The GT300/Fermi Thread

[kemo]

Cant find it on their webpage

http://www.techpowerup.com/reviews/N..._Architecture/
Wizard posted it here earlier

[saaya]

Cant find it on their webpage

here ya go
http://www.techpowerup.com/reviews/N..._Architecture/

Here i put the Far Cry2 numbers in an xbitlabs graph, xbit tests with a 975 though, while nvidia supposedly tested with a 960. no big difference though, 3.33 vs 3.2ghz and we are talking 1920x1200 and 2560x1600 here...

So all signs so far point to being faster than a 5870 by 20-30% and tad slower than a 5970. Looks like won't be that big of a price war for us consumers then.

weeellll if those numbers ARE for a 360, then this is good news, if they are for a 380, then they are not that good news... i think it was a 380...

so basically fermi is as fast as a 295, maybe a tad faster... makes sense...
but considering they more than doubled the logic, i actually expected more...

[JohnZS]

here ya go
http://www.techpowerup.com/reviews/N..._Architecture/

Here i put the Far Cry2 numbers in an xbitlabs graph, xbit tests with a 975 though, while nvidia supposedly tested with a 960. no big difference though, 3.33 vs 3.2ghz and we are talking 1920x1200 and 2560x1600 here...


weeellll if those numbers ARE for a 360, then this is good news, if they are for a 380, then they are not that good news... i think it was a 380...

Holy smokes batman, look at the MINIMUM framerates. Fermi rocks the socks off all Ati cards in that test including the 5970. Hats off to nVidia

I only hope the Fermi's are more reliable than the GT200 series cards.
John

[Solus Corvus]

Holy smokes batman, look at the MINIMUM framerates. Fermi rocks the socks off all Ati cards in that test including the 5970. Hats off to nVidia

I only hope the Fermi's are more reliable than the GT200 series cards.
John

CPU has an effect on minimum framerates.

[saaya]

Holy smokes batman, look at the MINIMUM framerates. Fermi rocks the socks off all Ati cards in that test including the 5970. Hats off to nVidia

I only hope the Fermi's are more reliable than the GT200 series cards.
John

5970 =42, fermi =45... hows that rocking off the socks? :P

[JohnZS]

Yea Far Cry 2 doesn't seem to give Core i7 much love, or it gives extra love to Phenom II.

Only if you are using an ASUS based intel board....

[kgk]

Me.

Heh. But to be fair, if you want the details cranked up on 5xxx by 3xxx resolution for new games, you're going to need more than one 5800 or GF100 anyway.....

[kgk]

5970 =42, fermi =45... hows that rocking off the socks? :P

Not an NVidia fanboy, but if those numbers are true they are doing it with one GPU compared to ATI's 2.....

[ajaidev]

Prices of most 5850 and 5870 have decreased to $289 and $399 in newegg, so u can get 2*5850 for $578 not bad at all

Not an NVidia fanboy, but if those numbers are true they are doing it with one GPU compared to ATI's 2.....

lol the old 2 gpu vs 1 gpu, we should really compare it segment-wise 5970 is suppose to fight GF100 from the start. 4870 X2 was suppose to fight GTX280/285.

In principle ATi has no plans at all to fight with either GTX295 or GTX395

[Solus Corvus]

Not an NVidia fanboy, but if those numbers are true they are doing it with one GPU compared to ATI's 2.....

Again, CPU has an effect on minimum framerates.

[saaya]

Not an NVidia fanboy, but if those numbers are true they are doing it with one GPU compared to ATI's 2.....

good point, yeah... i hope they can fix the microstuttering problems... they managed to do it on 4870x2, didnt they?

lol the old 2 gpu vs 1 gpu, we should really compare it segment-wise 5970 is suppose to fight GF100 from the start. 4870 X2 was suppose to fight GTX280/285.

thats what ati and nvidia want, but it doesnt work like that... :P
single gpu cards still perform better than dual gpu cards in a way that fps alone cant meassure...

[ajaidev]

Heh. But to be fair, if you want the details cranked up on 5xxx by 3xxx resolution for new games, you're going to need more than one 5800 or GF100 anyway.....

http://www.hardocp.com/article/2010/...mance_review/1

resolution 5760*1200 the 5970 "dual gpu" and 5870 "single gpu" perform quite well in most modern games.

good point, yeah... i hope they can fix the microstuttering problems... they managed to do it on 4870x2, didnt they?

thats what ati and nvidia want, but it doesnt work like that... :P
single gpu cards still perform better than dual gpu cards in a way that fps alone cant meassure...

Yep with new driver releases micro shutter should get much less ass was the case with 4870 x2.

yes single card solution does work better but ATi has no single card solution in the same segment as the GF100 it is really odd to compare a Chevrolet Cruze to a Porsche 911?

[firas]

While disappointing that here are no easy to read benchmarks, it's possible to discern general performance from a few key points: The HAWX benchmark in the deep dive slides and the FARCRY2 GF100 vs. GTX285 video. (Check out guru3d's video of the Farcry2 benchmark for extra info). GF100 is ~ 68% faster in FC2, and 61% faster in HAWX, both using 4xAA, compared to a GTX285.

HD5870 is ~35% faster than a GTX285 1920 4xAA, according to Computerbase.de. Also, an HD5970 is ~90% faster than a GTX285. The tested GF100 part sits between a HD5870 & HD5970, closer towards the speed of the 5970. And it appears GF100 handles 8xAA better than HD5000. I rank them as so:

GTX285: 100%
HD5870: 135%
GTX3xx: 168%
HD5970: 190%

and for the price...

Nvidia scenario
GTX285: 100% - 400
HD5870: 135% - 400
GTX3xx: 168% - 672
HD5970: 190% - 700

ATI scenario
GTX285: 100% - 400
HD5870: 135% - 400
GTX3xx: 168% - 600
HD5970: 190% - 700

my scenario
GTX285: 100% - 400
HD5870: 135% - 400
GTX3xx: 168% - 650
HD5970: 190% - 700

edit: based on the Dark Void benchmark it's 180% faster than the 285 not 168%?

[saaya]

ROFL, whats up with H? half the page are nvidia ads

http://www.hardocp.com/article/2010/...mance_review/1

resolution 5760*1200 the 5970 "dual gpu" and 5870 "single gpu" perform quite well in most modern games.

ahum, they say crysis is playable at 5760x1200 with a 5970... and a min fps of 10... :P

[ajaidev]

Thats why i said most games :P

A 5870 does quite well in most games also but 5970 does excels in some games like GTA4,etc...

[saaya]

best quote about fermi from H:

I think GF100 is going to be faster than 5870

i like that H actually posted the entire presentation nvidia gave them, thats cool
most is useless, but i still prefer looking at it and making my own conclusion about them instead of somebody telling me this and that info is useless and not showing it...

[jfromeo]

2560x1600 users, rejoice

Definitely grabbing two 380GTX for 32xSSAA 16XAF 1600p awesomeness. I really hope they do scale well.

[annihilat0r]

Yeah, that'd definitely do better than fx5200 :P

[saaya]

2560x1600 users, rejoice

Definitely grabbing two 380GTX for 32xSSAA 16XAF 1600p awesomeness. I really hope they do scale well.

price will probably be 700$, since supply will be VERY low, the price might actually climb above that... so prepare to pay almost 2K us$ for 2 cards as you might have to import them

i recommend you to pre-order the cards, iirc the 280s were really in very short supply for the first weeks after the launch... either you buy it as soon as its out, or youll have to wait for a while until you can get them i think...

[annihilat0r]

price will probably be 700$, since supply will be VERY low, the price might actually climb above that... so prepare to pay almost 2K us$ for 2 cards as you might have to import them

i recommend you to pre-order the cards, iirc the 280s were really in very short supply for the first weeks after the launch... either you buy it as soon as its out, or youll have to wait for a while until you can get them i think...

of couse, all of this will be true IF the 512sp variant makes it to launch... and I am not even talking about a hard launch

[jaredpace]

based on the Dark Void benchmark it's 180% faster than the 285 not 168%?

Perhaps because GTX285 has 1/4th the gpgpu compute power of fermi and this benchmark is using compute + rendering.

For the results you see below, NVIDIA made use of a built-in benchmark that make heavy use of in-game physics and the engine’s long draw distances. This is one test where the GPU compute roots of the GF100 can really come into play by efficiently processing PhysX and rendering the scene at the same time. Performance is once again far beyond anything the GTX 285 can accomplish which makes should make things interesting come release.

[Xoulz]

Including architecture scaling, GAMING PERFORMANCE and a whole wack of other stuff.....


RIGHT HERE!!!!!

Thanks!

[jfromeo]

Well, I'm planning to update in early June, just after I'm done with the exams. I hope retailers have some stock by then.

Thanks for the tip though, saaya

[Nedjo]

Review up @ AT too.
http://www.anandtech.com/video/showdoc.aspx?i=3721

In short, here’s what we still don’t know and will not be able to cover today:

1. Die size
2. What cards will be made from the GF100
3. Clock speeds
4. Power usage (we only know that it’s more than GT200)
5. Pricing
6. Performance

At this point the final products and pricing are going to heavily depend on what the final GF100 chips are like.

However, right now there are unknown variables about the GF100. While we have a good idea about the inner workings of the GPU, there are other factors we simply do not know yet. Not much is really known about pixel shader performance. Without knowing how the clock domains are setup, and what the actual clock speeds are, we can’t even guess at what pixel shading performance will actually be like.

Then there are the variables that will make or break this GPU. We don’t know what the cost of video cards based on this GPU will be. We don’t know how much power this GPU is going to demand and what kind of power supply will be required to operate it. We don’t know the kind of heat output it will generate and what kind of cooling solutions will be required. We don’t know NVIDIA’s production and yields of this GPU. Lastly, we don’t know what availability is going to be like.

When all is said and done, the GF100 sounds impressive in terms of feature set and performance for gaming, but we are far away from being able to tell you what kind of gaming experience it is going to bring to the table. We are also far away from being able to tell you if it is a good value or not based on other physical factors. We’ve been burned before.

This has gotta be the most desperate attempt in the history of the GFX market to stop sales of competing product (and stalling the sales of your own lineup) by upping the dust around the phantom product that will not be out in the next two or three months! Bravo nVidia

[Jamesrt2004]

Not an NVidia fanboy, but if those numbers are true they are doing it with one GPU compared to ATI's 2.....

yet they will be same price.