Monday, December 14th 2009

NVIDIA Pitches GeForce GTX 300 Series to Clinch Performance Crown

NVIDIA's latest DirectX 11 compliant GPU architecture, codenamed "Fermi," is getting its first consumer graphics (desktop) implementation in the form of the GeForce GTX 300 series. The nomenclature turned from being obvious to clear, with a set of company slides being leaked to the media, carrying the GeForce GTX 300 series names for the two products expected to come out first: GeForce GTX 380, and GeForce GTX 360. The three slides in public domain as of now cover three specific game benchmarks, where the two graphics cards are pitted against AMD's Radeon HD 5870 and Radeon HD 5970, being part of the company's internal tests.

Tests include Resident Evil 5 (HQ settings, 1920x1200, 8x AA, DX10), STALKER Clear Sky (Extreme quality, No AA, 1920 x 1200, DX10), and Far Cry 2 (Ultra High Quality, 1920x1200, 8x AA, DX10). Other GPUs include GeForce GTX 295 and GTX 285 for reference, just so you know how NVIDIA is pitting the two against the Radeon HD 5000 GPUs, given that the figures are already out. With all the three tests, GTX 380 emerged on top, with GTX 360 performing close to the HD 5970. A point to note, however, is that the tests were run at 1920 x 1200, and tests have shown that the higher-end HD 5000 series GPUs, particularly the HD 5970, is made for resolutions higher than 1920 x 1200. AA was also disabled in STALKER Clear Sky. NVIDIA's GeForce GTX 300 will be out in Q1 2010.

Update (12/15): NVIDIA’s Director of Public Relations EMEAI told us that these slides are fake, but also "when it's ready it's going to be awesome".

Source: Guru3D
Add your own comment

189 Comments on NVIDIA Pitches GeForce GTX 300 Series to Clinch Performance Crown

#2
buggalugs
Nvidia can kiss my ass.:nutkick:
Posted on Reply
#3
laszlo
the slides taken from guru maybe fakes but the numbers used in it for the existing cards are real and confirmed by existing reviews;all we don't know is if fermi numbers are invented or not

i expect fermi to be better than cypress but i don't know the% so we must wait....
Posted on Reply
#4
Amok
It's not hard to collect numbers from existing reviews, make up some other numbers and make a powerpoint slide to resemble an nvidia official slide.


But, important point is that i never saw an official Nvidia or Ati slide to contain actual numbers, they only contain percentages.

Again, examples of official PR slides.






They never include actual performance numbers, just percentages.
Posted on Reply
#5
Imsochobo
sapetto said:
Ha again?! Lol ATI also showed similar slides and they were not very correct (in the slides 5870 outperforms GTX295).
And it does.

In fact, by a large margin in diffrent games, i would say 5870 is about on PAR with 295 with the latest driver.





I don't recall any of those games being more ati friendly than nvidia friendly.
Posted on Reply
#6
the54thvoid
I'm new here but have been an avid techpowerup watcher. These slides are not the real deal. I recall a great many bench reviews where Stalker has ATI cards outperfomring NV quite competently. A number of sites reviewed show the 5870 beat the 295. Toms Hardware is one of the exceptions.
You still get biased hardware sites that favour one camp. I had a 295 and ditched it for 2 5850's. I can speak for both camps. Based on the reviews i read before purchasing my 5850's (i read practically every english review on the web - i'm that careful) these benchies are fabricated. Stalker is the key here - NV performs poorly in clear sky - it's an ATI game. Here are some reviews that contain a 5870 beating a 295 in other games than stalker.
http://www.guru3d.com/article/radeon-hd-5870-review-test/18
http://www.guru3d.com/article/radeon-hd-5870-review-test/16
http://www.hardwarecanucks.com/forum/hardware-canucks-reviews/23415-sapphire-radeon-hd-5870-1gb-gddr5-review-16.html
http://www.legitreviews.com/article/1080/6/
and the most amusing from this very website you amnesic puppies
http://www.techpowerup.com/reviews/ATI/Radeon_HD_5870/20.html (for Stalker SOC)
http://www.techpowerup.com/reviews/ATI/Radeon_HD_5870/21.html (for stalker clear skies)

So, before the camps go to war, do your research... and beware false prophets.
Posted on Reply
#7
Csokis
Hmm... DX11 bench? :confused: :rolleyes:
Posted on Reply
#8
InnocentCriminal
Resident Grammar Amender
It's marketing guff - end of. It's when the card gets reviewed that I'll be interested.

Oh and welcome to the forums the54thvoid, don't forget to fill in your System Specification.
Posted on Reply
#9
Animalpak
It seems to me a joke the single GTX 380 is faster than the long and weighing 5970? If it is true I have to laugh ahaha.. Owned
Posted on Reply
#10
Lionheart
ZOMG, I just realized the Z in ZOMG doesnt mean anything lol:D:D:D, oh yeah about these benchmarks reviews, complete BS:p
Posted on Reply
#11
z1tu
:roll: what a pathetic attempt, they mean to tell me that the 5970 is just barely better than the 360 will be and in some cases even worse? :roll::roll::roll::roll: Who the hell falls for crap like this?:shadedshu
Posted on Reply
#12
crow1001
I like to see sites like tech get a fair share of traffic but this is not the way to do it, guru know they are fake but they know it will attract loads of hits, tech does not have to go down this root when its so obvious they are fake.
Posted on Reply
#13
HammerON
The Watchful Moderator
Nay sayers ~ let us wait and see:)
Posted on Reply
#14
blibba
225W TDP... gief GTX395 nao.
Posted on Reply
#15
qubit
Overclocked quantum bit
crow1001 said:
FAKE!!

Take em down, you all look like suckers.:slap:
They might be, but I wouldn't be surprised if the real figures are in this ballpark.

A GTX 300 seems to have about double the performance of a GTX 285, which is not unreasonable.
Posted on Reply
#16
crow1001
WTF you talking about " seems " keep guessing away, most NV fanboys do anyway. I'll wait for legit reviews.
Posted on Reply
#17
gigabit942007
from what i see this are only DX10 tests no real DX11 tests with tesselation and other dx11 stuff so from what i see they might be better in DX10 but they might suck in DX11 and these tests are not real !
Posted on Reply
#18
t77snapshot
crow1001 said:
FAKE!!

Take em down, you all look like suckers.:slap:
hey show some respect to our news Editors ok.:shadedshu

gigabit942007 said:
from what i see this are only DX10 tests no real DX11 tests with tesselation and other dx11 stuff so from what i see they might be better in DX10 but they might suck in DX11 and these tests are not real !
Well there are very little games right now that even utilize DX11.
Posted on Reply
#19
qubit
Overclocked quantum bit
crow1001 said:
WTF you talking about " seems " keep guessing away, most NV fanboys do anyway. I'll wait for legit reviews.
Just because I have an nvidia card does not make me a "fanboy". Thanks. The tests may well not be real and I said that, too.

And you gotta admit my point is reasonable. After all, the 5870 roughly doubled the performance of the 4890, didn't it?
Posted on Reply
#20
Flyordie
You guys do realize ATI isn't just gonna stand around and watch this unfold...

There will be an HD5980 or equivalent. ;-p

Probably a 512bit GDDR5 HD5890... As we all know the HD58xx series is very bandwidth starved... giving it that extra bandwidth should pull it ahead of almost everything Nvidia can throw out...
Posted on Reply
#21
Amok
Fakes.
Look at the closeup, someone pasted in the numbers in photoshop, it's really obvious.

You can notice the heavy pixelation around the numbers. Cheap stuff



Let me show you what i crafted in 5 minutes... Jen Hsuang gave me this bench himself. Do you believe me?



And i can make as many versions, with whatever scores you want.
Posted on Reply
#22
stasdm
Whichever the numbers are, they show that GTX3XX is a bad scaler.
380 looses more than 50% cores number gain when compared to 360.

That means there would not be any sence in dual GTX3XX configuration (the productivity gain will hardly be about 5%). So, in high-end gaming it's a looser.
Posted on Reply
#23
laszlo
Amok said:
Fakes.
Look at the closeup, someone pasted in the numbers in photoshop, it's really obvious.

You can notice the heavy pixelation around the numbers. Cheap stuff

http://i50.tinypic.com/34nhz6e.jpg

.
i don't see any pixel distortions in the posted ones by bta :confused:
Posted on Reply
#24
Amok
you can't see it because i zoomed in on the picture in photoshop and screenshoted it. At the normal low resolution it's not noticeable, but once you zoom in it's really obvious.

I made another zoom in, even further in...
See what i mean. Do it yourself, you will see it as well. Really shabby work from the guy who faked it.
Posted on Reply
#25
laszlo
Amok said:
you can't see it because i zoomed in on the picture in photoshop and screenshoted it. At the normal low resolution it's not noticeable, but once you zoom in it's really obvious.

I made another zoom in, even further in...
See what i mean. Do it yourself, you will see it as well. Really shabby work from the guy who faked it.
http://i45.tinypic.com/n71o9e.jpg
i zoom it at max ....nothing... take the posted picture not the uploaded:)
Posted on Reply
Add your own comment