Nvidia’s card is fast, but ATI’s is still faster.

The fastest card we’ve seen, but the performance boost doesn’t justify the price.

The fastest single-GPU card we've seen, but it can't shake ATI in the price war.

It’s fast, but not quite fast enough for the huge price.

A step behind ATI’s HD 4870 on price and performance.

A minor performance boost isn’t enough to fend off its rivals.

Hard to find, and not worth the typical price at all.

Still holding its own at this price, but note the HD 4830.

Not the perfect value card it once was, but still worth a look.

On its last legs, it costs as much as faster cards.

Decent power, but bigger, dearer and noisier than the HD 4850.

Can't handle gaming and beaten by cheaper media cards too.

Nvidia’s latest graphics cards and chips throw down the gauntlet to Intel and AMD. Go behind the scenes with this guide to the big technologies including PhysX, CUDA, and games like Far Cry 2 and S.T.A.L.K.E.R Clear Sky.

nVidia GeForce 256

Intel's Atom platform may provide the backbone of today's netbooks, but with rivals emerging, Intel could find itself playing a smaller part in the future, as players such as Nvidia's Tegra take hold.

The graphics market is a cutthroat game, and we have seen many casualties recently. Not all of them have been fatal, like Matrox, which has recently backed out of the consumer race, or Rendition, whic

This card is the same card as the Bravura Evil Master II, but without the retail extras thrown in. The card itself is a fairly plain-looking affair - a simple green PCB (printed circuit board) with a small cooling fan and heatsink over the GPU, and an array of RAM modules absent of the large RAM heatsinks now commonplace on the NVIDIA models.

AMD/ATI and its partners made quite a splash with the recent release of the 5 series GPUs, and is creating a lot of excitement about what future iterations of the cards will be able to do

There are a lot of GeForce3 cards on the market at the moment, with not an incredible amount of difference between them. This is indicative of the somewhat heavy-handed tactics of NVIDIA when setting down reference designs.