Nom du GPU | Raven | GP108 |
---|---|---|
Architecture | GCN 5.0 | Pascal |
Fonderie | GlobalFoundries | Samsung |
Taille du processus | 14 nm | 14 nm |
Transistors | 4,940 million | 1,800 million |
Taille des matrices | 210 mm² | 74 mm² |
GPU variant | — | N17S-G0-A1 |
Date de publication | May 10th, 2018 | — |
---|---|---|
Génération | Raven Ridge (Vega) | — |
Production | Active | — |
Bus interface | IGP | — |
Prédécesseur | Great Horned Owl | — |
Successeur | Picasso | — |
Base clock | 300 MHz | 1519 MHz |
---|---|---|
Boost clock | 1280 MHz | 1531 MHz |
Mémoire clock | System Shared | 1502 MHz 6 Gbps effective |
Taille de la mémoire | System Shared | 2 GB |
---|---|---|
Type de mémoire | System Shared | GDDR5 |
Bus mémoire | System Shared | 64 bit |
Bande passante | System Dependent | 48.06 GB/s |
Shading units | 384 | 256 |
---|---|---|
TMUs | 24 | 16 |
ROPs | 8 | 16 |
Compute units | 6 | — |
Comptage SM | — | 2 |
Cache L1 | — | 48 KB (per SM) |
Cache L2 | — | 512 KB |
Taux de pixel | 10.24 GPixel/s | 24.50 GPixel/s |
---|---|---|
Taux de texture | 30.72 GTexel/s | 24.50 GTexel/s |
FP16 (half) performance | 1.966 TFLOPS (2:1) | 12.25 GFLOPS (1:64) |
FP32 (float) performance | 983.0 GFLOPS | 783.9 GFLOPS |
FP64 (double) performance | 61.44 GFLOPS (1:16) | 24.50 GFLOPS (1:32) |
Taille Slot | IGP | IGP |
---|---|---|
TDP | 15 W | 10 W |
Sorties | No outputs | No outputs |
Connecteurs de puissance | — | None |
DirectX | 12 (12_1) | 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 21st, 2019 |
---|---|---|
Génération | — | GeForce MX (2xx) |
Production | — | End-of-life |
Bus interface | — | PCIe 3.0 x4 |