The GT300/Fermi Thread - Part 2!

Quote:

---

Originally Posted by saaya

why should nvidia sell it for such a low price if it performs as well as a 5970 or better?

---

performance.. we dont know.
Cost - 3B Tran@40nm - definitely very high.

I'm betting $600. Unless nVidia pulls a Sony - inlude a blu-ray, and make up loss by selling games LOL

Quote:

---

Originally Posted by FischOderAal

Yeah right :D At first everyone said "wait until Fermi is there before buying HD5k" and now you say "wait until HD6k is there before buying Fermi"? Jesus :D You should buy your freaking GFX whenever you need it, not when your glass globe tells you to.

---

Yeah my glass globe has been pretty smelly recently.. maybe its the floating goldfish.
I bought DX10 AFTER Crysis launched.. expect to do the same with DX11 when BadCompany2 and Crysis2 arrive.

Quote:

---

Originally Posted by saaya

i think ati could have sold quite some additional 5800 cards if they would have reduced the price 50-100$ .....

---

#cards @ 400 x profit(400) > #cards @ 300 x profit(300).
ie COST=200.. 100 x $200 = Y x $100,
Y = 200. (TWICE as many as 100)
Basically, I'm sure AMD marketing is smart enough to do the math, and probably came up that they wouldn't find twice as many gamer consumers if they lowered the price.

Quote:

---

Originally Posted by zerazax

Exactly, although ATI had the right idea by first testing out 40nm on the RV740 and going for a smaller design, which made their transition much smoother (relatively)

---

This goes back to 55nm RV670. AMD almost (RV630) skipped 65nm altogether. Likewise AMD ahead on implementing Tesselation, Vista/Win7 driver, GDDR5, 40nm, AVC/VC1 acceleration, DisplayPort... its a long list. l)

Deimos - Those were all excellent points above, you really hit the nail on the head :up:

less $ than 5970 dont mean less performance than 5970 lol

white paper fermi/gf100 specs dont add up to 3 billion transistors

lets just say they havent shown all their cards yet

Quote:

---

Originally Posted by NapalmV5

less $ than 5970 dont mean less performance than 5970 lol

white paper fermi/gf100 specs dont add up to 3 billion transistors

lets just say they havent shown all their cards yet

---

Ummm... with a chip like this, i.e. very costly to produce, yes it does.
If they could sell it for more they definitely would.

Number of trannies I speculated on back in Sept was ~3.1-3.4b(I pick 3.2b), going off the info that GF100 was G200 sized.

Quote:

---

Originally Posted by LordEC911

Not that simple. These designs are being worked on for 3-4 years before they are released. This isn't a simple, hey let's throw DX11 into the architecture and get it out next month since GF100 is having problems. There isn't much you can do as far as a backup plan in this industry the R&D costs are just way too high as it is.

---

Thats my sentiment on the issue more or less, there is little choice but to release and work through the issues for a product that has been in development win or loose which is what we've seen with other hardware in the industry. The only way to really have a backup plan would be to actually have a backup design of similar performance potential being developed in parallel in case the first design doesn't work out.

Quote:

---

Originally Posted by highoctane

Thats my sentiment on the issue more or less, there is little choice but to release and work through the issues for a product that has been in development win or loose which is what we've seen with other hardware in the industry. The only way to really have a backup plan would be to actually have a backup design of similar performance potential being developed in parallel in case the first design doesn't work out.

---

TL no workie, S3 S2000
the no show BitBoyz
the doomed Voodoo4.
hyped beyond belief, and barely propped up with questionable optimization, fx
circuit problem plagued X1800XT
power beast X2900XT

Launching a half-ready product could have worked in the last century.. nowadays with fierce competition and slim margins, the bad press will ruin your company. Only folks like Intel and Microsoft who are too big to fail, can get away with it (a few times).

Quote:

---

Originally Posted by highoctane

I gotcha, it really does seem like they fell asleep at the wheel in regards to getting any dx11 hardware out the door. Both ATI and Nvidia seem to release the latest on the highend part first before releasing the lower parts so with gt300 delay's you would have thought some midrange to lower end parts would have been scheduled to release by now.

---

very good point :up:
I've repeatedly asked the same question too - where are at least rumours of midrange/lowend DX11 tapeout?? Is nVidia abandoning that market segment? Obviously they can't make new Intel or AMD chipsets, so you'd think with resources concentrated on video cards, they'd be the ones ahead of AMD...:rolleyes:

FYI Fermi yields will obviously be worse than 5870. Question is how much, and the impact. Even if Fermi is launched at $299, low availability will create $1000 eBay market. ex X800XTPE, iPhone, PS3 etc..

Quote:

---

Originally Posted by ***Deimos***

I've repeatedly asked the same question too - where are at least rumours of midrange/lowend DX11 tapeout?? Is nVidia abandoning that market segment? Obviously they can't make new Intel or AMD chipsets, so you'd think with resources concentrated on video cards, they'd be the ones ahead of AMD...:rolleyes:

FYI Fermi yields will obviously be worse than 5870. Question is how much, and the impact. Even if Fermi is launched at $299, low availability will create $1000 eBay market. ex X800XTPE, iPhone, PS3 etc..

---

Rumors of midrange have always been late Q2, though no word on tape outs or anything is pretty strange.

Again, any GF100 product under $399 isn't going to happen anytime soon.

new stuff about the rocket sled demo.

http://www.youtube.com/watch?v=HjIzo...layer_embedded

Any folding ppd numbers out yet? With 512sps GTX 480 should be breaking 20k ppd easily.

Quote:

---

Originally Posted by LiquidReactor

Any folding ppd numbers out yet? With 512sps GTX 480 should be breaking 20k ppd easily.

---

There's been no new numbers for a while, I smell another delay, March starts to seem unlikely...

Quote:

---

Originally Posted by zalbard

There's been no new numbers for a while, I smell another delay, March starts to seem unlikely...

---

That is a load of crap. People start pulling stuff like that out of their rear and the rumors start spreading very easily due to a lot of anti-Nvidia feelings that seem to be very prevalent with a large number of people.

Quote:

---

Originally Posted by ElSel10

That is a load of crap. People start pulling stuff like that out of their rear and the rumors start spreading very easily due to a lot of anti-Nvidia feelings that seem to be very prevalent with a large number of people.

---

There is no official release date out so far. You have no facts to prove me wrong.
I am simply drawing the conclusion out of the current situation.

Quote:

---

Originally Posted by LiquidReactor

Any folding ppd numbers out yet? With 512sps GTX 480 should be breaking 20k ppd easily.

---

http://smiliesftw.com/x/bowdownred.gif if true

Quote:

---

Originally Posted by zalbard

There is no official release date out so far. You have no facts to prove me wrong.
I am simply drawing the conclusion out of the current situation.

---

Dont talk about outers lack of facts when they are commenting on the lack of yours :shakes:

Quote:

---

Originally Posted by LiquidReactor

Any folding ppd numbers out yet? With 512sps GTX 480 should be breaking 20k ppd easily.

---

Yeah, in single precision :p

*SCNR*

The wait on this card is very frustrating

Quote:

---

Originally Posted by NapalmV5

less $ than 5970 dont mean less performance than 5970 lol

white paper fermi/gf100 specs dont add up to 3 billion transistors

lets just say they havent shown all their cards yet

---

If Nvidia had a card that performed better than their competition, what makes you think they'd price it any less? For that matter, ATI as well?

The wait on anything is frustrating, waiting on my iPad to come so I can bash it in with a sledge hammer is frustrating.

As for fermi I am starting to wonder how massive the problems have been for Nvidia, and how long these problems could effect the future of PC gaming. Gf 100 is the archatecture that Nvidia, from what I have read was going to use for a long time to come, and from the looks of it, its not working out as they had hoped.

Nivida has a few ways then can go from here imho, they can pull a ATI and rework fermi in to something that sorta works, IE the 2900XT in to HD38XX cards, or they can try to build a new arch, which they should have already started back when Gf 100 started not working as expected.

I am not saying no one is going to buy Fermi, I bought a 2 2900Pros knowing darn well that a 8800 Ultra would have been a better buy, but I am saying is that if this was the first card of a new arch, fermi is not a repeat of the 8800, there is nothing here great enough for Nvidia to rename for the next 3 years.

I could be totally wrong and fermi could own everything, but I would suspect that if that was the case it would already be out, or at least there would be a few "real" leaks.

Its all bad though, fermi being late and costing a lot when it does come out just hurts PC gaming, Good gaming PC's need to be under 1,000 USD in order for PC gaming to make it back into the mainstream, and right now you can build a good Direct X 10 PC for around that, but throw in a 5850 or 5870 and you break the budget. Fermi needs to come out now, cost the right amount, and work at least haft as well as the fan-boys expect or PC gaming is going to be in trouble.

I am a AMD/ATI fan, I will admit it here, but I want Nvidia to be strong so that ATI has a reason to keep breaking new ground. The 8800 days were great, you could throw anything at a 8800 GTX and it just spited it back out. The 5870 is a great card, but its no 8800, and from the looks of it Fermi is no 8800 ether, (yes I know a 5870 will play most anything at HD+ res but thats more out of a lack of good PC games, then due to it being an amazing card) To me it looks like the next big step forward will be next gen with the Fermi 2 (or whatever they end up calling it) and the HD 68XX.

Fermi is late, really really really late, and that is bad for everyone involved, save ATI, so unless you work for ATI you have many reasons to be rereading this thread every day with the hope that the next post will be word that Fermi is out, it owns everything, and costs so little that ATI was forced to drop its entire line by 100 USD.

Great post CadESin, you made many good points!

In the Rocket Sled demo youtube video, they simulated 500 000 particles for the bridge, at about 5fps. The presenter says you can do millions.

My observations:
1. Fermi is running well. If you can run something that complex, no reason you can't launch.

2. 500 000 particles is almost as high as screen pixel count (2MP for 1080). That's huge!! In Particle Fluid Simulation GTX285 (CUDA) does 30fps with 60 000 particles. 2fps on Core i7.

3. Repeated mention of "CUDA" and "PhysX" instead of DX11. I guess nVidia will continue promoting proprietary path. So, future TWIMTBP games like Batman Arkham Asylum2 will continue to require nVidia card for all the features. AMD users get shafted?

4. Rumours, photos and tapeout of Clarkdale almost 12 months before launch. nVidia DX11 midrange/lowend... canceled?

Quote:

---

Originally Posted by ***Deimos***

In the Rocket Sled demo youtube video, they simulated 500 000 particles for the bridge, at about 5fps. The presenter says you can do millions.

My observations:
1. Fermi is running well. If you can run something that complex, no reason you can't launch.

2. 500 000 particles is almost as high as screen pixel count (2MP for 1080). That's huge!! In Particle Fluid Simulation GTX285 (CUDA) does 30fps with 60 000 particles. 2fps on Core i7.

---

So... very rough estimate 40% (1 - (500000 / (60000*6)) = .39) faster than GTX285 while needing to render a scene and do physics as well? Sounds pretty good. The whole demo looks sweet, but so did whiteout. In fact, I'd say whiteout still looks better than any game on the market and came out with the HD2900. Just shows that demos aren't always the best way to gauge power.

Quote:

---

Originally Posted by hurleybird

So... very rough estimate 40% (1 - (500000 / (60000*6)) = .39) faster than GTX285 while needing to render a scene and do physics as well? Sounds pretty good. The whole demo looks sweet, but so did whiteout. In fact, I'd say whiteout still looks better than any game on the market and came out with the HD2900. Just shows that demos aren't always the best way to gauge power.

---

YES the whiteout demo was amazing, here a link for a bit of a refresh for everyone http://www.youtube.com/watch?v=FL2hnBe6tiE

Based on http://www.youtube.com/watch?v=3-1GMbqzGX0, I am guessing its 3 GF100 cards, one for rendering, one for PhysX/CUDA, and one for Direct X11 Compute and Tessellation, or maybe its all 3 cards working together on everything, there is no way to tell I guess, Thus I agree with hurleybird, I am not sure if that demo is really anything to get that excited about, its cool tech, but if it takes 3 GF100 cards to run its really not going to impact most peoples gaming. Then again do demos like this ever really relate to what we will play this and next year?

http://www.nvidia.com/object/gf100.html is a fun read though to see how Nvidia sees GF100 at least.

ever since the 600$ 3dfx voodoo2 sli decade+ ago.. thats where the fastest video cards have been hoovering around

one fermi beats 1ghz 58xx @ lower price
one fermi beats 1ghz 59xx @ lower price
one fermi beats 1ghz 6000 @ this is the mofo around 600$

is that what you're hoping for? or is that what you know?

Quote:

---

Originally Posted by zalbard

There is no official release date out so far. You have no facts to prove me wrong.
I am simply drawing the conclusion out of the current situation.

---

Official in the sense that even the bottom PR people are commenting on a Q1 '10 release???

I personally thought that the latest they could be was end of '09. While it might have been mentioned, it was nothing that could be traced back to Nvidia. We have video and tech media that have direct quotes from Nvidia PR on a Q1 '10 launch. If they miss that, they are in a world of more hurt than even Charlie could imagine.

Quote:

---

Originally Posted by ***Deimos***

3. Repeated mention of "CUDA" and "PhysX" instead of DX11. I guess nVidia will continue promoting proprietary path. So, future TWIMTBP games like Batman Arkham Asylum2 will continue to require nVidia card for all the features. AMD users get shafted?

---

They made a big deal about Tessellation in some other videos but honestly, DX11 has a much more subtle effect than something like PhysX does which is in your face and easy to explain to the masses. Besides, what company wants to focus on stuff that everyone else can do as well? It's all about highlighting the distinguishing features of your product.

Quote:

---

Originally Posted by LordEC911

Official in the sense that even the bottom PR people are commenting on a Q1 '10 release???

I personally thought that the latest they could be was end of '09. While it might have been mentioned, it was nothing that could be traced back to Nvidia. We have video and tech media that have direct quotes from Nvidia PR on a Q1 '10 launch. If they miss that, they are in a world of more hurt than even Charlie could imagine.

---

Hope will be out for Crysis II ... :p: :D

I joke but nVIDIA is really really late. If they have no GTX 4X0 cards ( with several tech specs etc ...) to introduce in CeBIT that will begin to stink really hard IMHO ...

Quote:

---

Originally Posted by CadESin

Its all bad though, fermi being late and costing a lot when it does come out just hurts PC gaming, Good gaming PC's need to be under 1,000 USD in order for PC gaming to make it back into the mainstream, and right now you can build a good Direct X 10 PC for around that, but throw in a 5850 or 5870 and you break the budget. Fermi needs to come out now, cost the right amount, and work at least haft as well as the fan-boys expect or PC gaming is going to be in trouble.

---

Very true. What worries me is the scalability of Fermi architecture.
Being the top GPU is nice, being a folding monster is great as well, but if mainstream gamers aren't going to be able to enjoy some of these features within 3-6 months after the launch then not many gaming devs will feel like rushing with the new features...

Quote:

---

Originally Posted by CadESin

Good gaming PC's need to be under 1,000 USD in order for PC gaming to make it back into the mainstream, and right now you can build a good Direct X 10 PC for around that, but throw in a 5850 or 5870 and you break the budget.

---

That's what the 5770 and 5750 are for.
They are good and cheap DX11 cards.

Quote:

---

Originally Posted by NapalmV5

ever since the 600$ 3dfx voodoo2 sli decade+ ago.. thats where the fastest video cards have been hoovering around

one fermi beats 1ghz 58xx @ lower price
one fermi beats 1ghz 59xx @ lower price
one fermi beats 1ghz 6000 @ this is the mofo around 600$

---

are you smoking something when posting replies ?

No doubt that Fermi (top model) will be a great performer,but don't get your hopes too high. It will beat single Cypress ,but 5970 will be more than enough to take on it. Even a 1Ghz single Cypress will be competitive in many cases.
Fermi design policy was primarily GPGPU computing capability ,after that gaming. It doesn't help that it is huge and power hungry and that it had design issues(Charlie talked about issues NV had early on with Fermi),and it doesn't help that TSMC's 40nm node is not top notch yields-wise.

Quote:

---

Originally Posted by informal

No doubt that Fermi (top model) will be a great performer,but don't get your hopes too high. It will beat single Cypress ,but 5970 will be more than enough to take on it. Even a 1Ghz single Cypress will be competitive in many cases.

---

Is this a fact or just your speculation ? :)

This can be clearly seen from the leaked numbers so far(can be found on the previous pages of this thread). BTW,nothing is "fact" until the product is launched and tested,so speculation fits perfectly.

Quote:

---

Originally Posted by dengyong

Is this a fact or just your speculation ? :)

---

I think its speculated fact based on the performance shown by fermi in farcry 2 bench.

Quote:

---

Originally Posted by CadESin

YES the whiteout demo was amazing, here a link for a bit of a refresh for everyone http://www.youtube.com/watch?v=FL2hnBe6tiE

---

Quote:

---

Originally Posted by CadESin

Based on http://www.youtube.com/watch?v=3-1GMbqzGX0, I am guessing its 3 GF100 cards, one for rendering, one for PhysX/CUDA, and one for Direct X11 Compute and Tessellation, or maybe its all 3 cards working together on everything, there is no way to tell I guess, Thus I agree with hurleybird, I am not sure if that demo is really anything to get that excited about, its cool tech, but if it takes 3 GF100 cards to run its really not going to impact most peoples gaming. Then again do demos like this ever really relate to what we will play this and next year?

http://www.nvidia.com/object/gf100.html is a fun read though to see how Nvidia sees GF100 at least.

---

That Ruby demo looks good still..and has falling ice, kicking up snow, etc, and guess what? NO PISSX, isn't that crazy?;)

Also, yes I think I read somewhere that the Sled demo was running 3 Fermis SLI'd. So, no, not too impressed.

Quote:

---

Originally Posted by orangekiwii

is that what you're hoping for? or is that what you know?

---

anything less and they can go pack up and go home.. do i make myself clear ?!?

6 months later: 5870 ~ fermi < 5970

whats the point of/in that^ ?!?

Quote:

---

Originally Posted by informal

No doubt that Fermi (top model) will be a great performer,but don't get your hopes too high. It will beat single Cypress ,but 5970 will be more than enough to take on it. Even a 1Ghz single Cypress will be competitive in many cases.
Fermi design policy was primarily GPGPU computing capability ,after that gaming. It doesn't help that it is huge and power hungry and that it had design issues(Charlie talked about issues NV had early on with Fermi),and it doesn't help that TSMC's 40nm node is not top notch yields-wise.

---

dont get your hopes too high on 1ghz 58xx/59xx for they shall soon be six feet deep

Quote:

---

Originally Posted by zalbard

Very true. What worries me is the scalability of Fermi architecture.
Being the top GPU is nice, being a folding monster is great as well, but if mainstream gamers aren't going to be able to enjoy some of these features within 3-6 months after the launch then not many gaming devs will feel like rushing with the new features...

---

True True

A must read:
Inside look at how RV770 got made
http://www.anandtech.com/video/showdoc.aspx?i=3469&p=5
Inside look at how RV870 got made
http://www.anandtech.com/video/showdoc.aspx?i=3740&p=3

Look at the die size comparisons, and read the part about power budgets, and yields (harvesting). In the ultimate quest to be #1, nVidia seems to be making same mistake as last time.

Quote:

---

Originally Posted by informal

This can be clearly seen from the leaked numbers so far(can be found on the previous pages of this thread). BTW,nothing is "fact" until the product is launched and tested,so speculation fits perfectly.

---

Fair enough.

Quote:

---

Originally Posted by informal

No doubt that Fermi (top model) will be a great performer,but don't get your hopes too high. It will beat single Cypress ,but 5970 will be more than enough to take on it. Even a 1Ghz single Cypress will be competitive in many cases.
Fermi design policy was primarily GPGPU computing capability ,after that gaming. It doesn't help that it is huge and power hungry and that it had design issues(Charlie talked about issues NV had early on with Fermi),and it doesn't help that TSMC's 40nm node is not top notch yields-wise.

---

Agreed, I'd put fermi just above the 5870 in games. And as always, nVidia will do better in some games, ATI will do better in a few others. tesselation will give a big advantage to nVidia but games that don't use it, or only use it a little bit will be very close.

/speculation, but based on evidence

Quote:

---

Sacrificing performance to meet the 16mm x 16mm die size targets wasn’t going to happen, but what ultimately convinced Carrell to go with a larger die this time around was something that ATI didn’t get nearly enough praise for: the ability to launch 4 different 40nm DirectX 11 GPUs in less than 6 months.

Remember that Carrell’s beef with building the biggest GPU possible is that it takes too long for the majority of customers to get access to derivatives of that GPU. Look at how long it took G80 or GT200 to scale down. And who knows when we’ll see $150 Fermi/GF100 derivatives.

---

Quote:

---

Engineering eventually figured a way to fix most of the leakage problem through some changes to the RV740 design. The performance was still a problem and the RV740 was mostly lost as a product because of the length of time it took to fix all of this stuff. But it served a much larger role within ATI. It was the pipe cleaner product that paved the way for Cypress and the rest of the Evergreen line.

As for how all of this applies to NVIDIA, it’s impossible to say for sure. But the rumors all seem to support that NVIDIA simply didn’t have the 40nm experience that ATI did. Last December NVIDIA spoke out against TSMC and called for nearly zero via defects.

---

imagine the sly smiles on ATI engineers faces when they hear nVidia is having same problems that they did - which the ATI guys already fixed months ago.

Quote:

---

Chen's speech included a call to "my friends at TSMC to give me more 40 nm parts," and a plea for improved via defectivity. D(i)ck James, a technology analyst at Chipworks (Ottowa, Canada), said via defects have shown up on ICs manufactured by TSMC. Chipworks has inspected products from graphics vendor ATI, now part of Advanced Micro Devices (AMD, Sunnyvale, Calif.). "The problem appears to be that when they cut a via, a residue of photoresist gets on the edge of the via, which creates a ring-shaped discontinuity in the metal," James said. "The discontinuity could create electromigration issues. We've seen the same problem on the upper metal levels on the ATI chips we've studied. It creates a reliability failure mode."

---

engineers dont spend their time laughing at the competition, they do their jobs. chipworks is a company that analyzes competitors products.

Quote:

---

Originally Posted by NapalmV5

dont get your hopes too high on 1ghz 58xx/59xx for they shall soon be six feet deep

---

must be some strong stuff :yepp:

This is a couple weeks old but a good article from Damien.
http://www.behardware.com/articles/7...evolution.html

Quote:

---

Originally Posted by eric66

must be some strong stuff :yepp:

---

Yeah... will they work? As these don't yet :down:

Quote:

---

Originally Posted by NapalmV5

anything less and they can go pack up and go home.. do i make myself clear ?!?

6 months later: 5870 ~ fermi < 5970

whats the point of/in that^ ?!?

---

So NVIDIA should just say "sorry dudes and dudettes, but Fermi won't beat HD 5970 so we ain't gonna relase it"? What would be the point of that?

Fermi might not be the new messiah for gamers, but as quite few already pointed out Fermi has unique HPC-capabilities.

A product that might not sell well to gamers but do "quite" good in HPC-applications is better than no product at all, don't you think?

Sometimes you've got to release what you got, no matter what. Don't you think ATI didn't like what R600 turned out to be? RV670 was basically R600 done right but wasn't really faster. Should they not have released RV670 because it was late and not fast enough?

Wake up NapalmV5! WTH happened to you?!

Quote:

---

Originally Posted by hurleybird

So... very rough estimate 40% (1 - (500000 / (60000*6)) = .39) faster than GTX285 while needing to render a scene and do physics as well? Sounds pretty good. The whole demo looks sweet, but so did whiteout. In fact, I'd say whiteout still looks better than any game on the market and came out with the HD2900. Just shows that demos aren't always the best way to gauge power.

---

Yeah, but you also have to consider that these particles may be rendered differently. They talk about all the different forces they are taking into account...

Idk how much that effects it, but just wanted to throw that out there. :D

Quote:

---

Originally Posted by highoctane

I gotcha, it really does seem like they fell asleep at the wheel in regards to getting any dx11 hardware out the door. Both ATI and Nvidia seem to release the latest on the highend part first before releasing the lower parts so with gt300 delay's you would have thought some midrange to lower end parts would have been scheduled to release by now.

---

i dont think they fell asleep, they didnt think dx11 was that important, or, most likely, didnt think dx11 matters all that much since the gaming industry has slowed down. they probably checked the plans of major game devs and most of them said they will focus on dx10.1 at most, if not even dx9 cause thats what consoles are stuck at for now, so they decided not to focus on dx11 yet...

and i think they are right... dx11 doesnt matter all that much and isnt necessary yet... BUT that doesnt mean they dont need it... its very important for marketing... who will buy a new 10.1 or 10.0 vga 2 years after he already bought a 10.0 vga? it either has to be cheaper or notably faster than the old card... for most people perf isnt all that important since games arent as demanding right now, so dx11 does matter as a selling point.

and i think nvidia underestimated how many game devs ati can convince to actually adopt dx11, even if its only patched on...

Quote:

---

Originally Posted by NapalmV5

ever since the 600$ 3dfx voodoo2 sli decade+ ago.. thats where the fastest video cards have been hoovering around

one fermi beats 1ghz 58xx @ lower price
one fermi beats 1ghz 59xx @ lower price
one fermi beats 1ghz 6000 @ this is the mofo around 600$

---

itll take you a long time to regain some credibility in the eyes of everybody who read this post :P

Quote:

---

Originally Posted by LordEC911

We have video and tech media that have direct quotes from Nvidia PR on a Q1 '10 launch. If they miss that, they are in a world of more hurt than even Charlie could imagine.

---

im not sure... i think they have until Q2 to come out with fermi... if they still havent launched fermi properly (shipping in volume) in the middle of this year, then LOTS of investors will jump ship... cause thats what nvidia told all the investors really matters, and most fell for it... they all believe in the fairy tales of gpgpu and the huge margins and profits nvidia supposedly will cash in on with fermi. and nvidia promised fermi tesla cards in mid 2010, they never changed that date, they said mid 2010 even at the end of 2009 already, and they stuck to it... but if they miss THAT date, then their stock price will take a major hit...

Quote:

---

Originally Posted by NapalmV5

dont get your hopes too high on 1ghz 58xx/59xx for they shall soon be six feet deep

---

riiiight
just like G92 has been 6feet deep since gt200 launched...
oh wait, gt200 was shrunk to 55nm, its price was cut in half, it was shrunk to 40nm, had dx10.1 glued on, cut in price again and G92 is still a kick4ss gaming card and unrivaled in price perf... :P

Quote:

---

Originally Posted by Chumbucket843

engineers dont spend their time laughing at the competition, they do their jobs. chipworks is a company that analyzes competitors products.

---

if they have a sense of humor, they sure do :D

FYI, i asked around regarding fermis @ cebit and didnt get a single reply...
Looks like there wont be any fermis at cebit, at least not in public...
they will probably do another deep dive like event or something like that... i just wonder if it will be a 100% repeat of the event in vegas or if there is anything new... at all... :shrug:

they better show something new...

http://www.semiaccurate.com/2010/02/...ding-examined/

looks like the financial markets know something we dont?
they seem to have lost most of their faith in nvidia and expect "ok" to bad results for 2010...

nvidia really cut their rnd budget? didnt they just brag about how much they spent on rnd over the past years and that they are investing more and more?

if they cut rnd it sounds like they are definately not working on a new architecture but are refining fermi... they are probably preparing a 32nm or 28nm fermi?

Quote:

---

Originally Posted by saaya

FYI, i asked around regarding fermis @ cebit and didnt get a single reply...
Looks like there wont be any fermis at cebit, at least not in public...

---

I hope to see some behind the booth action (with some free grog :p) would be a major letdown for me if there weren't any cards on display

Quote:

---

Originally Posted by Leeghoofd

I hope to see some behind the booth action (with some free grog :p) would be a major letdown for me if there weren't any cards on display

---

me2... if i see something i will def let you know :D
pm me your mobile no and ill send you a text :D heheh

Quote:

---

Originally Posted by saaya

http://www.semiaccurate.com/2010/02/...ding-examined/

looks like the financial markets know something we dont?
they seem to have lost most of their faith in nvidia and expect "ok" to bad results for 2010...

nvidia really cut their rnd budget? didnt they just brag about how much they spent on rnd over the past years and that they are investing more and more?

if they cut rnd it sounds like they are definately not working on a new architecture but are refining fermi... they are probably preparing a 32nm or 28nm fermi?

---

yeah true they ont need rnd if they only need a shrink.. might be a good sign that they HAVE finished fermi and just need to get it out their now

Read page 9 of Anand's excellent article on Cypress:
http://anandtech.com/video/showdoc.aspx?i=3740&p=9

Maybe TSMC was the straw that choked Fermi.

Quote:

---

Originally Posted by saaya

looks like the financial markets know something we dont?
they seem to have lost most of their faith in nvidia and expect "ok" to bad results for 2010...

---

Take an objective look at NVIDIA's position today:
- GT200b is pretty much EOL right now.
- NVIDIA exclusive partners are making funny products: EVGA's GTX275 coop (GT200b + G92b), Galaxy dual G92b.
- Pretty much all mobile GeForce 2xx have been renamed to GeForce 3xx, I suspect Fermi based mobile parts will not come so soon.
- The only good news from NVIDIA lately is Tegra 2 (which is quite cool) and the mobile Optimus technology.

With no solid product, I simply cannot see how NVIDIA can be profitable this quarter or even the next. 2010 does indeed look bad for NVIDIA.

Quote:

---

Originally Posted by Jamesrt2004

yeah true they ont need rnd if they only need a shrink.. might be a good sign that they HAVE finished fermi and just need to get it out their now

---

But why would big green do that? They could shout "full steam ahead" for the next architecture. A Fermi-shrink might be a (early) 2011 card but what comes afterwards? Cutting R&D isn't a good idea imho.

I find it ironic how clearly nvidia forgot what made them a successful company. If memory serves me correctly ATI's strategy is the one that nvidia used to bludgoen 3dfx into bankruptcy because 3dfx was always late; always missed targets and had features people didnt want at a price that wasn't competitive.

I have strong suspicious that Fermi will never be made in volume in the consumer market and we wont see a real wide spread DX11 chip from nvidia until summer. Reason I suspect this is because of how quiet they have gotten all of a sudden about its release....unusually so.

No leaked benchmarks of any substance, and only a handful of "reference" boards. I mean hell we havent even seen a PCB shot yet; what's up with that?

NV finally found final CLOCKS of GTX 470/480. Cards are ready to launch, will be in second half of March in reviews. There is one word: Gemini, when GF100 will be launched, remember for that word! Bye fans!

Quote:

---

Originally Posted by Zed_X

NV finally found final CLOCKS of GTX 470/480. Cards are ready to launch, will be in second half of March in reviews. There is one word: Gemini, when GF100 will be launched, remember for that word! Bye fans!

---

Less then a month to go, time flies.

Gemini? You mean dual GF100?

If I remember correctly GeCube used "Gemini" for their Dual-GPU-cards, didn't they?

rumor goes around:
199euro for GTX 470
299euro for GTX 480

Quote:

---

Originally Posted by blob

rumor goes around:
199euro for GTX 470
299euro for GTX 480

---

Source of rumor?

Quote:

---

Originally Posted by blob

rumor goes around:
199euro for GTX 470
299euro for GTX 480

---

Add another 100€ and you may find the true prices.

link
Check his track record

Quote:

---

Originally Posted by blob

rumor goes around:
199euro for GTX 470
299euro for GTX 480

---

Rumor goes around:

the Cake is a Fermi!




Sorry, but these prices are highly unlikely imho. I don't think NVIDIA is giving away Fermi that cheap, unless GF100 really does suck balls...

Source of the info:

Quote:

---

Ik kan het verder ook niet helpen, deze informatie komt rechstreeks uit de prijslijst van een nVidia partner.

---

"I can't help myself either, this information comes directly from a pricelist of one of NVIDIA's partners." (Correct me if I'm wrong)
http://gathering.tweakers.net/forum/...71807#33471807

(In the dutch thread some are doubting the numbers as well from what I can read and understand)

I concur, the Cake is a Fermi! (please note the capitalization for future references ;) )

Maybe it's the price for Fermi per chip, not board? If it's chip price, then it's OK I guess.

Why is "Cake" written in capitals here? Sorry, English is not my mother tongue :( But thanks for correcting me :)

That was just a joke. The CAKE in Portal (game) is sacred ;)

It's the price of the Wooden Screw edition.

Gemini = dual fermi AFAIK

i think somebody just took the us dollar price and calculated what that is in euros... unfortunately thats not how it works...
whatever something costs in us$ in the us is usually what it costs in europe in euros... :D

200E=275us$
300E=400us$

the real euro price would then be around 275E and 400E...
even that sounds a bit low...

Add 50% more and that'd be about right, yeah.

If it's $400, it will be 400€ in Europe, at least at launch.

Quote:

---

Originally Posted by FischOderAal

So NVIDIA should just say "sorry dudes and dudettes, but Fermi won't beat HD 5970 so we ain't gonna relase it"? What would be the point of that?

Fermi might not be the new messiah for gamers, but as quite few already pointed out Fermi has unique HPC-capabilities.

A product that might not sell well to gamers but do "quite" good in HPC-applications is better than no product at all, don't you think?

Sometimes you've got to release what you got, no matter what. Don't you think ATI didn't like what R600 turned out to be? RV670 was basically R600 done right but wasn't really faster. Should they not have released RV670 because it was late and not fast enough?

Wake up NapalmV5! WTH happened to you?!

---

You are right about the parallels to R600.. big, power hungry, late and overhyped.

In the end though, regardless what the outcome of Fermi is, the market will move on to new products.

Already 2 years ago nVidia engineers created architecture for DX11 successor, and right now they are designing it with lessons learned from Fermi.

But, nVidia's make chips bigger and bigger mentality is becoming their acheles heel. As the inside look into AMD showed, changing course (convincing EVERYONE to do things differently) is difficult. 90-95% chance that GT400/"Fermi2" will be big monolithic design... probably 28-32nm shrink of Fermi.

After that, in 2012, (if Fermi doesnt do it) nVidia might be first to break 645.16 mm^2 barrier. ie a chip so big its 1 square inch, on a wafer thats only 37.7 inch^2.

JUST IN CASE BACK IN SEPT 30, 2009 YOU WEREN'T SHOCKED THAT FERMI IS 3BT.. some perspective:
45nm Phenom II 6MB L3 = 258mm2
45nm Core i7 8MB L3 = 263mm2 - tiny in comparison ;)
45nm 6core Dunnington 16MB L3 (Core2) = 503mm2
45nm 8core SMT POWER7 32MB eDRAM = 567mm2
65nm 240SP GT200 = 576mm2 - redefining ridiculously huge
45nm 8core SMT Tukwila Itanium 24MB L3 ~700mm^2 and 2BT.
40nm 512SP Fermi = ?? record die size ?? - 3BT is record!!??

BTW: if anandtech Inside RV870 article is true about TSMC 40nm, bad news for Fermi. Via defects, and doubling to compensate, and having more redundant logic/ALU will bloat an already enormous die size. Likewise transistor channel length varience bigger problem for Fermi since chip already hot enough.

Quote:

---

Originally Posted by saaya

i think somebody just took the us dollar price and calculated what that is in euros... unfortunately thats not how it works...
whatever something costs in us$ in the us is usually what it costs in europe in euros... :D

200E=275us$
300E=400us$

the real euro price would then be around 275E and 400E...
even that sounds a bit low...

---

Quote:

---

Originally Posted by zalbard

Add 50% more and that'd be about right, yeah.

---

Quote:

---

Originally Posted by jfromeo

If it's $400, it will be 400€ in Europe, at least at launch.

---

Are you guys talking about these prices with a competing GPU from ATI or without it? You guys have to clarify this, because the price is always directly affected by competition.

If Fermi beats the performance of current ATI-line up, an ATI just watch it (with those big plans for 86x-series which comes later , and even bigger and later plans for 28nm :p:) then a superior Fermi won't be cheap, for sure.
But if ATI can match the Fermi-performance with a refresh (or new GPU) at the same time, then we are looking at good prices and cheap Fermi.
What scenario is the base of your price speculation guys?

Assuming 5870 < GTX480 < 5970 performance wise, right?
Plus it's new, expensive to make, and features sweet GPGPU all cool kids totally must have.
So I suppose it'll be priced around $600 at launch, $150 more expensive than 5870 (and 50% higher than the proposed 300 Euro = $400).

From previous releases over the years I think your estimate is very accurate (at least given the current compatition from AMD/ATI). In my opinion I wouldn't be surprised if the GTX480 launches at a slighty higher retail price, somewhere closer to $649+-?

Recent HD5450 got me thinking about the enormous and still growing gap between low and high end.

Low end shader performance improvement from 2002->201: 4 (9500) -> 16 (HD5450)... 4x

Code:

---

R9600        4 ROP/TMU        4 PixelShader
X300        4 ROP/TMU        4 PixelShader
X1300        4 ROP/TMU        4 PixelShader
HD23/24        4 ROP/TMU        8 5wide SP
HD34xx        4 ROP/TMU        8 5wide SP
HD43/45        4 ROP 8 TMU        16 5wide SP
HD54xx        4 ROP 8 TMU        16 5wide SP

---

High-end shader performance improvement from 2002->2010: 8 (9700) -> 640 (HD5970)... 80x!!

Anybody who thinks Fermi will cost less than what G200's cost (ie those 299 euro folks) is totally dellusional. Put down the kool-aid/mushrooms/acid. Unless as I said long ago, nVidia starts drinking Sony's kool-aid and starts "giving away" Fermi at a loss to try to reach bankruptcy ASAP.

PS: 199 euro FERMI at launch - WTF how ridiculous is that!! Its 21 fricken months after G200 launched, and a crippled GTX275 is US$250-$300.. if you can find it all, since pretty much discontinued (EOL).

PS2: Remember HD5870 is $379 MSRP ($400 in stores), but Fermi is at least 50% bigger, and %50 more memory. Think about it.

Quote:

---

Originally Posted by zalbard

Assuming 5870 < GTX480 < 5970 performance wise, right?
Plus it's new, expensive to make, and features sweet GPGPU all cool kids totally must have.
So I suppose it'll be priced around $600 at launch, $150 more expensive than 5870 (and 50% higher than the proposed 300 Euro = $400).

---

Assuming 5870 < GTX480 < 5970 performance.
In GPU-world 20% more performance means at least 60% more price for the top-cards. (ATi is not adding 20% to the 5850 price for calculating the price of 5870 either). So based on that assumption, if GTX 480 is ~20% better than 5870 then you are right on spot, but if it can match the 5970-performance and ATi just watch it (with those big plans for 86x-series which comes later , and even bigger and later plans for 28nm ), then I'm afraid that you have to recalculate those numbers, big time.

Quote:

---

Originally Posted by ***Deimos***

40nm 512SP Fermi = ?? record die size ?? - 3BT is record!!??

---

It's already been confirmed that the die size is somewhere between the 65nm GT200 and the 55nm GT200b.

Quote:

---

Originally Posted by Sam_oslo

but if it can match the 5970-performance

---

It didn't even in NVIDIA's biased demos...
And 600$ is minimum (ATI's prices might drop), if it really rocks it can be priced as high as 700-800$...

Quote:

---

Originally Posted by zalbard

It didn't even in NVIDIA's biased demos...
And 600$ is minimum (ATI's prices might drop), if it really rocks it can be priced as high as 700-800$...

---

I don't think a a price-drop will help at all, how cares about a second grade outdated GPU anyways? ATi needs to have a card to compete on performance (it can be either 5970, or a refresh/new GPU) that can match/beat Fermi, otherwise Fermi won't be cheap.

Fermi won't be cheap no matter what happens. :p:
Nvidia isn't doing a giveaway after working so hard and for so long, not to mention the size of the thing... At least they can sell a lot of cool key chains with their supposedly poor yields! :D

I don't agree. ATi is going to play a big role on the price of Fermi, just as nVidia did in last round. ATi had no competing GPU and you see what happend to 5-series prices?

I hope ATi manges to put up a fight now (not with 68-series in 6 months or 28nm in next year) because if these guys are avoiding to compete and beat each other, then we all are going to suffer.

Well, IMO cheap (for an ultimate high end gfx card) is under $499. Do you believe in that? At least within the first month after the launch. I don't.

Quote:

---

Originally Posted by Zed_X

NV finally found final CLOCKS of GTX 470/480. Cards are ready to launch, will be in second half of March in reviews. There is one word: Gemini, when GF100 will be launched, remember for that word! Bye fans!

---

:toast:

:up:

Quote:

---

Originally Posted by zalbard

Fermi won't be cheap no matter what happens. :p:
Nvidia isn't doing a giveaway after working so hard and for so long, not to mention the size of the thing... At least they can sell a lot of cool key chains with their supposedly poor yields! :D

---

Both 256bit boards. Both 1GB GDDR5. Both <200W.
RV870 die size about 25% more than RV770.

HD5870/50 launch price: $379, $259 - AMD's 2nd test on defect prone 40nm
HD4870/50 launch price: $299, $199 - very mature 55nm, exellent yields

If it costs AMD more to make 40nm chips, why would you think nVidia would be an exception? All those TSMC issues are much worse for a big chip.

I would be shocked if by Sept'10, nVidia had good availability and didn't rely extensively on cutdown models.

nVidia future options:
1. different fab? unlikely.. at least 6-9 months for a product.
2. different node.. ie 32nm.. more redesign/floorplanning required.. at least 6-9 months.
3. do like "9600GT": make smaller 320 SP (10 clusters) or even 384 SP (12 clusters) chip using 256bit bus.
75% of SP, 75% bus.. other trimmings.. bam.. ~330mm2 chip like Cypress.:up:

Quote:

---

Originally Posted by zalbard

Well, IMO cheap (for an ultimate high end gfx card) is under $499. Do you believe in that? At least within the first month after the launch. I don't.

---

The age av $499 high-end GPU started already last year, because we had a historical competition, and ATi had finally manged to beat the nVidia-performance. But nVidia chose to "jump over" the GPU who should compete with 5-series, and then ..... you look at the 4870x2 vs 5970 launch-prices and you tell me what append?.

I hope ATI and Nividia has not agreed on a "jump-over" game, because that's the only game that can bring back the $600-GPU from the death.

Quote:

---

Originally Posted by Zed_X

NV finally found final CLOCKS of GTX 470/480. Cards are ready to launch, will be in second half of March in reviews. There is one word: Gemini, when GF100 will be launched, remember for that word! Bye fans!

---

Gemini is the dual GF100 card, that was leaked a while ago :D

Can't wait!!!

Quote:

---

Originally Posted by Sam_oslo

The age av $499 high-end GPU started already last year, because we had a historical competition, and ATi had finally manged to beat the nVidia-performance. But nVidia chose to "jump over" the GPU who should compete with 5-series, and then ..... you look at the 4870x2 vs 5970 launch-prices and you tell me what append?.

I hope ATI and Nividia has not agreed on a "jump-over" game, because that's the only game that can bring back the $600-GPU from the death.

---

death of $600 video card? maybe by $1.
MSRP of 5970 is $599.

Lack of competition and availability:
newegg avg price $700.

Only $830 8800 ULTRA was more.
Even dual G200, GTX295 MSRP was **only** $499 (IMHO nVidia partners complained about losing A LOT of money on each one compared to GTX280/275/260).

Quote:

---

Originally Posted by ElSel10

Gemini is the dual GF100 card, that was leaked a while ago :D

Can't wait!!!

---

yeh gemini goes back to november

but "gemini" dont always mean dual lol

then again maybe it is dual fermi :)

Quote:

---

Originally Posted by NapalmV5

yeh gemini goes back to november

but "gemini" dont always mean dual lol

then again maybe it is dual fermi

---

Stop...posting...rubbish.... :p:

Quote:

---

Originally Posted by ***Deimos***

death of $600 video card? maybe by $1.
MSRP of 5970 is $599.

Lack of competition and availability:
newegg avg price $700.

Only $830 8800 ULTRA was more.
Even dual G200, GTX295 MSRP was **only** $499 (IMHO nVidia partners complained about losing A LOT of money on each one compared to GTX280/275/260).

---

You see that $599-monster came back from the death just because nVidia "jumped over" the GPU that should compete with it. If you remember, ATi had announced $499 for 5970 to start with, But the lack of competition not only brought back that $600-monster from the death, but also ended up selling for over $700 too.

So, the conclusion is AMD's high prices are nVidia's fault for not delivering competition.

Sounds very logical.

Quote:

---

Originally Posted by Luka_Aveiro

So, the conclusion is AMD's high prices are nVidia's fault for not delivering competition.

Sounds very logical.

---

Exactly. When Nvidia prices its cards high because of a lack of competition, they are greedy monsters out for the money in our pockets. But when AMD does the same this is only fair because NV couldn't put up a competitive product.

http://i47.tinypic.com/21mzp89.jpg

So true annihilat0r ... :yepp:

LOL, nobody does charity work for nerds these days.
Prices are based on competition, and without a proper completion everybody will try to squeeze the most out of us, and they call it business, you know.

Quote:

---

Originally Posted by Tim

Stop...posting...rubbish.... :p:

---

you guys...just dont...get it.... :p:

read zedx post above a few times

dont worry itll come to you come march

till then :toast:

Quote:

---

Originally Posted by annihilat0r

Exactly. When Nvidia prices its cards high because of a lack of competition, they are greedy monsters out for the money in our pockets. But when AMD does the same this is only fair because NV couldn't put up a competitive product.

---

People are more sympathetic with AMD simply because AMD probably needs the money more than Nvidia

Could be that he meant Gemini the Zodiac sign. In which case, Gemini would be May 21 to June 20.