GPU model | GP104 | GP102 |
---|---|---|
GPU variant | GP104-140-KA-A1 | — |
Architecture | Pascal | Pascal |
Foundry | TSMC | TSMC |
Manufacturing process | 16 nm | 16 nm |
Number of transistors | 7,200 million | 11,800 million |
Die size | 314 mm² | 471 mm² |
Release date | Dec 25th, 2016 | Sep 13th, 2016 |
---|---|---|
Generation | GeForce 10 | Tesla |
Predecessor | GeForce 900 | — |
Successor | GeForce 20 | — |
Production | End-of-life | End-of-life |
Launch price | 199 USD | 5,699 USD |
Interface | PCIe 3.0 x16 | PCIe 3.0 x16 |
Reviews | 117 in our database | — |
Base clock | 1506 MHz | 1303 MHz |
---|---|---|
Boost clock | 1708 MHz | 1531 MHz |
Memory clock | 2002 MHz 8 Gbps effective | 1808 MHz 14.5 Gbps effective |
Memory size | 3 GB | 24 GB |
---|---|---|
Memory type | GDDR5 | GDDR5X |
Memory bus width | 192 bit | 384 bit |
Bandwidth | 192.2 GB/s | 694.3 GB/s |
Shading units | 1152 | 3840 |
---|---|---|
TMUs | 72 | 240 |
ROPs | 48 | 96 |
SM count | 9 | 30 |
Cache L1 | 48 KB (per SM) | 48 KB (per SM) |
Cache L2 | 1536 KB | 3 MB |
Pixel rate | 81.98 GPixel/s | 147.0 GPixel/s |
---|---|---|
Texture rate | 123.0 GTexel/s | 367.4 GTexel/s |
FP16 (half) performance | 61.49 GFLOPS (1:64) | 183.7 GFLOPS (1:64) |
FP32 (float) performance | 3.935 TFLOPS | 11.76 TFLOPS |
FP64 (double) performance | 123.0 GFLOPS (1:32) | 367.4 GFLOPS (1:32) |
Slot width | Dual-slot | Dual-slot |
---|---|---|
Length | 250 mm 9.8 inches | 267 mm 10.5 inches |
TDP | 120 W | 250 W |
Suggested PSU | 300 W | 600 W |
Outputs | 1x DVI1x HDMI3x DisplayPort | No outputs |
Power connectors | 1x 6-pin | 1x 6-pin + 1x 8-pin |
Width | — | 111 mm 4.4 inches |
Board number | — | PG610 SKU 200 |
DirectX | 12 (12_1) | 12 (12_1) |
---|---|---|
OpenGL | 4.6 | 4.6 |
OpenCL | 3.0 | 3.0 |
Vulkan | 1.2 | 1.2 |
CUDA | 6.1 | 6.1 |
Shader model | 6.4 | 6.4 |