NVIDIA GeForce FX 5600 vs NVIDIA Tesla V100 FHHL
Processeur graphique
| Nom du GPU | NV31 | GV100 |
|---|---|---|
| GPU variant | 5600 | — |
| Architecture | Rankine | Volta |
| Fonderie | TSMC | TSMC |
| Taille du processus | 130 nm | 12 nm |
| Transistors | 80 million | 21,100 million |
| Taille des matrices | 121 mm² | 815 mm² |
Carte graphique
| Date de publication | Mar 6th, 2003 | Mar 27th, 2018 |
|---|---|---|
| Génération | GeForce FX | Tesla |
| Production | End-of-life | End-of-life |
| Bus interface | AGP 8x | PCIe 3.0 x16 |
| Critiques | 2 in our database | — |
| Prédécesseur | GeForce 4 Ti | — |
| Successeur | GeForce 6 AGP | — |
Vitesse clock
| GPU clock | 325 MHz | — |
|---|---|---|
| Mémoire clock | 250 MHz 500 Mbps effective | 810 MHz 1620 Mbps effective |
| Base clock | — | 937 MHz |
| Boost clock | — | 1290 MHz |
Mémoire
| Taille de la mémoire | 128 MB | 16 GB |
|---|---|---|
| Type de mémoire | DDR | HBM2 |
| Bus mémoire | 128 bit | 4096 bit |
| Bande passante | 8.000 GB/s | 829.4 GB/s |
Render Config
| Pixel shaders | 4 | — |
|---|---|---|
| Vertex shaders | 2 | — |
| TMUs | 4 | 320 |
| ROPs | 4 | 128 |
| Shading units | — | 5120 |
| Comptage SM | — | 80 |
| Tensor cores | — | 640 |
| Cache L1 | — | 128 KB (per SM) |
| Cache L2 | — | 6 MB |
Performance théorique
| Taux de pixel | 1.300 GPixel/s | 165.1 GPixel/s |
|---|---|---|
| Vertex rate | 162.5 MVertices/s | — |
| Taux de texture | 1.300 GTexel/s | 412.8 GTexel/s |
| FP16 (half) performance | — | 26.42 TFLOPS (2:1) |
| FP32 (float) performance | — | 13.21 TFLOPS |
| FP64 (double) performance | — | 6.605 TFLOPS (1:2) |
Board Design
| Taille Slot | Single-slot | Single-slot |
|---|---|---|
| TDP | 37 W | 250 W |
| Suggestions pour le PSU | 200 W | 600 W |
| Sorties | 1x DVI1x VGA1x S-Video | No outputs |
| Connecteurs de puissance | None | 1x 8-pin |
Caractéristiques graphiques
| DirectX | 9.0a | 12 (12_1) |
|---|---|---|
| OpenGL | 2.0 | 4.6 |
| OpenCL | — | 3.0 |
| Vulkan | — | 1.2 |
| Pixel shader | 2.0a | — |
| Vertex shader | 2.0a | — |
| CUDA | — | 7.0 |
| Shader model | — | 6.6 |
