Nom du GPU | Raven | TU104 |
---|---|---|
Architecture | GCN 5.0 | Turing |
Fonderie | GlobalFoundries | TSMC |
Taille du processus | 14 nm | 12 nm |
Transistors | 4,940 million | 13,600 million |
Taille des matrices | 210 mm² | 545 mm² |
GPU variant | — | N19E-Q3-A1 |
Date de publication | Apr 19th, 2018 | — |
---|---|---|
Génération | Raven Ridge (Vega) | — |
Production | Active | — |
Bus interface | IGP | — |
Prédécesseur | Great Horned Owl | — |
Successeur | Picasso | — |
Base clock | 300 MHz | 1110 MHz |
---|---|---|
Boost clock | 1101 MHz | 1560 MHz |
Mémoire clock | System Shared | 1750 MHz 14 Gbps effective |
Taille de la mémoire | System Shared | 8 GB |
---|---|---|
Type de mémoire | System Shared | GDDR6 |
Bus mémoire | System Shared | 256 bit |
Bande passante | System Dependent | 448.0 GB/s |
Shading units | 512 | 2560 |
---|---|---|
TMUs | 32 | 160 |
ROPs | 8 | 64 |
Compute units | 8 | — |
Comptage SM | — | 40 |
Tensor cores | — | 320 |
RT cores | — | 40 |
Cache L1 | — | 64 KB (per SM) |
Cache L2 | — | 4 MB |
Taux de pixel | 8.808 GPixel/s | 99.84 GPixel/s |
---|---|---|
Taux de texture | 35.23 GTexel/s | 249.6 GTexel/s |
FP16 (half) performance | 2.255 TFLOPS (2:1) | 15.97 TFLOPS (2:1) |
FP32 (float) performance | 1,127 GFLOPS | 7.987 TFLOPS |
FP64 (double) performance | 70.46 GFLOPS (1:16) | 249.6 GFLOPS (1:32) |
Taille Slot | IGP | IGP |
---|---|---|
TDP | 35 W | 110 W |
Sorties | No outputs | No outputs |
Connecteurs de puissance | None | None |
DirectX | 12 (12_1) | 12 Ultimate (12_2) |
---|---|---|
OpenGL | 4.6 | 4.6 |
OpenCL | 2.1 | 3.0 |
Vulkan | 1.2 | 1.2 |
Shader model | 6.4 | 6.6 |
CUDA | — | 7.5 |
Date de publication | — | May 27th, 2019 |
---|---|---|
Génération | — | Quadro Mobile (Tx000) |
Production | — | Active |
Bus interface | — | PCIe 3.0 x16 |