The GT300/Fermi Thread

**randomizer** · 01-16-2010, 10:10 PM

Originally Posted by -=DVS=-

If im not mistake NDA expires 9pm on Sunday , right now is only Saturday

Quite so. It's only 5pm on Sunday here in NSW, Aus, and we're on DST as well, so really 4pm in this time zone.

Mk · 01-16-2010, 11:30 PM

Facebook NVIDIA's Photos - PDX LAN 2010

**iTravis** · 01-16-2010, 11:51 PM

You miss it on purpose?

The sign is hilarious: ATI = Always Trouble Inside?

**eric66** · 01-16-2010, 11:54 PM

never seen so much nerds together :P

**WeeMaan** · 01-17-2010, 12:13 AM

People are saying it's 9pm pacific time. :\ That would suck.

**zerazax** · 01-17-2010, 12:17 AM

It's the typical time they give... Nvidia is on PST after all

**zerazax** · 01-17-2010, 12:45 AM

Originally Posted by GoldenTiger

In random-scaling microstutter land, just like a GTX 295 does.

You should read the other micro-stuttering thread.... microstutter = overhyped/talked about at best, downright false at worse..

Originally Posted by annihilat0r

there was a Rahja guy posting similar info, I have seen that

Yep that was Rahja, and so much for his NDA expiring on the 12th

**Tim** · 01-17-2010, 12:59 AM

Originally Posted by Mk

Facebook NVIDIA's Photos - PDX LAN 2010

What....the....

And I thought I was a nerd!!!

**Andrew LB** · 01-17-2010, 01:17 AM

Originally Posted by Tim

What....the....

And I thought I was a nerd!!!

Yea. Same here. After taking one look at that bunch, I've come to the realization that I have absolutely nothing to worry about.

The guy with both fists in the air reminds me of when I first got laid. But in his case.... it's over a video card.

**ajaidev** · 01-17-2010, 01:30 AM

Of i was thinking those box's were of GF100 but when i read 250 in the end my heart broke

G92 when will you die...

Oh ya did u see this the Nvidia eyefidelity

so does nvidia have anything against ATI's 6 screen tech?

**neliz** · 01-17-2010, 01:42 AM

Lol@nFinity

restricted to 1920x1080 displays.

**Macadamia** · 01-17-2010, 01:48 AM

Single link DVI.

rofl patchjob

**Jamesrt2004** · 01-17-2010, 01:57 AM

Originally Posted by jazkat

some info i found on another forum

AvP3 and Crysis2 can be taken with a grain of salt, since they are not finalized and drivers aren't finalized yet. DiRT2 may improve or decline slightly, depends on final drivers.

Those tests were run using a Corei7 920, 6GBs of DDR3 1333, and an Intel 64GB SSD paired with a single GF100 card. The tests were run at 1920x1200 with 4x SSAA and 16xAF. Happy? If you want more details, wait for a benchmarking site to run benchmarks like everyone else. I am spoiling you as it is. Don't bite the hand that feeds.

The GF100 card length is 10.61"
The GF104 is going to be larger, around 5870 length
The GF104 is the faster of the two cards.

At midnight (EST, -5 GMT), I will fill in the blanks.... MUHAHAHAA.

GF100 outperforms ATi's 5870 by 46% on average
GF100 outperforms ATi's 5970 by 8% on average

The GF100 gets 148 fps in DiRT2
The GF100 gets 73 fps in Crysis2
The GF100 gets 82 fps in AvP3

maybe you guys have already seen this?

it's wrong if it was 46% faster.. nvidia would be ALL over it not just saying 20%...

**Teemax** · 01-17-2010, 02:04 AM

Originally Posted by Macadamia

Single link DVI.

rofl patchjob

I don't think so. NVIDIA 3D requires 120hz refresh rate which doubles the bandwidth from the standard 60hz. So dual link DVI wouldn't have enough bandwidth to drive 2560x1600 resolution in 3D.

I wouldn't trade higher resolution for 3D though. Given that 3D glasses also hinders the performance monitors; I don't think 3D is anywhere ready for prime-time.

**Jamesrt2004** · 01-17-2010, 02:11 AM

Originally Posted by Teemax

I don't think so. NVIDIA 3D requires 120hz refresh rate which doubles the bandwidth from the standard 60hz. So dual link DVI wouldn't have enough bandwidth to drive 2560x1600 resolution in 3D.

I wouldn't trade higher resolution for 3D though. Given that 3D glasses also hinders the performance monitors; I don't think 3D is anywhere ready for prime-time.

well nvidia's implimentation is a bad way, there is that cimpany at CES which has 3d without the need for glasses and just a 120hz res which is WAYYYY better and worked just as good... thats the way they should do it...

**neliz** · 01-17-2010, 02:12 AM

Originally Posted by Jamesrt2004

well nvidia's implimentation is a bad way, there is that cimpany at CES which has 3d without the need for glasses and just a 120hz res which is WAYYYY better and worked just as good... thats the way they should do it...

I had a 3D monitor with nvidia support back in 2005 :/

**Jamesrt2004** · 01-17-2010, 02:17 AM

Originally Posted by neliz

I had a 3D monitor with nvidia support back in 2005 :/

? still need to use those god awful glasses though, im saying they should change the way they implement it to the same as the company at CES which don't need the shutter glasses or anything.. it just works

**saaya** · 01-17-2010, 02:35 AM

Originally Posted by neliz

one is 725, one is 750 (i hope I'm not reading fake fact sheets now btw)

too high clocks i think...

Originally Posted by neliz

It says that GF100 scores P22000 in Vantage.. which.. you already knew for a couple of hours...

5k above a 5870, which is around 25% faster...

as fast as 285sli and a 5970... makes sense... i guess...
but isnt a P score kinda pointless for such a highend card?
how about X?

Originally Posted by Sam_oslo

22k is behind the GTX 295, but can be a good start for a single card with early (and totally new) drivers..

295 scores more than 22K? where?

Originally Posted by jazkat

The GF100 card length is 10.61"
The GF104 is going to be larger, around 5870 length
The GF104 is the faster of the two cards.

At midnight (EST, -5 GMT), I will fill in the blanks.... MUHAHAHAA.

GF100 outperforms ATi's 5870 by 46% on average
GF100 outperforms ATi's 5970 by 8% on average

The GF100 gets 148 fps in DiRT2
The GF100 gets 73 fps in Crysis2
The GF100 gets 82 fps in AvP3

cheers

so 100=360 and 104=380?
with this kind of performance... why would anybody want sli or a dual card unless its for benching?

but those are nvidia pr numbers...
the real numbers are probably notably lower... theres a reason so many people said ~25% faster than 5870, and i dont think it can beat the 5970 consistently... if it would be 48% faster than a 5870 it would have to score over 25K in vantage, and a card performing better compared to the competition in games than in 3dmark? thats not going to happen... 3dmark has turned into a 3D sisoft sandra benchmark, showing the maximum theoretical performance which is always higher than realworld performance, and the gap between theoretical max and actual max grows with every new card. so a new card that is 30% faster in 3dmark compared to another card, is most likely only 25-29% faster in games.

Originally Posted by neliz

Lol@nFinity

restricted to 1920x1080 displays.

amazing visual realism and advanced dynamic realism... terrible marketing speak...

Originally Posted by Jamesrt2004

well nvidia's implimentation is a bad way, there is that cimpany at CES which has 3d without the need for glasses and just a 120hz res which is WAYYYY better and worked just as good... thats the way they should do it...

what tech is that? and have you seen it with your own eyes? i saw all the 3d stuff at the display tradeshow here in taipei over half a year ago, CMO CTM and AUO showed their panel tech back then, which samsung sony lg etc used to build displays they now show at CES... and the 3d tech that didnt require glasses was terrible... it was distorted and flickered and low res, and when you moved your head or even just your eyes slightly, the image changes...

**neliz** · 01-17-2010, 02:41 AM

100 = 360/380 104 = 350
X score was 15K

but it seems bogus anyway

**Nightcover** · 01-17-2010, 02:41 AM

How many hours back?

**Mechanical Man** · 01-17-2010, 03:14 AM

~7,5h

**saaya** · 01-17-2010, 03:22 AM

Originally Posted by neliz

100 = 360/380 104 = 350
X score was 15K

but it seems bogus anyway

yeah that makes more sense...
15K X is very good... that is indeed 48% faster than a 5870 then...

btw, remember this?
everybody said it was fake, but the deep dive numbers seem to be about the same?

http://www.youtube.com/watch?v=kKbq6...eature=related