NVIDIA GF100 & Friends speculation

For the best case? Why not?

 

If the 470 was faster than the 5870 (or even equal, for that matter) and the 480 significantly faster, Nvidia would have designed that slide accordingly for sure. It's a PR/marketing slide afterall, it's supposed to say "look how good our products are and how bad theirs are!". You don't place the 470 so far below the 5870 in such a slide unless you absolutely have to because everything else would be revealed as a flat out lie once the NDA lifts.

 

How old, or new is that slide?

 

Depends on what they are talking about is "$"

 

It sort of puts a limit to 470's performance - Do you REALLY think PR would put it next to 5850 (slightly above, but below a bar that wouldn't have to be there), if it was actually more or less on par with the 5870?

Interesting numbers from the nv ftp btw (time to look for reliable references regarding the first 4 colums) - but I would recommend nvidia to 'mark' the ftp accounts by moving the watermark around a bit, instead off adding removable numbers

 

Good one - but an X-Fi can do some more things, so it'd add value on its own yet again.

BTT: Most people i know are still using regularly connected 2.1 - 5.1 speakers, not routed with several HDMI-extenders through their receivers in their living rooms and back to the room where the PC sits.

It's a great-to-have feature though for living room/HT PCs, but in my personal opinion it is not very useful for high-end gaming cards. It is no more decisive for my buying decision as for example the video decoders capabilities (again: for high-end gaming cards that sit in high-end gaming rigs).

As I said: We're considering it right now. But a, no two - we added 4870 X2 too - bad decisions from half a year back don't necessarily mean, we have to keep with that until all time ends. After all, we've quit benchmarking Quake 3 also without being called names.

 

x86 IS LOCKED IN! All the libraries, the tools, the drivers, flows, etc. Are basically x86 based now. This doesn't even get into the quality of things like the compilers vs other architectures.

yes, and your point is?

And here I was thinking free and supported with drivers.

 

they insulted you for that? that sucks >.<

 

What I mean is that there is vastly less "lock-in" of x86 processors. The main reason that they're so popular is because they're often genuinely better for the cost, a cost that is subsidized by the desktop/laptop markets for AMD and Intel, a cost reduction not available to other companies. However, if there is a real, cost-effective alternative to x86 processors, I don't think it will take long for HPC's to take advantage of it. If, for instance, nVidia could work with some other CPU company to bundle a CPU with their GPU, leveraging the GPU R&D from the consumer space, it is conceivable they could put out a cost-competitive, high-performance integrated design that could well be quite useful there. It would be a bit of an uphill battle, but still within reason.

But for desktop systems, there just isn't any reasonable way for any significant move away from x86, because Windows is so pervasive.

As for drivers, the majority of Linux drivers these days are built into the kernel. I'd be rather surprised if those aren't cross-platform by large.

 

So wait, if I understand this correctly - you will only ever benchmark dual-gpu cards against other dual-gpu cards from now on?

I mean if you aren't going to bench them against single gpu cards then surely that only leaves dual gpu cards?

 

You guaiz seen this?

 

So true.

 

The taster article observes:

A more useful comparison for these crossfire-on-a-stick cards is probably to include other Crossfire and SLI solutions.

So Carsten, as we seem to be in request mode in this topic....

 

Bad Company 2 gets tested twice and with diff results at the same resolution. That's impressive.

 

I think you mean same results, different res.

 

Probably just a type since the guy had to copy it by hand. It matches up with the other resolution.

The numbers don't seem unreasonable. 10-50% faster averaging 25% is pretty much where I expected the GTX480 to be.

 

that one smells of nvidia marketing i'm not trusting that since there's no way any reviewer's gotten 2 480s and a 470 by now

 

but i kinda hope its real

 

Sure, they are clearly labeled as nvidia slides, source here: http://futuremark.yougamers.com/forum/showthread.php?t=125009 So most probably legit, but of course with all the usual bias regarding games, settings and drivers we can expect from marketing slides (from either company). Not something I would draw an average from.

 

Me too, if it is ATi will have to respond with something better than OC versions. I want a 5890 for less than £300 dammit!!