GF100 (Fermi) previews and discussion

Page 7 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.
Status
Not open for further replies.

I was being sarcastic and disturbing 😀
 

I realised that and I feel I may require a substantial amount of expensive therapy to suppress the images that your post conjured up, but then at the same time I was also thinking of the Mars Edition and the soiled undergarments that seemed to cause (then and now).
 

Its not the finest creation in the land , but if it carries on as a disturbance at least its lasting , man can only hope that he achieves the goal of a lasting creation. You know if this newly aquired symptom passes in any short while ill be back at the drawing board pulling my hair out and yelling at everything. :lol:
 
Fermi looks good, but ATI has a huge head start on Nvidia. So many people (including me) have already bought HD 5000 cards, I doubt they will spend another $300+ on Fermi so soon. I just bought an HD 5850, and I'm going to keep that card for as long as possible.
 
Thats nVidias fault tho, by their late release. They need Fermi to be compelling, but it doesnt look to be a deal breaker yet.
Knowing an answer is right around the corner from ATI, if even needed, it all comes down to pricing, where nVidia has a disadvantage there
 
i stumbled on this on toms couple of day ago and they quickly removed it in a matter of minutes.. i prey that this is true and it goes with mine asumptions witch have always bean +,- 90% corect. it would bee nice that this is true and i bet it would.




 
Funny, I first saw here as well, in and from this forum as well, and then reality hit in, and successful troll was found to be not so successful, as these have long been found to be faked, and no one yet has any HW from nVidia.
As W1zzard has said, hed be surprised that nVidia has even made 400 cards to date, and thats including the first 2 steppings, tho the first stepping was to have been less than 10 according to rumors.
nVidias partners currently have 0/zero/none/nada chips yet, so, in order to have made one, itd have to have come straight from nVidia itself, and they just dont do that, not on unfinished chips.
Latest rumors are, Revision A3 wont be seen other than by review sites, and maybe a handful out to other sources, but the true retails will be B1 or some such
 



I did not said that this is genuine or not.this is fermi tread so i just posted what i found nothin more.if this is fake then autor of this has no integrity and honor.i hope that i will see performance atleast as that in regard of deleying and waiting of fermi and than we can sit back and enjoy in price wars that will benefit all of us,fanboy or not and we can see that without competition ati is raising their prices witch suprise me because ati has always bean afordable and now they play nvidias role with prices,odd,and that only tell us that they are the same in the game off slaping the money from our pocket.
 
Well, heres some more credible infos:

The Pro version is the oil cooled one, which will end up with the four 5970 GPU's oil cooled and a Fermi GPU also oil cooled. According to my supplier the Fermis are buggers for heat. Even the single entry level GPU will run 200 watts.
http://milkyway.cs.rpi.edu/milkyway/forum_thread.php?id=1508&nowrap=true#35947

So, it appears the lower end 360 will be pulling 200 watts, that makes the 380 at least 225
 
new psu here i come or make it two psu-s :) just kidding,as long it performs right i dont mind with temp or how much will they pull juice from psu i never take power consumption in my calculation for new build only performance but as this global crysis crunch my wallet price is definetly in my calculation from now on.people never should have invented money... :)
 

The price for a IN STOCK 5970 = 800......But you all knew that.
You use this line, go away alot, I'm not your mother asking you to clean your room. I'm showing you proof that ATI will gouge you just as much as Nividia might for a video card.
800 dollars when they told reviewers it would be 600. Thats worse than jacking the price up on the 5850 to 320 from 269.00. But ATI is for the people, yeah right.
 
but as this global crysis crunch my wallet price is definetly in my calculation from now on

I hear that. I really wanted a Core i7-940 and HD 5870 in my new PC, but this economy is too tight to spend that kind of money. People in my area are getting laid off left and right. Ah well, the Phenom II 945 and HD 5850 will keep me gaming for at least 2 years.
 

Sounds pretty bitter here. Were you one of the "lucky" G260/280 early buyers than spent shedloads on them, only to see the price plummet, and when nVidia offered rebates, you werent eligible?
I dont care how it performs, if people are claiming pricing is high on ATI, if its Halo, it has no competition, just like the 5870, so, it shouldnt be higher than the 5870s highest price, as there may still be a x2 part coming from nVidia for their top card, ala the 5970
 

the 5970 is high for any card but look at the i7 extreme , thats how they profit and AMD hasnt been able to get that kind of margin in a long time; I say more power to them because when fermi comes out AMD will be back t smaller margins.
 
Ummm, no.
The margins are thin in this market, and the silicon costs enter in in a large way.
Having 50% more silicon, and lower yields doesnt make a great place to have low pricing. Especially on such low yields. Look at the 285 pricing as an example, and how hard they are to find, whereas the dumpoff 260s are plentiful, and nVidias margins are cut severly.
Ive been warning people of all these percieved price cuts from ATI to maybe slow it down some, as I dont see large cuts, smaller ones certainly, but also, I remind you, the 5 series is priced higher than the 4 series ever was, and lowering prices would only get them into those release ranges for the 4 series. And thats what I expect.
nVidia is in no position to have a pricing war, and so far, Fermi doesnt look that compelling.

What nVidia fans have been saying over and over, theres no DX11 games, so why buy a ATI DX11 card also holds true with Fermi, and IF it has a somewhat better solution for DX11/tesselation (currently, all we have is nVidias PR and running synthetics), theres still no games for them, and as far as perf goes, theyre soo far behind in getting to the shelves, many of their sales are gone anyways, and a ATI refresh is right around the corner, add in the low yields to actually get some marketshare, and nVidias hurting, no matter what color you paint it, even if its green
 

you are saying tsmc still has yield issues even after the president denied it?

I realize several places are talking 40nm woes but you need to also realise that GF is contracted to handle a certain portion of ATI lith, and the TSMC president , to my knowledge has not confirmed low yields claims.
 
No, theyre slowing getting it right, but its also held nVidia back, and trying to do a 3.2 billion tranny chip for high yield, mass production, on a brand new arch, its hurt them, and theyre having their own problems besides.
And yes, the 40nm is hitting the wall, without HKMG, so this is why IMO all these bad yields from TSMC.
But more to the point, its nVidias problem of doing such a chip, plus the TSMC problems and delays.
The pace of the gfx market dictates a quicker response than what nVidias shown here, and unless perf is incredible, it wont save them, and if it is incredible, ATIs margins will be fine, as nVidias pricing will be high, and if it isnt incredible in perf, itll only be competitive, and having such higher costs, nVidia wont be able to price themselves competitively, so again, prices wont be that low IMO.
So, either way, ATIs pricing is pretty solid from their POV, with only slight movements downwords.
Also, regarding just the 5970, weve seen nothing to show Fermi will topple this card, only the 5870, which is expected
 
president of tsmc reported that all 40nm yields problems are solved so there are not going to be problem with yield(bugs and low perf) and availability with fermi cards and thats legit info-news.
 
I think that delaying is the result of that they had to cut some of the original fermi cpu parts because 40nm is little tight for what they intended to put in(512bit+higher clocks 1ghz cpu 2ghz cuda and + few stuffs) but i think that original fermi is stil waiting 32nm and that is gona be out in december 2010\2011 and i think that we will have a tough competition when this 40nm fermi is out from what i read within white paper and the fact 2,3 x faster with antialiasing 8x than gtx285 and 8x computional power but that is if the exact described actual product from white paper comes out and i mean 100% exact with no hickups and bugs.
 
Well, the AA is supposedly 2.33 times, and who knows what that means, as G200 did poorly at 26x16 res with 8xAA, and didnt have the DX10.1 1 less pass, which makes it 10-20% faster already.
In other words, add 20% to Fermis, plus 60% perf, plus a better high res solution for 8xAA, and you can see how theyd get 2.33 times better.
As for the DP compute, thats only on Tesla, as theyve removed the DP ability for the gfx desktop cards, and isnt needed for games
 
Status
Not open for further replies.