Nom du GPU | Baffin | M12 |
---|---|---|
GPU variant | Baffin PRO | — |
Architecture | GCN 4.0 | Rage 9 |
Fonderie | GlobalFoundries | TSMC |
Taille du processus | 14 nm | 130 nm |
Transistors | 3,000 million | 76 million |
Taille des matrices | 123 mm² | 92 mm² |
Date de publication | Sep 27th, 2016 | Jun 1st, 2004 |
---|---|---|
Génération | Embedded (9000) | M1x (Mobility 9000) |
Production | Active | End-of-life |
Bus interface | PCIe 3.0 x8 | AGP 4x |
Critiques | — | 2 in our database |
Prédécesseur | — | M9 |
Successeur | — | M2x |
Base clock | 1090 MHz | — |
---|---|---|
Boost clock | 1200 MHz | — |
Mémoire clock | 1750 MHz 7 Gbps effective | 243 MHz 486 Mbps effective |
GPU clock | — | 250 MHz |
Taille de la mémoire | 4 GB | 64 MB |
---|---|---|
Type de mémoire | GDDR5 | DDR |
Bus mémoire | 128 bit | 64 bit |
Bande passante | 112.0 GB/s | 3.888 GB/s |
Shading units | 896 | — |
---|---|---|
TMUs | 48 | 4 |
ROPs | 16 | 4 |
Compute units | 14 | — |
Cache L1 | 16 KB (per CU) | — |
Cache L2 | 1024 KB | — |
Pixel shaders | — | 4 |
Vertex shaders | — | 2 |
Taux de pixel | 19.20 GPixel/s | 1.000 GPixel/s |
---|---|---|
Taux de texture | 57.60 GTexel/s | 1.000 GTexel/s |
FP16 (half) performance | 2.150 TFLOPS (1:1) | — |
FP32 (float) performance | 2.150 TFLOPS | — |
FP64 (double) performance | 134.4 GFLOPS (1:16) | — |
Vertex rate | — | 125.0 MVertices/s |
Taille Slot | Single-slot | — |
---|---|---|
Longueur | 168 mm 6.6 inches | — |
Taille | 111 mm 4.4 inches | — |
TDP | 80 W | unknown |
Sorties | No outputs | No outputs |
Connecteurs de puissance | None | — |
DirectX | 12 (12_0) | 9.0 |
---|---|---|
OpenGL | 4.6 | 2.0 |
OpenCL | 2.1 | — |
Vulkan | 1.2 | — |
Shader model | 6.4 | — |
Pixel shader | — | 2.0 |
Vertex shader | — | 2.0 |