Nom du GPU | Neptune | Polaris 22 |
---|---|---|
GPU variant | Neptune XT (216-0847000) | VegaM XL (D715A714) |
Architecture | GCN 1.0 | GCN 4.0 |
Fonderie | TSMC | GlobalFoundries |
Taille du processus | 28 nm | 14 nm |
Transistors | 2,800 million | 5,000 million |
Taille des matrices | 212 mm² | 208 mm² |
Date de publication | May 14th, 2013 | Feb 1st, 2018 |
---|---|---|
Génération | Solar System (HD 8900M) | Vega (Vega M) |
Production | End-of-life | Active |
Bus interface | PCIe 3.0 x16 | IGP |
Prédécesseur | London | — |
Successeur | Crystal System | — |
Base clock | 850 MHz | 931 MHz |
---|---|---|
Boost clock | 900 MHz | 1011 MHz |
Mémoire clock | 1200 MHz 4.8 Gbps effective | 700 MHz 1400 Mbps effective |
Taille de la mémoire | 4 GB | 4 GB |
---|---|---|
Type de mémoire | GDDR5 | HBM2 |
Bus mémoire | 256 bit | 1024 bit |
Bande passante | 153.6 GB/s | 179.2 GB/s |
Shading units | 1280 | 1280 |
---|---|---|
TMUs | 80 | 80 |
ROPs | 32 | 32 |
Compute units | 20 | 20 |
Cache L1 | 16 KB (per CU) | 16 KB (per CU) |
Cache L2 | 512 KB | 1024 KB |
Taux de pixel | 28.80 GPixel/s | 32.35 GPixel/s |
---|---|---|
Taux de texture | 72.00 GTexel/s | 80.88 GTexel/s |
FP32 (float) performance | 2.304 TFLOPS | 2.588 TFLOPS |
FP64 (double) performance | 144.0 GFLOPS (1:16) | 161.8 GFLOPS (1:16) |
FP16 (half) performance | — | 2.588 TFLOPS (1:1) |
Taille Slot | MXM Module | IGP |
---|---|---|
TDP | 100 W | 65 W |
Sorties | No outputs | No outputs |
Board number | — | D136 |
DirectX | 12 (11_1) | 12 (12_0) |
---|---|---|
OpenGL | 4.6 | 4.6 |
OpenCL | 1.2 | 2.1 |
Vulkan | 1.2 | 1.2 |
Shader model | 5.1 | 6.4 |