NVIDIA launched the GeForce GTX series 200, this launch includes both the GeForce GTX 260 and 280, now in several flavors.
The new GTX series 200 GPUs are big man, they have roughly 1400 million transistors. It's the biggest chip that NVIDIA has ever built, with 1400M transistors it's a freak on a leash. To compare to, 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 the biggest one they've ever made.
It resulted in 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 do expect NVIDIA to move to a smaller fab process (55nm) later in Q4 2008.
Let's walk through some of the main features (you need to stamp into your head) of this new beast:
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.
216 cores on the GeForce GTX 260 Core 216
192 cores on the GeForce GTX 260
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 frame buffer memory. This is a significant improvement in the architecture. You already spotted it, the new GeForce GTX 260 Core 216 has 216 Shader Processors (SPs).
Let's do some Core 216 math !
Inside that GPU, the shaders cores are clustered in three blocks of eight shader processors.
Now we know that there are ten clusters totaling up towards the 240 shader units for the GeForce GTX 280.
So the GTX 280 has 10x3x8= 240 Shader Processors. Therefore a standard GTX 260 has to have 8x3x8= 192 SPs.
Now with the info above that I just mentioned, you can do the math with me real quick. The regular GeForce GTX 260 has got to have 8 shader clusters and thus 192 shader processors which you can utilize for gaming ... or computing other stuff.
What does this mean, well:
Now I think it's pretty safe to say that the entire GTX 200 series is all one and the same chip. This new Core 216 GPU is in fact a full fetched GTX 280 core with one cluster laser-cut and deactivated. Check out the photo below.
GeForce Series 280 architecture - Look closely and spot the 10 shader clusters. On the GTX 260 Core 216 one has been disabled.
Other than the increased shader count, that should provide a significant boost to the shader compute power, other GPU denominators such as clock speeds remain the same. The GeForce GTX Core 216 SKUs will feature 72 texturing units and 28 ROPs. The updated GPU is technically called G200-103-A2 (the older core was G200-100-A2).The card still has a 448-bit (7x64-bit) wide GDDR3 memory bus with 896 MB of framebuffer/memory. It comes with 1.0 ns memory chips unless you stumble into OC models from difference board partners.
So the extra 24 shader processor cores will give the product a little more bite. Especially in this price-range.
GeForce 8800 Ultra
GeForce 9800 GTX
GeForce GTX 260
GeForce GTX 260 Core 216
GeForce GTX 280
Stream (Shader) Processors
Core Clock (MHz)
Shader Clock (MHz)
Memory Clock (MHz) x2
Two Dual link DVI
The clock frequencies for the GTX 260 and Core 216 version remain the same, 576MHz on the GPU, and 999MHz for the GDDR3 memory. All that on a 448-bit memory interface. The power for the GTX 260 is fed by two six-pin connectors. Power consumption for this 10.5-inch board is 182W.
GeForce GTX 980 2 and 3-way SLI review We review GeForce GTX 980 but this time in a 2-way and 3-way SLI. We'll FLIR them, Overclock them, look at ultra HD performance as well as a micro stuttering analysis with the help of FCAT. Join us i...
GeForce GTX 970 SLI review We review Nvidias money maker, the GeForce GTX 970 but this time in a 2-way SLI. As such we'll be going from fast to faaaaast. In this review we'll run the standard benchmarks, but we will also have...
Palit GeForce GTX 970 Jetstream review Palit is back in the house as we review their GeForce GTX 970 Jetstream edition. The product comes factory overclocked with a boost clock of 1304 MHz. The product has a custom and very small PCB, it'...
ASUS GeForce GTX 970 Strix review We review the new ASUS GeForce GTX 970 Strix edition. A product that has been overhauled in terms of PCB and cooling. ASUS also clocks the product a little faster then reference. The new DirectCU II b...