GPU model | G80 | GF108 |
---|---|---|
GPU variant | G80-100-K0-A2 | GF108-400-A1 |
Architecture | Tesla | Fermi |
Foundry | TSMC | TSMC |
Manufacturing process | 90 nm | 40 nm |
Number of transistors | 681 million | 585 million |
Die size | 484 mm² | 116 mm² |
Release date | Nov 8th, 2006 | Oct 11th, 2010 |
---|---|---|
Generation | GeForce 8 | GeForce 400 |
Production | End-of-life | End-of-life |
Launch price | 449 USD | — |
Interface | PCIe 1.0 x16 | PCIe 2.0 x16 |
Reviews | 403 in our database | 33 in our database |
Predecessor | GeForce 7 PCIe | GeForce 200 |
Successor | GeForce 9 | GeForce 500 |
GPU clock | 513 MHz | 700 MHz |
---|---|---|
Shader clock | 1188 MHz | 1400 MHz |
Memory clock | 792 MHz 1584 Mbps effective | 800 MHz 1600 Mbps effective |
Memory size | 640 MB | 2 GB |
---|---|---|
Memory type | GDDR3 | DDR3 |
Memory bus width | 320 bit | 128 bit |
Bandwidth | 63.36 GB/s | 25.60 GB/s |
Shading units | 96 | 96 |
---|---|---|
TMUs | 24 | 16 |
ROPs | 20 | 4 |
SM count | 12 | 2 |
Cache L2 | 80 KB | 256 KB |
Cache L1 | — | 64 KB (per SM) |
Pixel rate | 10.26 GPixel/s | 2.800 GPixel/s |
---|---|---|
Texture rate | 24.62 GTexel/s | 11.20 GTexel/s |
FP32 (float) performance | 228.1 GFLOPS | 268.8 GFLOPS |
FP64 (double) performance | — | 22.40 GFLOPS (1:12) |
Slot width | Dual-slot | Single-slot |
---|---|---|
Length | 267 mm 10.5 inches | 145 mm 5.7 inches |
TDP | 143 W | 49 W |
Suggested PSU | 300 W | 200 W |
Outputs | 2x DVI1x S-Video | 1x DVI1x HDMI1x VGA |
Power connectors | 1x 6-pin | None |
Board number | P356 | — |
DirectX | 11.1 (10_0) | 12 (11_0) |
---|---|---|
OpenGL | 3.3 | 4.6 |
OpenCL | 1.1 (1.0) | 1.1 |
Vulkan | — | — |
CUDA | 1.0 | 2.1 |
Shader model | 4.0 | 5.1 |