Nom du GPU | Polaris 22 | GP106 |
---|---|---|
GPU variant | VegaM XT (C7398773) | GP106-875-A1 |
Architecture | GCN 4.0 | Pascal |
Fonderie | GlobalFoundries | TSMC |
Taille du processus | 14 nm | 16 nm |
Transistors | 5,000 million | 4,400 million |
Taille des matrices | 208 mm² | 200 mm² |
Date de publication | Feb 1st, 2018 | — |
---|---|---|
Génération | Vega (Vega M) | — |
Production | Active | — |
Bus interface | IGP | — |
Base clock | 1063 MHz | 1076 MHz |
---|---|---|
Boost clock | 1190 MHz | 1480 MHz |
Mémoire clock | 800 MHz 1600 Mbps effective | 1752 MHz 7 Gbps effective |
Taille de la mémoire | 4 GB | 5 GB |
---|---|---|
Type de mémoire | HBM2 | GDDR5 |
Bus mémoire | 1024 bit | 160 bit |
Bande passante | 204.8 GB/s | 140.2 GB/s |
Shading units | 1536 | 1024 |
---|---|---|
TMUs | 96 | 64 |
ROPs | 64 | 40 |
Compute units | 24 | — |
Cache L1 | 16 KB (per CU) | 48 KB (per SM) |
Cache L2 | 1024 KB | 1280 KB |
Comptage SM | — | 8 |
Taux de pixel | 76.16 GPixel/s | 59.20 GPixel/s |
---|---|---|
Taux de texture | 114.2 GTexel/s | 94.72 GTexel/s |
FP16 (half) performance | 3.656 TFLOPS (1:1) | 47.36 GFLOPS (1:64) |
FP32 (float) performance | 3.656 TFLOPS | 3.031 TFLOPS |
FP64 (double) performance | 228.5 GFLOPS (1:16) | 94.72 GFLOPS (1:32) |
Taille Slot | IGP | Single-slot |
---|---|---|
TDP | 100 W | 75 W |
Sorties | No outputs | 4x DisplayPort |
Longueur | — | 196 mm 7.7 inches |
Taille | — | 111 mm 4.4 inches |
Suggestions pour le PSU | — | 250 W |
Connecteurs de puissance | — | None |
Board number | — | PG410 SKU 502 |
DirectX | 12 (12_0) | 12 (12_1) |
---|---|---|
OpenGL | 4.6 | 4.6 |
OpenCL | 2.1 | 3.0 |
Vulkan | 1.2 | 1.2 |
Shader model | 6.4 | 6.4 |
CUDA | — | 6.1 |
Date de publication | — | Feb 6th, 2017 |
---|---|---|
Génération | — | Quadro |
Production | — | End-of-life |
Bus interface | — | PCIe 3.0 x16 |