Nom du GPU | CW16800-A | GA106 |
---|---|---|
Architecture | Wonder | Ampere |
Taille du processus | 800 nm | 8 nm |
Transistors | unknown | 13,250 million |
Taille des matrices | 90 mm² | 276 mm² |
Fonderie | — | Samsung |
Date de publication | Aug 4th, 1986 | — |
---|---|---|
Génération | MDA/CGA | — |
Successeur | EGA | — |
Production | End-of-life | — |
Bus interface | PCI | — |
GPU clock | 10 MHz | — |
---|---|---|
Mémoire clock | 5 MHz | 1750 MHz 14 Gbps effective |
Base clock | — | 900 MHz |
Boost clock | — | 1425 MHz |
Taille de la mémoire | 32 KB | 6 GB |
---|---|---|
Type de mémoire | DRAM | GDDR6 |
Bus mémoire | 32 bit | 192 bit |
Bande passante | 20.00 MB/s | 336.0 GB/s |
Pixel shaders | 0 | — |
---|---|---|
Vertex shaders | 0 | — |
TMUs | 0 | 120 |
ROPs | 0 | 48 |
Shading units | — | 3840 |
Comptage SM | — | 30 |
Tensor cores | — | 120 |
RT cores | — | 30 |
Cache L1 | — | 128 KB (per SM) |
Cache L2 | — | 3 MB |
Taux de pixel | 0 MPixel/s | 68.40 GPixel/s |
---|---|---|
Taux de texture | 0 MTexel/s | 171.0 GTexel/s |
FP16 (half) performance | — | 10.94 TFLOPS (1:1) |
FP32 (float) performance | — | 10.94 TFLOPS |
FP64 (double) performance | — | 171.0 GFLOPS (1:64) |
Taille Slot | Single-slot | — |
---|---|---|
TDP | unknown | 80 W |
Suggestions pour le PSU | 200 W | — |
Sorties | No outputs | No outputs |
Board number | 168230 | — |
Connecteurs de puissance | — | None |
DirectX | — | 12 Ultimate (12_2) |
---|---|---|
OpenGL | — | 4.6 |
OpenCL | — | 3.0 |
Vulkan | — | 1.2 |
Pixel shader | — | — |
Vertex shader | — | — |
CUDA | — | 8.6 |
Shader model | — | 6.6 |
Date de publication | — | Jan 12th, 2021 |
---|---|---|
gpu.details.availability | — | Feb 2nd, 2021 |
Génération | — | GeForce 30 Mobile |
Prédécesseur | — | GeForce 20 Mobile |
Production | — | Active |
Bus interface | — | PCIe 4.0 x16 |
Critiques | — | 1 in our database |