Nom du GPU | Aldebaran | Vega 12 |
---|---|---|
GPU variant | Aldebaran XT | Vega 12 XLA |
Architecture | CDNA 2.0 | GCN 5.0 |
Fonderie | TSMC | GlobalFoundries |
Taille du processus | 6 nm | 14 nm |
Transistors | 58,200 million | unknown |
Taille des matrices | unknown | unknown |
Date de publication | Nov 8th, 2021 | — |
---|---|---|
Génération | Radeon Instinct | — |
Production | unknown | — |
Bus interface | PCIe 4.0 x16 | — |
Base clock | 1000 MHz | 815 MHz |
---|---|---|
Boost clock | 1700 MHz | 1190 MHz |
Mémoire clock | 1600 MHz 3.2 Gbps effective | 1200 MHz 2.4 Gbps effective |
Taille de la mémoire | 128 GB | 4 GB |
---|---|---|
Type de mémoire | HBM2E | HBM2 |
Bus mémoire | 8192 bit | 1024 bit |
Bande passante | 3,277 GB/s | 307.2 GB/s |
Shading units | 14080 | 1024 |
---|---|---|
TMUs | 880 | 64 |
ROPs | 0 | 32 |
Compute units | 220 | 16 |
Cache L1 | 16 KB (per CU) | 16 KB (per CU) |
Cache L2 | 16 MB | 1024 KB |
gpu.details.mcm | 2 | — |
Taux de pixel | 0 MPixel/s | 38.08 GPixel/s |
---|---|---|
Taux de texture | 1,496 GTexel/s | 76.16 GTexel/s |
FP16 (half) performance | 383.0 TFLOPS (8:1) | 4.874 TFLOPS (2:1) |
FP32 (float) performance | 47.87 TFLOPS | 2.437 TFLOPS |
FP64 (double) performance | 47.87 TFLOPS (1:1) | 152.3 GFLOPS (1:16) |
Taille Slot | Dual-slot | IGP |
---|---|---|
Longueur | 267 mm 10.5 inches | — |
Taille | 111 mm 4.4 inches | — |
TDP | 500 W | 75 W |
Suggestions pour le PSU | 900 W | — |
Sorties | No outputs | No outputs |
Connecteurs de puissance | 2x 8-pin | — |
DirectX | — | 12 (12_1) |
---|---|---|
OpenGL | — | 4.6 |
OpenCL | 3.0 | 2.1 |
Vulkan | — | 1.2 |
Shader model | — | 6.3 |
Date de publication | — | Nov 14th, 2018 |
---|---|---|
Génération | — | Radeon Pro Mac (Vega Series) |
Production | — | Active |
Bus interface | — | PCIe 3.0 x16 |