The GT300/Fermi Thread - Part 2!

[mapel110]

[-Sweeper_]

621 mm² die size?

[zerazax]

The images are correct. There's no point in nitpicking over it, they did some analyzing at B3D and there's nothing up with either on the benchmark

[trinibwoy]

Yep, they made a chip that's bigger than what TSMC is physically capable of

[zalbard]

900MHz memory can't be right.

[icecpu]

memory only 900MHz ???,
the screen shot looks legit

[zalbard]

memory only 900MHz ???,
the screen shot looks legit

GPU-Z does not support Fermi yet, it is a fake.

[zerazax]

Well, if the GTX 470 is indeed 800 MHz GDDR5, then 900MHz from the 480 isn't out of the realm of possibility

Strange to see that a 256-bit 5870 might have matching bandwidth capability

[fellix_bg]

it is a fake.

Wrong word, it's real!

[Behemot]

900MHz memory can't be right.

Why not? It's the frequency which HD 4870 used. Remember it's first time NVIDIA has GDDR5. They can be just careful.

[-Sweeper_]

Why not? It's the frequency which HD 4870 used. Remember it's first time NVIDIA has GDDR5. They can be just careful.

nvidia already used 850mhz GDDR5 on GT240.

[ElSel10]

Is this real?

[eric66]

First of all, christ people, Unigine looks pretty much identical on both ATI and NVidia, if it werent, one camp or another would call out CHEATER!
Second of all

We have already got to the conclusion that using 2560x1600 on a 1GB card causes dramatic fps drop due to the framebuffer taking such a large amount of memory.
For the people with monstrous 30" displays there will be 2GB 5870 shortly in few different flavours(including OCed one).
So again these results arent exactly comparable.

yeah that pretty obvious 5870 1 gb s_cks for ultra high res now weird part is company that made this card had eyefinity tech as ad at least nvidia did smart move in tam choice kudos to them

[HelixPC]

The final clocks are not out yet, im sure the memory will be at 1100mhz minimum.

[NaMcO]

Lol @ release date. Wow it even has the DAY of release

[Marios]

That's OK. The BIOS version (gigabyte) and a lot more are not OK. Fake.

[Boissez]

yeah that pretty obvious 5870 1 gb s_cks for ultra high res now weird part is company that made this card had eyefinity tech as ad at least nvidia did smart move in tam choice kudos to them

Actually it's the other way around. ATI currently caters to 99% of the market while the remaining percent will have 2 GB in a short while. The 4XX series are needlessly expensive to make.

900MHz memory can't be right.

Makes sense - it's most likely low voltage GDDR (ie. like the one used in HD5970) to keep down TDP

[Teemax]

If what you said is true, then nVIDIA lied in a public technical specifications pdf.
And you're wrong about load balancing via the driver.
That would be the silliest move ever.
"CUDA cores" can be dynamically assigned to various tasks and load balancing should and is done by hardware not software.

Not an elegant thinking I'd say

Which part of the "specifications pdf" would be wrong if my claims are true?

NVIDIA claimed that GF100 has stellar tessellation performance, which might well be true. But they never claimed that tessallation wouldn't affect shader performance.

If you have info that disproves what I said, by all means do share it.

[Pontos]

Not sure if posted already, but anyway.

http://hothardware.com/News/NVIDIA-L...d-Vision-Demo/

So it beats the HD5870 on heavy tessellation, but it's a tie on "normal" usage.
Would that mean that the real advantage of Nvidia will be on future DX11 games, and not on current DX9/10 games?

[orangekiwii]

If the advantage is for future dx11 games, then thats not really an advantage for consumer usage. By the time there is any spectacular or even semi decent game (AvP is bad and runs poorly anyway) there will be MUCH better dx11 hardware out. What matters is current games, if gtx480 is the same as 5870, then whats the point for consumers? I want performance in current games not games in 2 years, if I did I'd buy a card in 2 years. If gtx480 does in fact more or less equal 5870, then theres no reason to buy any card from this generation as simply put its just not a big enough step up in performance on any level.

[Designer]

lol, my memory bandwith is better

[NaMcO]

If gtx480 does in fact more or less equal 5870, then theres no reason to buy any card from this generation as simply put its just not a big enough step up in performance on any level.

This is specially true for me. My 295 is faster than a 5870 at all levels, and drivers apart, (because CCC is a real PITA), if the 480 doesn't beat my 295 by a considerable margin there's no reason for me to change video card to a 480.

I want a bit more of juice than the 295 can deliver and the logical step would be a 5970 in case the 480 fails to impress. Thing is... The software sucks to be real honest about CCC, when will ATI deliver simple stuff like game profiles, i've been asking for this forever

I need AA8x for Crysis, but then i want to force AA16x in older games. Some games i want them with Vsync, some others i have to disable it (video card fails to deliver constant 60+fps) and i want all of this to be automatic when i start the game like nVidia drivers do.

I really like the 5970, don't get me wrong, but i always feel that something's missing when i use my other PC with the 4870x2 in it, this keeps me away from ATI for now

[annihilat0r]

lol@you guys actually thinking gpu-z "measures" things. It has a database for its information. If it says 620mm2 and 512 shaders, this doesn't actually mean it has measured the chip's size or anything. It just displays what its makers put in it as information.

[LesGrossman]

My 295 is faster than a 5870 at all levels

At all levels of Far Cry 2 right? Or do you have a special 295?

[zerazax]

lol@you guys actually thinking gpu-z "measures" things. It has a database for its information. If it says 620mm2 and 512 shaders, this doesn't actually mean it has measured the chip's size or anything. It just displays what its makers put in it as information.

Its fake, although clocks do get measured by GPU-Z (thats about the only thing actually measured though)