GPU model | GP102 | TU104 |
---|---|---|
Architecture | Pascal | Turing |
Foundry | TSMC | TSMC |
Manufacturing process | 16 nm | 12 nm |
Number of transistors | 11,800 million | 13,600 million |
Die size | 471 mm² | 545 mm² |
GPU variant | — | TU104-895-A1 |
Release date | Never Released | Sep 13th, 2018 |
---|---|---|
Generation | GeForce 10 | Tesla |
Predecessor | GeForce 900 | — |
Successor | GeForce 20 | — |
Production | End-of-life | Active |
Interface | PCIe 3.0 x16 | PCIe 3.0 x16 |
Reviews | 149 in our database | — |
Base clock | 1557 MHz | 585 MHz |
---|---|---|
Boost clock | 1670 MHz | 1590 MHz |
Memory clock | 1376 MHz 11 Gbps effective | 1250 MHz 10 Gbps effective |
Memory size | 10 GB | 16 GB |
---|---|---|
Memory type | GDDR5X | GDDR6 |
Memory bus width | 320 bit | 256 bit |
Bandwidth | 440.3 GB/s | 320.0 GB/s |
Shading units | 3200 | 2560 |
---|---|---|
TMUs | 200 | 160 |
ROPs | 80 | 64 |
SM count | 25 | 40 |
Cache L1 | 48 KB (per SM) | 64 KB (per SM) |
Cache L2 | 2.5 MB | 4 MB |
Tensor cores | — | 320 |
RT cores | — | 40 |
Pixel rate | 133.6 GPixel/s | 101.8 GPixel/s |
---|---|---|
Texture rate | 334.0 GTexel/s | 254.4 GTexel/s |
FP16 (half) performance | 167.0 GFLOPS (1:64) | 65.13 TFLOPS (8:1) |
FP32 (float) performance | 10.69 TFLOPS | 8.141 TFLOPS |
FP64 (double) performance | 334.0 GFLOPS (1:32) | 254.4 GFLOPS (1:32) |
Slot width | Dual-slot | Single-slot |
---|---|---|
Length | 267 mm 10.5 inches | 168 mm 6.6 inches |
Width | 112 mm 4.4 inches | — |
Height | 40 mm 1.6 inches | — |
TDP | 250 W | 70 W |
Suggested PSU | 600 W | 250 W |
Outputs | 1x HDMI3x DisplayPort | No outputs |
Power connectors | 1x 6-pin + 1x 8-pin | None |
Board number | PG611 SKU 10 | PG183 SKU 200 |
DirectX | 12 (12_1) | 12 Ultimate (12_2) |
---|---|---|
OpenGL | 4.6 | 4.6 |
OpenCL | 3.0 | 3.0 |
Vulkan | 1.2 | 1.2 |
CUDA | 6.1 | 7.5 |
Shader model | 6.4 | 6.6 |