Tom's Hardware does Fermi.

http://www.tomshardware.com/reviews/geforce-gtx-480,2585.html

Nvidia's next-gen graphics card finally gets reviewed by Tom's. Will the $500 GTX 480 and the $350 GTX 470 live up to their worth? You decide.

My personal opinion is anything above $200 is a waste and you should just wait for it to drop in value.

I think it very much depends on your income bracket. The last time I bought a card (7 years ago.. no longer gaming seriously) it was just above $300, and that was a HUGE investment for me. But with the salary I expect to get when I enter the industry, I might just buy two of those cards because I can.

The card is kind of disappointing though, I was expecting more from Nvidia with all the extra time they have had to produce their new "flagship"

For the new gen Nvidia cards.

Once it's out, I'll buy it, make it run so fast that it might self-destruct and post a review here.

If it's good, I'll recommend it.

If it's bad, I'll just give it to my cousin.

And have your cousin send me his old card =)

The card itself is pretty good, the GTX 480 is the most powerful single GPU card on the market and the GTX 470 has also a good share of power. However, it has been plagued by high production costs and low yields, as well as a rallying army of ATI users who don't want to see their 5-series purchase going to waste.

The GTX 400 series is known to use a significant amount of power as well as produce a good deal of heat, which seems to be one of the great weaknesses of the cards. Although they have a lot of overclocking headroom because of the high power usage, the cards have to remain cool which means expensive aftermarket cooling is required. Also the cards accel at computations (Folding@home, CUDA, OpenCL) far higher than the ATI cards do, but unfortunately this does not affect graphics performance... yet.

I don't think that Fermi was a failure. This is because the cards themselves are good (and better than the ATI alternatives), but unfortunately they may in the long run, cost more and end up being much like the last GTX 200 series. Was the GTX 200 series a failure? I don't think so either. The problem is price at this point, because the GTX 400 series perform competitively with the ATI 5-series. And unfortunately, ATI probably has the advantage with lower production costs, and the capability to lower their prices. From a performance standpoint, the GTX 400 is fine but from a business/price per performance standpoint, I believe Nvidia lost this one.

Those are some very good points Rothen. My gaming setup has a weak CPU (AMD dual-core 2.2Ghz) but an awesome graphics card (Nvidia GTX 260 216-core) and because Nvidia cards have CUDA, I am able to offload all the processing to the card for HD movies (with the CoreAVC codec) that I would otherwise bog my processor to holy hell. Looking at those benches and how much each of those cards retail for, I think Nvidia is catching up in terms of price/performance which traditionally has always been in ATI's favor. Hopefully later iterations of this GF100-based card will make improvements in performance and they will figure out better ways to cut costs to bring it down to the level of the budget-conscious gamer.

His 2x 8800 GTs?

Anyways, here is my plans for the test:

World in Conflict
Battlefield: Bad Company 2
Crysis
Collin McRae's DiRT 2
Company of Heroes
Half-Life 2 Episode 2
Supreme Commander 2
STALKER: Call of Pripyat

That should give it a run for it's money. :P