AMD Radeon Instinct MI8 vs NVIDIA Quadro K5000
Processeur graphique
| Nom du GPU | Fiji | GK104 |
|---|---|---|
| GPU variant | Fiji XT | — |
| Architecture | GCN 3.0 | Kepler |
| Fonderie | TSMC | TSMC |
| Taille du processus | 28 nm | 28 nm |
| Transistors | 8,900 million | 3,540 million |
| Taille des matrices | 596 mm² | 294 mm² |
Carte graphique
| Date de publication | Dec 12th, 2016 | Aug 17th, 2012 |
|---|---|---|
| Génération | Radeon Instinct | Quadro |
| Production | End-of-life | End-of-life |
| Bus interface | PCIe 3.0 x16 | PCIe 2.0 x16 |
| Prix de lancement | — | 2,499 USD |
| Critiques | — | 26 in our database |
Vitesse clock
| GPU clock | 1000 MHz | 706 MHz |
|---|---|---|
| Mémoire clock | 500 MHz 1000 Mbps effective | 1350 MHz 5.4 Gbps effective |
Mémoire
| Taille de la mémoire | 4 GB | 4 GB |
|---|---|---|
| Type de mémoire | HBM | GDDR5 |
| Bus mémoire | 4096 bit | 256 bit |
| Bande passante | 512.0 GB/s | 172.8 GB/s |
Render Config
| Shading units | 4096 | 1536 |
|---|---|---|
| TMUs | 256 | 128 |
| ROPs | 64 | 32 |
| Compute units | 64 | — |
| Cache L1 | 16 KB (per CU) | 16 KB (per SMX) |
| Cache L2 | 2 MB | 512 KB |
| Comptage SMX | — | 8 |
Performance théorique
| Taux de pixel | 64.00 GPixel/s | 22.59 GPixel/s |
|---|---|---|
| Taux de texture | 256.0 GTexel/s | 90.37 GTexel/s |
| FP16 (half) performance | 8.192 TFLOPS (1:1) | — |
| FP32 (float) performance | 8.192 TFLOPS | 2.169 TFLOPS |
| FP64 (double) performance | 512.0 GFLOPS (1:16) | 90.37 GFLOPS (1:24) |
Board Design
| Taille Slot | Dual-slot | Dual-slot |
|---|---|---|
| Longueur | 152 mm 6 inches | 267 mm 10.5 inches |
| Taille | 111 mm 4.4 inches | 111 mm 4.4 inches |
| TDP | 175 W | 122 W |
| Suggestions pour le PSU | 450 W | 300 W |
| Sorties | No outputs | 2x DVI2x DisplayPort |
| Connecteurs de puissance | 1x 8-pin | 1x 6-pin |
| Board number | — | P2004 |
Caractéristiques graphiques
| DirectX | 12 (12_0) | 12 (11_0) |
|---|---|---|
| OpenGL | 4.6 | 4.6 |
| OpenCL | 2.0 | 3.0 |
| Vulkan | 1.2 | 1.1 |
| Shader model | 6.3 | 5.1 |
| CUDA | — | 3.0 |
