GPU model | GF104 | GP100 |
---|---|---|
GPU variant | GF104-300-KB-A1 | — |
Architecture | Fermi | Pascal |
Foundry | TSMC | TSMC |
Manufacturing process | 40 nm | 16 nm |
Number of transistors | 1,950 million | 15,300 million |
Die size | 332 mm² | 610 mm² |
Release date | Jul 12th, 2010 | Jun 20th, 2016 |
---|---|---|
Generation | GeForce 400 | Tesla |
Production | End-of-life | End-of-life |
Launch price | 199 USD | 4,599 USD |
Interface | PCIe 2.0 x16 | PCIe 3.0 x16 |
Reviews | 154 in our database | — |
Predecessor | GeForce 200 | — |
Successor | GeForce 500 | — |
GPU clock | 675 MHz | — |
---|---|---|
Shader clock | 1350 MHz | — |
Memory clock | 900 MHz 3.6 Gbps effective | 715 MHz 1430 Mbps effective |
Base clock | — | 1190 MHz |
Boost clock | — | 1329 MHz |
Memory size | 768 MB | 12 GB |
---|---|---|
Memory type | GDDR5 | HBM2 |
Memory bus width | 192 bit | 3072 bit |
Bandwidth | 86.40 GB/s | 549.1 GB/s |
Shading units | 336 | 3584 |
---|---|---|
TMUs | 56 | 224 |
ROPs | 24 | 96 |
SM count | 7 | 56 |
Cache L1 | 64 KB (per SM) | 24 KB (per SM) |
Cache L2 | 384 KB | 3 MB |
Pixel rate | 9.450 GPixel/s | 127.6 GPixel/s |
---|---|---|
Texture rate | 37.80 GTexel/s | 297.7 GTexel/s |
FP32 (float) performance | 907.2 GFLOPS | 9.526 TFLOPS |
FP64 (double) performance | 75.60 GFLOPS (1:12) | 4.763 TFLOPS (1:2) |
FP16 (half) performance | — | 19.05 TFLOPS (2:1) |
Slot width | Dual-slot | Dual-slot |
---|---|---|
Length | 210 mm 8.3 inches | 267 mm 10.5 inches |
TDP | 160 W | 250 W |
Suggested PSU | 450 W | 600 W |
Outputs | 2x DVI1x mini-HDMI | No outputs |
Power connectors | — | 1x 8-pin |
DirectX | 12 (11_0) | 12 (12_1) |
---|---|---|
OpenGL | 4.6 | 4.6 |
OpenCL | 1.1 | 3.0 |
Vulkan | — | 1.2 |
CUDA | 2.1 | 6.0 |
Shader model | 5.1 | 6.4 |