GeForce GTX 280 review

GeForce GTX series 200

The product we actually review today is the GeForce GTX 280, a raw beast of a processor for sure.

The new GTX series 200 GPUs amounts up-to 1400 million transistors. It's the biggest goblin that NVIDIA has ever built, seriously with 1400m transistors it's a freak on a leash. 1.4 billion transistors.. think about that for a minute guys, and keep in mind that the GeForce 8800 'only' had roughly 700 Million processors. So that's roughly doubling up the previous transistor count. Interestingly enough, that would also double up the die-size of the processor and so you'd expect NVIDIA to move to a smaller fabrication process for this graphics processor. They did not as the new architecture is still based on a 65nm fabrication size. The chip is being made at TSMC and according to them ... the biggest one they've ever made. Yes NV, we believe you alright.

Unconfirmed info states that NVIDIA has come up with a huge die measuring 24 x 24 mm. And not many chips will actually fit on a 300 mm wafer, since , resulting in a die area size of 576 mm2. We'll show you a photo I took of that die area later on, but I measured and it's indeed easily 5.5 cm. I have no clue how a chip this big really yields, but this certainly is one expensive graphics processor to make. We expect NVIDIA to move to a smaller fab process (55nm) pretty soon.

Let's walk through some of the main features (you need to stamp into your head) of this new beast:

• 1.4 billion transistors
• 993 GigaFLOP processing power
• 240 processing (shader) cores (GTX 280)
• 192 processing (shader) cores (GTX 260)
• DirectX 10
• New power management enhancements
• CUDA parallel processing
• GeForce PhysX

So the thing you'll wonder about is, where are all these transistors going? Obviously a big chunk of the transistors are being utilized for the shader cores. And shader cores the product surely has, 240 of them on the GeForce GTX 280. The new shader architecture has some cool new features. Sitting in-between them now is an integration of local cache memory (16k software managed cache). It is sitting in-between a block with 8 shader cores. So simply put, what helps here is that the data / instruction doesn't have to leave the GPU anymore to crunch it's data (normally in the regular framebuffer memory. This is a very significant improvement in the architecture.

Let me show you some building blocks on that information. Light green, the shader cores, yellow the new 16k shared memory cache, brown texture filtering. As you can see the shaders are clustered in three blocks of eight shader processors. Then there are ten clusters totaling up towards the 240 shader units for the GeForce GTX 280. And if you do the math with me real quick then the GeForce GTX 260 has to have 8 shader clusters with a total of 192 shader processors.