GPU Name | GA102 | GV100 |
---|---|---|
GPU Variante | GA102-890-A1 | — |
Architektur | Ampere | Volta |
Hersteller | Samsung | TSMC |
Prozessgröße | 8 nm | 12 nm |
Transistoren | 28,300 million | 21,100 million |
Chipgröße | 628 mm² | 815 mm² |
Veröffentlichungsdatum | Apr 12th, 2021 | Mar 27th, 2018 |
---|---|---|
gpu.details.availability | 2021 | — |
Generation | Tesla | Tesla |
Produktion | Active | End-of-life |
Businterface | PCIe 4.0 x16 | PCIe 3.0 x16 |
Basis-Taktfrequenz | 885 MHz | 1230 MHz |
---|---|---|
Boost-Takt | 1695 MHz | 1380 MHz |
Speicher-Taktung | 1563 MHz 12.5 Gbps effective | 876 MHz 1752 Mbps effective |
Speichergröße | 24 GB | 32 GB |
---|---|---|
Speichertyp | GDDR6 | HBM2 |
Speicherbus | 384 bit | 4096 bit |
Bandbreite | 600.2 GB/s | 897.0 GB/s |
Shading-Einheiten | 9216 | 5120 |
---|---|---|
TMUs | 288 | 320 |
ROPs | 96 | 128 |
Anzahl SM | 72 | 80 |
Tensor Cores | 288 | 640 |
RT-Kerne | 72 | — |
Cache L1 | 128 KB (per SM) | 128 KB (per SM) |
Cache L2 | 6 MB | 6 MB |
Pixelrate | 162.7 GPixel/s | 176.6 GPixel/s |
---|---|---|
Texturrate | 488.2 GTexel/s | 441.6 GTexel/s |
FP16 (half) Leistung | 31.24 TFLOPS (1:1) | 28.26 TFLOPS (2:1) |
FP32 (float) Leistung | 31.24 TFLOPS | 14.13 TFLOPS |
FP64 (double) Leistung | 976.3 GFLOPS (1:32) | 7.066 TFLOPS (1:2) |
Steckplatzbreite | Single-slot | Dual-slot |
---|---|---|
Länge | 267 mm 10.5 inches | — |
Breite | 112 mm 4.4 inches | — |
TDP | 150 W | 250 W |
Vorgeschlagenes Netzteil | 450 W | 600 W |
Ausgabe | No outputs | No outputs |
Stromverbindung | 8-pin EPS | 2x 8-pin |
Board-Nummer | PG133 SKU 215 | PG500 SKU 200 |
DirectX | 12 Ultimate (12_2) | 12 (12_1) |
---|---|---|
OpenGL | 4.6 | 4.6 |
OpenCL | 3.0 | 3.0 |
Vulkan | 1.2 | 1.2 |
CUDA | 8.6 | 7.0 |
Shader-Modell | 6.6 | 6.6 |