Matrox Millennium G200 vs NVIDIA Tesla K8
Processeur graphique
| Nom du GPU | Eclipse | GK104 |
|---|---|---|
| GPU variant | MGA-G200A-B | — |
| Architecture | G200 | Kepler |
| Fonderie | UMC | TSMC |
| Taille du processus | 350 nm | 28 nm |
| Transistors | 10 million | 3,540 million |
| Taille des matrices | unknown | 294 mm² |
Carte graphique
| Date de publication | 1998 | Sep 16th, 2014 |
|---|---|---|
| Génération | G Series | Tesla |
| Production | End-of-life | End-of-life |
| Bus interface | PCI | PCIe 2.0 x16 |
| Critiques | 5 in our database | — |
Vitesse clock
| GPU clock | 84 MHz | — |
|---|---|---|
| Mémoire clock | 143 MHz | 1250 MHz 5 Gbps effective |
| Base clock | — | 693 MHz |
| Boost clock | — | 811 MHz |
Mémoire
| Taille de la mémoire | 8 MB | 8 GB |
|---|---|---|
| Type de mémoire | SGR | GDDR5 |
| Bus mémoire | 64 bit | 256 bit |
| Bande passante | 1.144 GB/s | 160.0 GB/s |
Render Config
| Pixel shaders | 1 | — |
|---|---|---|
| Vertex shaders | 1 | — |
| TMUs | 1 | 128 |
| ROPs | 1 | 32 |
| Shading units | — | 1536 |
| Comptage SMX | — | 8 |
| Cache L1 | — | 16 KB (per SMX) |
| Cache L2 | — | 512 KB |
Performance théorique
| Taux de pixel | 84.00 MPixel/s | 25.95 GPixel/s |
|---|---|---|
| Vertex rate | 21.00 MVertices/s | — |
| Taux de texture | 84.00 MTexel/s | 103.8 GTexel/s |
| FP32 (float) performance | — | 2.491 TFLOPS |
| FP64 (double) performance | — | 103.8 GFLOPS (1:24) |
Board Design
| Taille Slot | Single-slot | Single-slot |
|---|---|---|
| Longueur | 150 mm 5.9 inches | 241 mm 9.5 inches |
| TDP | unknown | 100 W |
| Suggestions pour le PSU | 200 W | 300 W |
| Sorties | 1x DVI1x VGA | No outputs |
| Connecteurs de puissance | None | 1x 6-pin |
| Board number | — | P2004 |
Caractéristiques graphiques
| DirectX | 5.0 | 12 (11_0) |
|---|---|---|
| OpenGL | None | 4.6 |
| OpenCL | None | 3.0 |
| Vulkan | — | 1.1 |
| Pixel shader | None | — |
| Vertex shader | None | — |
| CUDA | — | 3.0 |
| Shader model | — | 5.1 |
