Nom du GPU | Vega 20 | TU117 |
---|---|---|
GPU variant | Vega 20 GLXT | N18P-G61-MP2 |
Architecture | GCN 5.1 | Turing |
Fonderie | TSMC | TSMC |
Taille du processus | 7 nm | 12 nm |
Transistors | 13,230 million | 4,700 million |
Taille des matrices | 331 mm² | 200 mm² |
Date de publication | Nov 18th, 2018 | — |
---|---|---|
Génération | Radeon Instinct | — |
Production | Active | — |
Bus interface | PCIe 4.0 x16 | — |
Base clock | 1200 MHz | 1380 MHz |
---|---|---|
Boost clock | 1746 MHz | 1515 MHz |
Mémoire clock | 1000 MHz 2 Gbps effective | 1500 MHz 12000 MHz effective |
Taille de la mémoire | 16 GB | 4 GB |
---|---|---|
Type de mémoire | HBM2 | GDDR6 |
Bus mémoire | 4096 bit | 128 bit |
Bande passante | 1,024 GB/s | 192.0 GB/s |
Shading units | 3840 | 1024 |
---|---|---|
TMUs | 240 | 64 |
ROPs | 64 | 32 |
Compute units | 60 | — |
Cache L1 | 16 KB (per CU) | 64 KB (per SM) |
Cache L2 | 4 MB | 1024 KB |
Comptage SM | — | 16 |
Taux de pixel | 111.7 GPixel/s | 48.48 GPixel/s |
---|---|---|
Taux de texture | 419.0 GTexel/s | 96.96 GTexel/s |
FP16 (half) performance | 26.82 TFLOPS (2:1) | 6.205 TFLOPS (2:1) |
FP32 (float) performance | 13.41 TFLOPS | 3.103 TFLOPS |
FP64 (double) performance | 6.705 TFLOPS (1:2) | 96.96 GFLOPS (1:32) |
Taille Slot | Dual-slot | — |
---|---|---|
Longueur | 267 mm 10.5 inches | — |
TDP | 300 W | 50 W |
Suggestions pour le PSU | 700 W | — |
Sorties | 1x mini-DisplayPort | No outputs |
Connecteurs de puissance | 1x 6-pin + 1x 8-pin | None |
DirectX | 12 (12_1) | 12 (12_1) |
---|---|---|
OpenGL | 4.6 | 4.6 |
OpenCL | 2.1 | 1.2 |
Vulkan | 1.2 | 1.2.140 |
Shader model | 6.4 | 6.5 |
CUDA | — | 7.5 |
Date de publication | — | Apr 15th, 2020 |
---|---|---|
Génération | — | GeForce 16 Mobile |
Production | — | Active |
Bus interface | — | PCIe 3.0 x16 |
Critiques | — | 44 in our database |