GPU Name | GA102 | GA102 |
---|---|---|
GPU Variante | GA102-300-A1 | — |
Architektur | Ampere | Ampere |
Hersteller | TSMC | Samsung |
Prozessgröße | 7 nm | 8 nm |
Transistoren | 40,000 million | 28,300 million |
Chipgröße | 627 mm² | 628 mm² |
Veröffentlichungsdatum | 2020 | Apr 12th, 2021 |
---|---|---|
Generation | GeForce 30 | Quadro |
Produktion | Unreleased | Active |
Originalpreis | 1,399 USD | — |
Businterface | PCIe 4.0 x16 | PCIe 4.0 x16 |
Basis-Taktfrequenz | 1410 MHz | 1170 MHz |
---|---|---|
Boost-Takt | 1740 MHz | 1695 MHz |
Speicher-Taktung | 2375 MHz 19000 MHz effective | 2000 MHz 16 Gbps effective |
Speichergröße | 12 GB | 24 GB |
---|---|---|
Speichertyp | GDDR6X | GDDR6 |
Speicherbus | 384 bit | 384 bit |
Bandbreite | 912.0 GB/s | 768.0 GB/s |
Shading-Einheiten | 5120 | 8192 |
---|---|---|
TMUs | 320 | 256 |
ROPs | 96 | 96 |
Anzahl SM | 80 | 64 |
Tensor Cores | 640 | 256 |
RT-Kerne | 160 | 64 |
Cache L1 | 192 KB (per SM) | 128 KB (per SM) |
Cache L2 | 6 MB | 6 MB |
Pixelrate | 167.0 GPixel/s | 162.7 GPixel/s |
---|---|---|
Texturrate | 556.8 GTexel/s | 433.9 GTexel/s |
FP16 (half) Leistung | 35.64 TFLOPS (2:1) | 27.77 TFLOPS (1:1) |
FP32 (float) Leistung | 17.82 TFLOPS | 27.77 TFLOPS |
FP64 (double) Leistung | 556.8 GFLOPS (1:32) | 867.8 GFLOPS (1:32) |
Steckplatzbreite | Dual-slot | Dual-slot |
---|---|---|
Länge | 267 mm 10.5 inches | 267 mm 10.5 inches |
TDP | 320 W | 230 W |
Vorgeschlagenes Netzteil | 700 W | 550 W |
Ausgabe | 1x HDMI3x DisplayPort1x USB Type-C | 4x DisplayPort |
Stromverbindung | 1x 12-pin | 1x 8-pin |
Board-Nummer | PG132 SKU 20 | — |
Breite | — | 112 mm 4.4 inches |
DirectX | 12 Ultimate (12_2) | 12 Ultimate (12_2) |
---|---|---|
OpenGL | 4.6 | 4.6 |
OpenCL | 2.0 | 3.0 |
Vulkan | 1.2.140 | 1.2 |
CUDA | 8.5 | 8.6 |
Shader-Modell | 6.5 | 6.6 |