATi, even if ATi is the current celebrated company of enthusiasts).

Just my 0.02. Reading the comments of dguy6789 and evolucion8 made me a little bit "inspired" to post a little bit of my thoughts tonight. :thumbsup:

 

[Lonyo]

It's too much to quote, so I'll just mention the references.

It takes two to tango, so what we have now (as I understand it, and I certainly could be wrong and/or you most certainly can have a very different opinion and we're still good) is because of the dance between nVidia and ATi. To me, things seemed to go as any outside observer expects it. nVidia, being the giant in the playground, will play relatively safer, and will focus on maximizing profits and being outrageously innovative will play second fiddle, and the focus on marketing becomes more obvious (it's always that way, even in most other fields. Take SAP. Far different market, same symptom: makes world-renowned enterprise software that some love, some hate. Commonality? They've got a ton of sales people, and they reek of marketing.)

ATi, on the other hand, is the small player (more so back then, now not quite as much but still not a giant the equal of nVidia, but seems to be making the right moves and can get there), so their focus is to get recognition, to win new customers, to grab marketshare. All these things aren't for "safe" moves, so they've got to one-up nVidia - first to deploy DX version-something, introduce tessellation or whatever feature, anything to make them stand out, to make them seem like "not just another player". In their dance, nVidia is leading, doing the relatively safer moves in general and acting like a giant company (which they are), while ATi does the dance of underdog.

ATi did what they had to, back then and now, and nVidia simply acted not too far from how one would expect given their standing (and you can't fault them; for all our disappointments with them due to any Fermi concerns we may have, they are pretty cash rich and their sales figures are still good, and no doubt their stockholders are currently very happy, maybe even happier than those of ATi, even if ATi is the current celebrated company of enthusiasts).

Just my 0.02. Reading the comments of dguy6789 and evolucion8 made me a little bit "inspired" to post a little bit of my thoughts tonight. :thumbsup:

You missed one important part which has really come to the fore with Fermi.
NV is a GPU company. It was a GPU/chipset.
ATI is a GPU/chipset company owned by a CPU company.

That has a serious implication for future products, and it what (IMO) is driving current and future design ideas.

 

[tvdang7]

nothing wait for next gen. Ati flopped with 2900xt and they are the underdog. They are still alive today.

 

[NoQuarter]

You missed one important part which has really come to the fore with Fermi.
NV is a GPU company. It was a GPU/chipset.
ATI is a GPU/chipset company owned by a CPU company.

That has a serious implication for future products, and it what (IMO) is driving current and future design ideas.

Very true, imo AMD bought ATI for 2 reasons, chipsets and to get GPU patents/knowledge for their future APU design (CPU + GPU chip). They didn't buy ATI to make money off their graphics card division at all.. it was all for what ATI could bring to AMD's x86 line and we will see how it finally pays off soon.

Nvidia can't compete in x86 market without patents, which is why they fought Intel so hard in lawsuits, and are now steering their products toward HPC market and establishing their GPUs as general purpose parallel processors.

 

[ultimatebob]

I highly doubt that it will "fail", unless it's released with a serious firmware bug that causes the entire product line to be recalled.

If NVidia's fastest part isn't as fast as ATI's fastest part, NVidia will just lower the price on their part until it's comparable with an ATI part of equal performance. You know... the same thing that both video card manufacturers have been doing for years.

 

[v8envy]

ATI plays the price to performance game, NV not so much. They just crank the hype and viral marketing machine up a notch.

You know, the crack marketing department that's been milking VACUUM for over six months to keep the Fermi buzz alive. I have no doubts those guys could move a bunch of GT210s restickered as GTX590s for $800 each, and Fermi flopping (if that happens) will be a complete noop as far as gaming hardware goes.

 

[Anarchist420]

nvidia will certainly survive a fermi flop. Fermi will flop anyway.

 

[Stoneburner]

If ATI introduced tesselation capable hardware in 2002, why is it Nvidias fault that we haven't had tesselation enabled software until 8 years later? Don't you thinki it might have been prudent for ATI to push this since their very first tesselator hit their hardware? I believe it was called TruForm back then. Why didn't they work with devs to furiously push this technology to its limits for the better part of a decade?
And now hardware based physics as well? AMD refused to support PhysX. Where do you think we would be today in our games if they had? Or do you think ATI jumping onboard wouldn't have made any difference?
Look over the last 3 years bud, and sit there and try to tell me who is the bigger innovator in this field. NO contest. Sorry.

This is odd. First, you are saying ATI didn't MARKET its technology. Then you say it didn't push PHYSX, which was made by a third party, not Nvidia. Nvidia is more innovative because it markets better and it buys out companies?

You should retract this post.

 

[zerocool84]

If ATI introduced tesselation capable hardware in 2002, why is it Nvidias fault that we haven't had tesselation enabled software until 8 years later? Don't you thinki it might have been prudent for ATI to push this since their very first tesselator hit their hardware? I believe it was called TruForm back then. Why didn't they work with devs to furiously push this technology to its limits for the better part of a decade?
And now hardware based physics as well? AMD refused to support PhysX. Where do you think we would be today in our games if they had? Or do you think ATI jumping onboard wouldn't have made any difference?
Look over the last 3 years bud, and sit there and try to tell me who is the bigger innovator in this field. NO contest. Sorry.

Umm because Nvidia has and had a bigger market share and had the muscle to get rid of tessellation out of DX10 so for all we know they made sure other devs didn't use it early before either. If Nvidia really wanted to get Physx out there, they would allow everybody and every card to use it but we all know how that is.

 

[Keysplayr]

ATi's market share was smaller at that time than now, and yet a couple of games came out with Truform like Serious Sam, a big hit. And it looked much better, also Unreal Tournament, but I don't remember which version it was, from there is where the dildo gun joke came from.

Serious Sam and Unreal Tournament? How many years ago was that? And if it were such a "hit" as you say, WHY didn't ATI keep pushing it??? How many years have they let go by without trying to push for that standard? Years man.

 

[Cuular]

And now hardware based physics as well? AMD refused to support PhysX. Where do you think we would be today in our games if they had? Or do you think ATI jumping onboard wouldn't have made any difference?

To bad the above statement is not true according to AMD/ATI.

http://www.bit-tech.net/bits/interviews/2010/01/06/interview-amd-on-game-development-and-dx11/1

See the bottom of the page. AMD/ATI have tried to talk to NVIDIA about PHYSX and as the article says there, they were told to go "whistle".

PHYSX could be a real mainstream API but for whatever reason NV isn't ready for that to happen.

 

[Lonyo]

To bad the above statement is not true according to AMD/ATI.

http://www.bit-tech.net/bits/interviews/2010/01/06/interview-amd-on-game-development-and-dx11/1

See the bottom of the page. AMD/ATI have tried to talk to NVIDIA about PHYSX and as the article says there, they were told to go "whistle".

PHYSX could be a real mainstream API but for whatever reason NV isn't ready for that to happen.

PhysX can't be an API.
It can be a piece of middleware which runs on freely available to use APIs, such as OpenCL or DirectCompute, but it can't be an API itself.
PhysX is middleware which presently, in accelerated form runs on the CUDA API and whatever lets it run on PPUs.

 

[NoQuarter]

PhysX can't be an API.
It can be a piece of middleware which runs on freely available to use APIs, such as OpenCL or DirectCompute, but it can't be an API itself.
PhysX is middleware which presently, in accelerated form runs on the CUDA API and whatever lets it run on PPUs.

It's still an API, API is just a term for programming libraries/functions, which is exactly what PhysX is. Basically the PhysX SDK consists of the PhysX API and some tools to help implement it.. even NVidia calls it an API interchangeably.

 

[OCGuy]

Fermi isnt flopping, much to Charlie's dismay.

It will be the fastest chip, and they can only screw it up by pulling a GT200 and pricing too high at launch.

Either way, there is a ton of enthusiast $$$ on the sideline waiting for the 480. Just looking at the EVGA box art got my Debit Card trembling...

 

[Hacp]

The biggest problem with Nvidia is that its trying to do too much, mainly because it doesn't have a CPU division to partner up with. GPGPU is a nice idea and all but that just adds to the size of the GPU and increases the price. We like cheap prices.

So when will Nvidia just use the cash and buy via already?

 

[soccerballtux]

If ATI introduced tesselation capable hardware in 2002, why is it Nvidias fault that we haven't had tesselation enabled software until 8 years later? Don't you thinki it might have been prudent for ATI to push this since their very first tesselator hit their hardware? I believe it was called TruForm back then. Why didn't they work with devs to furiously push this technology to its limits for the better part of a decade?
And now hardware based physics as well? AMD refused to support PhysX. Where do you think we would be today in our games if they had? Or do you think ATI jumping onboard wouldn't have made any difference?
Look over the last 3 years bud, and sit there and try to tell me who is the bigger innovator in this field. NO contest. Sorry.

Member of Nvidia Focus Group
NVIDIA Focus Group Members receive free software and/or hardware from NVIDIA from time to time
to facilitate the evaluation of NVIDIA products. However, the opinions expressed are solely those of the Members.
Intel i7 860 @ 3.4Ghz, Asus Maximus III P55, 8GB DDR3, 2x GTX295 (QSLI)
Heat 68-0-0

 

[Keysplayr]

Member of Nvidia Focus Group
NVIDIA Focus Group Members receive free software and/or hardware from NVIDIA from time to time
to facilitate the evaluation of NVIDIA products. However, the opinions expressed are solely those of the Members.
Intel i7 860 @ 3.4Ghz, Asus Maximus III P55, 8GB DDR3, 2x GTX295 (QSLI)
Heat 68-0-0

Ummmmm, thanks? Did I mis-spell something?
Or was this just a mighty fine quality bait attempt?

I'm not bitin dude. So stew.

 

[ronnn]

If fermi continues to be a long term flop. Expect nvidia pr to continue with a major fud campaign along with a focus to physx.

Really gaming computers with expensive graphic cards are getting rare. The war is over, Sony and Microsoft have won. Nvidia and amd will have to make big profits elsewhere.

 

[Kenmitch]

Fermi isnt flopping, much to Charlie's dismay.

It will be the fastest chip, and they can only screw it up by pulling a GT200 and pricing too high at launch.

Either way, there is a ton of enthusiast $$$ on the sideline waiting for the 480. Just looking at the EVGA box art got my Debit Card trembling...

Guess you meant this as a joke.

As the lack of artwork on the box could also be saying what evga thinks of it.

I guess you have a point on the $$$'s sitting on the sidelines waiting to see if fermi was worth the wait.

Given the amount of time fermi has been delayed and the lack of nvidia hype it doesn't look good for fermi.

But I guess the diehard nvidia fanboys always will have fermi 2 to wait for. Or will just buy a slightly worse or better card because it's a nvidia!

I don't care either way if it flops or not as I'm not in the market for a video card anyways.

I only brought up the subject to see what others thought about the situation. It's best for all parties concerned to have multiple players in the market as it spurs innovation and helps keep prices in line.

 

[Tempered81]

It's not a flop if you ask me. It's late, big & hot, and gpgpu-centric, but it's still going to be faster, and the number 1 single gpu. It will have strengths in the low-end, minimum fps, tessellation, and anti-aliasing. It probably won't peak as high as a 5870 or 5970 in common game settings, but it is a powerful chip. I think the issue with it being spotty & voltage needy has put a damper on the clock potential. High clocks would give good framerates at common benchmark settings. As it stands now, it doesn't look to get more than 20% on the 5870, and I could still see the 5870 winning a few of them. The only way it could topple a 5970 is huge resolution, 8xAA, maximum tessellation scene where the 5970 chokes, and gtx480 shows it's strengths.

Powerful chip, and good design for compute, cuda, c++ and all that jazz. Not a flop, but still not a powerful gaming gpu. And probably won't ship in volume until they respin/redesign, and come out with a better "Fermi II" later on.

I still give props to Nvidia for innovation and thinking outside the box. Big risks like that increase the probability of a flop. Perhaps they didn't expect Ati to double rv770 (which still isn't as big a jump as rv670 to rv770 was). Can't wait to see the look on their face when fermi II refresh/respin/shrink has to fight against the "hechtoneries" or whatever rv970 is code-named.

And everyone has the cypress refresh to look forward too, along with the 2gb eyefinity6, so it may be a tough sell for nvidia.

 

[cbn]

The only way it could topple a 5970 is huge resolution, 8xAA, maximum tessellation scene where the 5970 chokes, and gtx480 shows it's strengths.

I don't know if a 480 GTX could beat a HD5970 in tessellation (even in the heaven benchmark).

It might be close though.

 

[nitromullet]

IMO the more interesting question is what happens if ATI produces a card that is a flop. The reason I say this is not because I have any desire for this happen, but because if NV flops a product line, Jen-Hsun will push NV to do better next time.

NV has other markets, but gpu/gpgpu is their core business and their future relies on the technology they produce. ATI on the other hand is just the high end graphics division of a cpu/chipset/gpu maker. Should the ATI graphics division start to hemorrhage cash, how long will AMD keep them on life support? I'm pretty sure this is not lost on ATI management. For ATI, I think being successful is a do or die situation. For NV, it just means they have to dig in, and get back to work.

 

[nitromullet]

As the lack of artwork on the box could also be saying what evga thinks of it.

Not that box art plays a big role in my decision making process, but EVGA has never really been a snarling dogs, sword wielding hottie, or crazy madman on the box kind of company. Where one sees "lack of artwork", another sees "tastefully executed design". To each is own, I guess.

 

[bryanW1995]

you guys have all missed jhh's evil plan: delay fermi, spread bad rumors about it, bump it up from gtx 3xx to gtx 4xx to increase "geek water cooler" talk, release stillborn gtx 4xx, then, finally, release fermi refresh as...gtx 5899. all the nvidia fanbois will buy it b/c, like, ZOMG it's nvidia, and all the others will buy it b/c 5899>5890. amd is so screwed right now...