Nom du GPU | Mach32 | GP108 |
---|---|---|
GPU variant | Mach32-03 (2106880003) | N17S-G2-A1 |
Architecture | Mach | Pascal |
Taille du processus | 700 nm | 14 nm |
Transistors | unknown | 1,800 million |
Taille des matrices | 90 mm² | 74 mm² |
Fonderie | — | Samsung |
Date de publication | 1992 | — |
---|---|---|
Génération | Mach 32 | — |
Production | End-of-life | — |
Prix de lancement | 799 USD | — |
Bus interface | PCI | — |
Prédécesseur | Mach 8 | — |
Successeur | Mach 64 | — |
GPU clock | 10 MHz | — |
---|---|---|
Mémoire clock | 10 MHz | 1502 MHz 6 Gbps effective |
Base clock | — | 1519 MHz |
Boost clock | — | 1582 MHz |
Taille de la mémoire | 2 MB | 2 GB |
---|---|---|
Type de mémoire | DRAM | GDDR5 |
Bus mémoire | 64 bit | 64 bit |
Bande passante | 80.00 MB/s | 48.06 GB/s |
Pixel shaders | 1 | — |
---|---|---|
Vertex shaders | 0 | — |
TMUs | 0 | 24 |
ROPs | 1 | 16 |
Shading units | — | 384 |
Comptage SM | — | 3 |
Cache L1 | — | 48 KB (per SM) |
Cache L2 | — | 512 KB |
Taux de pixel | 10.00 MPixel/s | 25.31 GPixel/s |
---|---|---|
Taux de texture | 0 MTexel/s | 37.97 GTexel/s |
FP16 (half) performance | — | 18.98 GFLOPS (1:64) |
FP32 (float) performance | — | 1,215 GFLOPS |
FP64 (double) performance | — | 37.97 GFLOPS (1:32) |
Taille Slot | Single-slot | IGP |
---|---|---|
TDP | unknown | 25 W |
Suggestions pour le PSU | 200 W | — |
Sorties | 1x VGA | No outputs |
Board number | 193 | — |
Connecteurs de puissance | — | None |
DirectX | — | 12 (12_1) |
---|---|---|
OpenGL | — | 4.6 |
OpenCL | — | 3.0 |
Vulkan | — | 1.2 |
Pixel shader | — | — |
Vertex shader | — | — |
CUDA | — | 6.1 |
Shader model | — | 6.4 |
Date de publication | — | Feb 21st, 2019 |
---|---|---|
Génération | — | GeForce MX (2xx) |
Production | — | End-of-life |
Bus interface | — | PCIe 3.0 x4 |