Nom du GPU | GA102 | GK210 |
---|---|---|
GPU variant | GA102-300-A1 | GK210-885-A1 |
Architecture | Ampere | Kepler 2.0 |
Fonderie | TSMC | TSMC |
Taille du processus | 7 nm | 28 nm |
Transistors | 40,000 million | 7,100 million |
Taille des matrices | 627 mm² | 561 mm² |
Date de publication | 2020 | Nov 17th, 2014 |
---|---|---|
Génération | GeForce 30 | Tesla |
Production | Unreleased | End-of-life |
Prix de lancement | 1,399 USD | — |
Bus interface | PCIe 4.0 x16 | PCIe 3.0 x16 |
Base clock | 1410 MHz | 562 MHz |
---|---|---|
Boost clock | 1740 MHz | 824 MHz |
Mémoire clock | 2375 MHz 19000 MHz effective | 1253 MHz 5 Gbps effective |
Taille de la mémoire | 12 GB | 12 GB |
---|---|---|
Type de mémoire | GDDR6X | GDDR5 |
Bus mémoire | 384 bit | 384 bit |
Bande passante | 912.0 GB/s | 240.6 GB/s |
Shading units | 5120 | 2496 |
---|---|---|
TMUs | 320 | 208 |
ROPs | 96 | 48 |
Comptage SM | 80 | — |
Tensor cores | 640 | — |
RT cores | 160 | — |
Cache L1 | 192 KB (per SM) | 16 KB (per SMX) |
Cache L2 | 6 MB | 1536 KB |
Comptage SMX | — | 13 |
Taux de pixel | 167.0 GPixel/s | 42.85 GPixel/s |
---|---|---|
Taux de texture | 556.8 GTexel/s | 171.4 GTexel/s |
FP16 (half) performance | 35.64 TFLOPS (2:1) | — |
FP32 (float) performance | 17.82 TFLOPS | 4.113 TFLOPS |
FP64 (double) performance | 556.8 GFLOPS (1:32) | 1,371 GFLOPS (1:3) |
Taille Slot | Dual-slot | Dual-slot |
---|---|---|
Longueur | 267 mm 10.5 inches | 267 mm 10.5 inches |
TDP | 320 W | 300 W |
Suggestions pour le PSU | 700 W | 700 W |
Sorties | 1x HDMI3x DisplayPort1x USB Type-C | No outputs |
Connecteurs de puissance | 1x 12-pin | 1x 8-pin |
Board number | PG132 SKU 20 | P2080 SKU 200 |
DirectX | 12 Ultimate (12_2) | 12 (11_1) |
---|---|---|
OpenGL | 4.6 | 4.6 |
OpenCL | 2.0 | 3.0 |
Vulkan | 1.2.140 | 1.1 |
CUDA | 8.5 | 3.7 |
Shader model | 6.5 | 5.1 |