GPU Name | GP104 | TU104 |
---|---|---|
GPU Variante | GP104-895-A1 | TU104-895-A1 |
Architektur | Pascal | Turing |
Hersteller | TSMC | TSMC |
Prozessgröße | 16 nm | 12 nm |
Transistoren | 7,200 million | 13,600 million |
Chipgröße | 314 mm² | 545 mm² |
Veröffentlichungsdatum | Sep 13th, 2016 | Sep 13th, 2018 |
---|---|---|
Generation | Tesla | Tesla |
Produktion | End-of-life | Active |
Businterface | PCIe 3.0 x16 | PCIe 3.0 x16 |
Basis-Taktfrequenz | 886 MHz | 585 MHz |
---|---|---|
Boost-Takt | 1114 MHz | 1590 MHz |
Speicher-Taktung | 1502 MHz 6 Gbps effective | 1250 MHz 10 Gbps effective |
Speichergröße | 8 GB | 16 GB |
---|---|---|
Speichertyp | GDDR5 | GDDR6 |
Speicherbus | 256 bit | 256 bit |
Bandbreite | 192.3 GB/s | 320.0 GB/s |
Shading-Einheiten | 2560 | 2560 |
---|---|---|
TMUs | 160 | 160 |
ROPs | 64 | 64 |
Anzahl SM | 20 | 40 |
Cache L1 | 48 KB (per SM) | 64 KB (per SM) |
Cache L2 | 2 MB | 4 MB |
Tensor Cores | — | 320 |
RT-Kerne | — | 40 |
Pixelrate | 71.30 GPixel/s | 101.8 GPixel/s |
---|---|---|
Texturrate | 178.2 GTexel/s | 254.4 GTexel/s |
FP16 (half) Leistung | 89.12 GFLOPS (1:64) | 65.13 TFLOPS (8:1) |
FP32 (float) Leistung | 5.704 TFLOPS | 8.141 TFLOPS |
FP64 (double) Leistung | 178.2 GFLOPS (1:32) | 254.4 GFLOPS (1:32) |
Steckplatzbreite | Single-slot | Single-slot |
---|---|---|
Länge | 168 mm 6.6 inches | 168 mm 6.6 inches |
TDP | 75 W | 70 W |
Vorgeschlagenes Netzteil | 250 W | 250 W |
Ausgabe | No outputs | No outputs |
Stromverbindung | None | None |
Board-Nummer | — | 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-Modell | 6.4 | 6.6 |