The Fermi Thread - Part 3

[onethreehill]

/YAWN

Since when those are THE final specs? Where are the clocks? Whats new there really? Just the 512 or 480 debate. The rest was pretty much known. So much fanfarre for nothing, IMO.

Do you have a source/link to confirm/reveal SP, TDP and price?

[Picao84]

Do you have a source/link to confirm/reveal SP, TDP and price?

Price is not a spec of the card is it?
And we know Fermi has 512 cores for how long? The only debate was rumour it would have only 480 enabled. 448 cores for GTX470 was pretty much a given, because Tesla parts have 448. So it would make sence to be next downgrade step.

Really, this is not really new info, except for the price, which is NOT a spec.

What i mean is really that without the clocks you dont know THE final specs. Shader count and TDP are NOT enough to define card specs.

[h0bbes]

I thought the 295W TDP was supposed to be lower, so that is news (to me at least)

[annihilat0r]

If 512sp is correct (which it seems to be right now) then the 700mhz clocks can't be true otherwise the performance difference would be a lot more...

[cegras]

Poop sling fest ahoy.

[Picao84]

I thought the 295W TDP was supposed to be lower, so that is news (to me at least)

Well, some days ago the same site was claiming it to be 298W. Its not that far is it?

[Picao84]

If 512sp is correct (which it seems to be right now) then the 700mhz clocks can't be true otherwise the performance difference would be a lot more...

But power is close to 300W. And performance is still yet to see. Saying its 5-10% higher than HD5870 is vague at best. In what games and settings? Contrary to things like shaders cores, memory ammount, TDP and price, it varies on the situation being tested.

[Sn@ke:~]

I think we'll agree on this finally since we know this happens in reality.

The question is, what happens if you have 4GB VRAM on your graphics? The system won't boot as I assumed?

I'll sure try it out if i can get my hand on 2 x gf100 one day.......! .

I'm not totally sure, but i would tend to believe it keeps the minimum required to boot and the rest for hardware. Therefore all hardware would operate with the remainder and probably would probably crash or something like address error, BSD or what not.

Is there anyone on Extreme that has 2 x 5970 running on XP 32bit ? Just ask them to try it out ! lmao

[Vardant]

So they don't know the clocks, but they know how fast they are. Sure, I'll believe that.

It seems more like NV is keeping everyone in the dark, so they all went the Charlie way, covering every possibility so they can state, they've been right all along.

[SKYMTL]

Wow, the number of false leads and misinformation in this thread and the links / pictures posted is simply stunning. Mind blowing actually.

[jtdigital]

Nvidia always charges more, doubt those prices are accurate

[hennyo]

I think we'll agree on this finally since we know this happens in reality.

The question is, what happens if you have 4GB VRAM on your graphics? The system won't boot as I assumed?

Well, that is really simple, the issue only arises with addressed memory, not available memory. The same way you can stick 8 GB of ram into a computer running 32bit windows, it doesn't have an issue booting or anything like that, but it still can't address more memory than it is capable of.

Edit: I should note I am referring to Windows Xp, and not Server editions of Windows which have no trouble addressing much more than 4GB.

[Funky]

Wow, the number of false leads and misinformation in this thread and the links / pictures posted is simply stunning. Mind blowing actually.

what did you expect?

Not many know the facts, those select few who have genuine info are not allowed to talk. And others are happy pretending they know something.

[GoldenTiger]

Wow, the number of false leads and misinformation in this thread and the links / pictures posted is simply stunning. Mind blowing actually.

Bah, looking forward to your review at HardwareCanucks once the NDA drops !

[chuchnit]

Wow, the number of false leads and misinformation in this thread and the links / pictures posted is simply stunning. Mind blowing actually.

I doubt anybody in this thread truly knows anything considering the cards aren't even in partners hands.

[Marios]

The heatpipes increase the height of the GTX480 out of specs.
Should I expect the dual Fermi to have double height? LOL.

Wow, the number of false leads and misinformation in this thread and the links / pictures posted is simply stunning. Mind blowing actually.

It's not false. It's just links to legitimate sites like VrZone and Semiaccurate or personal hopes.
Blame the linked sites in case of any misinformation posted here.

[Zed_X]

... hhehehe, half of my posts are BS? No, i only play game with you, who dont know anything! I know everything, but its funny play with you, nobodies.

That spec are BS, reality is different, only 480s picture on FAKE slides is correct, nothing more.

[Sn0wm@n]

blame nvidia for all that misinfo ... its their fault

[gatecrasherlok]

... hhehehe, half of my posts are BS? No, i only play game with you, who dont know anything! I know everything, but its funny play with you, nobodies.

That spec are BS, reality is different, only 480s picture on FAKE slides is correct, nothing more.

Until the NDA is lifted, everything that is presented to us is BS as far as I am concrned. I think there are many guys and girls like me who will veer away from this type of topic. Endless discussion and debate without source is meaningless atm.

[SKYMTL]

blame nvidia for all that misinfo ... its their fault

So NVIDIA started the rumor-mongering certain sites feel they need to publish for the sole purpose of driving up their traffic numbers? Really? Tell me more...

[nsegative]

http://vr-zone.com/articles/nvidia-g...aled/8635.html

GeForce GTX 480 : 512 SP, 384-bit, 295W TDP, US$499

GeForce GTX 470 : 448 SP, 320-bit, 225W TDP, US$349

Internal benchmarks reveal that GeForce GTX 470 is some 5-10% faster than Radeon HD 5850 and similiar for GeForce GTX 480 over the Radeon HD 5870. Interestingly, the TDP of GeForce GTX 480 is almost similar to Radeon HD 5970 which is a dual GPU card. Interestingly, our sources revealed that there are indeed plans for dual Fermi cards and the TDP of the card is probably gonna be mind blowing.

[annihilat0r]

So NVIDIA started the rumor-mongering certain sites feel they need to publish for the sole purpose of driving up their traffic numbers? Really? Tell me more...

I think that he meant Nvidia's extreme tight lippedness caused all those false info circulating the net so close to the launch.

[MnM]

It seems that there is a lot of "myths" around the 32bit and 4gb+ memory addressing.
The facts are:
32 bit operating systems ARE able to use and address more than 4GB of ram. As mentioned before Enterprise software must be use (and I use the word Enterprise in order to describe Microsoft Windows OS - such as Microsoft Windows 2003 Enterprise Edition or Windows 2008 Enterprise Edition 32 bit which actually supports up to 64GB - yes 64gb! in 32 bit see here http://msdn.microsoft.com/en-us/libr...78(VS.85).aspx). However going over 4GB comes with a performance hit.
Also I can verify the above as I have setup some servers with 32bit Windows 2003 Enterprise Edition with 16GB of Ram and I’ve seen the usage on those gone as high as 12GB of physical memory used (Citrix servers with 150+ users per server).

So it is clear that ANY 32 bit Operating System is capable of supporting and using more than 4gb - including all the XP, Vista, 7 we use at home. Why hasn’t this feature been enabled? Nobody knows but rumours around tend to point to certain licensing costs on Microsoft in order to use this technology - costs which MS will recoup through the more expen$$ive OS-es.

Well, that is really simple, the issue only arises with addressed memory, not available memory. The same way you can stick 8 GB of ram into a computer running 32bit windows, it doesn't have an issue booting or anything like that, but it still can't address more memory than it is capable of.

[Dalten]

I think that he meant Nvidia's extreme tight lippedness caused all those false info circulating the net so close to the launch.

The simplest answer is most people flat out lie about what they know. Not that anyone would do that right? right?!

XS should make a "prophet" title, and anyone who posts info ahead of time with 110% accuracy gets it, that way we'd know ahead of time who's "most probably" got actual info and not just speaking out of their festering asses as is the normal among these kinds of threads..

[tajoh111]

If we don't even have the clocks how can we know the performance. So many rumors out there. Honestly besides the partners, I guess no one knows the specs, not even AMD. I think anything leaked so far are all from engineering samples so anything is possible for retail at the moment.

I don't know if their is hype for this thing as much as their is mystery. I bet not a single reviewer has this card in their hands yet, at least the final one. This NDA/security on this card is tight.