Nom du GPU | Coffee Lake GT1 | GA102 |
---|---|---|
Architecture | Generation 9.5 | Ampere |
Fonderie | Intel | TSMC |
Taille du processus | 14 nm+++ | 7 nm |
Transistors | unknown | 40,000 million |
Taille des matrices | unknown | 627 mm² |
GPU variant | — | GA102-300-A1 |
Date de publication | Apr 3rd, 2018 | — |
---|---|---|
Génération | HD Graphics-M (Coffee Lake) | — |
Production | Active | — |
Bus interface | Ring Bus | — |
Critiques | 1 in our database | — |
Base clock | 300 MHz | 1410 MHz |
---|---|---|
Boost clock | 900 MHz | 1740 MHz |
Mémoire clock | System Shared | 2375 MHz 19000 MHz effective |
Taille de la mémoire | System Shared | 12 GB |
---|---|---|
Type de mémoire | System Shared | GDDR6X |
Bus mémoire | System Shared | 384 bit |
Bande passante | System Dependent | 912.0 GB/s |
Shading units | 96 | 5120 |
---|---|---|
TMUs | 12 | 320 |
ROPs | 2 | 96 |
Unités d'exécution | 12 | — |
Comptage SM | — | 80 |
Tensor cores | — | 640 |
RT cores | — | 160 |
Cache L1 | — | 192 KB (per SM) |
Cache L2 | — | 6 MB |
Taux de pixel | 1.800 GPixel/s | 167.0 GPixel/s |
---|---|---|
Taux de texture | 10.80 GTexel/s | 556.8 GTexel/s |
FP16 (half) performance | 345.6 GFLOPS (2:1) | 35.64 TFLOPS (2:1) |
FP32 (float) performance | 172.8 GFLOPS | 17.82 TFLOPS |
FP64 (double) performance | 43.20 GFLOPS (1:4) | 556.8 GFLOPS (1:32) |
Taille Slot | IGP | Dual-slot |
---|---|---|
TDP | 15 W | 320 W |
Sorties | No outputs | 1x HDMI3x DisplayPort1x USB Type-C |
Longueur | — | 267 mm 10.5 inches |
Suggestions pour le PSU | — | 700 W |
Connecteurs de puissance | — | 1x 12-pin |
Board number | — | PG132 SKU 20 |
DirectX | 12 (12_1) | 12 Ultimate (12_2) |
---|---|---|
OpenGL | 4.6 | 4.6 |
OpenCL | 3.0 | 2.0 |
Vulkan | 1.2 | 1.2.140 |
Shader model | 6.4 | 6.5 |
CUDA | — | 8.5 |
Date de publication | — | 2020 |
---|---|---|
Génération | — | GeForce 30 |
Production | — | Unreleased |
Prix de lancement | — | 1,399 USD |
Bus interface | — | PCIe 4.0 x16 |