GPU model | GM200 | TU104 |
---|---|---|
GPU variant | GM200-895-A1 | TU104-895-A1 |
Architecture | Maxwell 2.0 | Turing |
Foundry | TSMC | TSMC |
Manufacturing process | 28 nm | 12 nm |
Number of transistors | 8,000 million | 13,600 million |
Die size | 601 mm² | 545 mm² |
Release date | Nov 10th, 2015 | Sep 13th, 2018 |
---|---|---|
Generation | Tesla | Tesla |
Production | End-of-life | Active |
Interface | PCIe 3.0 x16 | PCIe 3.0 x16 |
Base clock | 948 MHz | 585 MHz |
---|---|---|
Boost clock | 1112 MHz | 1590 MHz |
Memory clock | 1502 MHz 6 Gbps effective | 1250 MHz 10 Gbps effective |
Memory size | 24 GB | 16 GB |
---|---|---|
Memory type | GDDR5 | GDDR6 |
Memory bus width | 384 bit | 256 bit |
Bandwidth | 288.4 GB/s | 320.0 GB/s |
Shading units | 3072 | 2560 |
---|---|---|
TMUs | 192 | 160 |
ROPs | 96 | 64 |
SMM count | 24 | — |
Cache L1 | 48 KB (per SMM) | 64 KB (per SM) |
Cache L2 | 3 MB | 4 MB |
SM count | — | 40 |
Tensor cores | — | 320 |
RT cores | — | 40 |
Pixel rate | 106.8 GPixel/s | 101.8 GPixel/s |
---|---|---|
Texture rate | 213.5 GTexel/s | 254.4 GTexel/s |
FP32 (float) performance | 6.832 TFLOPS | 8.141 TFLOPS |
FP64 (double) performance | 213.5 GFLOPS (1:32) | 254.4 GFLOPS (1:32) |
FP16 (half) performance | — | 65.13 TFLOPS (8:1) |
Slot width | Dual-slot | Single-slot |
---|---|---|
Length | 267 mm 10.5 inches | 168 mm 6.6 inches |
TDP | 250 W | 70 W |
Suggested PSU | 600 W | 250 W |
Outputs | No outputs | No outputs |
Power connectors | 1x 6-pin + 1x 8-pin | None |
Board number | PG600 SKU 202 | PG183 SKU 200 |
DirectX | 12 (12_1) | 12 Ultimate (12_2) |
---|---|---|
OpenGL | 4.6 | 4.6 |
OpenCL | 3.0 | 3.0 |
Vulkan | 1.1 | 1.2 |
CUDA | 5.2 | 7.5 |
Shader model | 6.4 | 6.6 |