ATI Radeon HD 5970 vs NVIDIA Tesla P100 PCIe 12 GB
Processeur graphique
| Nom du GPU | Hemlock | GP100 |
|---|---|---|
| GPU variant | Hemlock XT (215-0735075) | — |
| Architecture | TeraScale 2 | Pascal |
| Fonderie | TSMC | TSMC |
| Taille du processus | 40 nm | 16 nm |
| Transistors | 2,154 million | 15,300 million |
| Taille des matrices | 334 mm² | 610 mm² |
Carte graphique
| Date de publication | Nov 18th, 2009 | Jun 20th, 2016 |
|---|---|---|
| Génération | Evergreen | Tesla |
| Production | End-of-life | End-of-life |
| Prix de lancement | 699 USD | 4,599 USD |
| Bus interface | PCIe 2.0 x16 | PCIe 3.0 x16 |
| Critiques | 74 in our database | — |
| Prédécesseur | Radeon R700 | — |
| Successeur | Northern Islands | — |
Vitesse clock
| GPU clock | 725 MHz | — |
|---|---|---|
| Mémoire clock | 1000 MHz 4 Gbps effective | 715 MHz 1430 Mbps effective |
| Base clock | — | 1190 MHz |
| Boost clock | — | 1329 MHz |
Mémoire
| Taille de la mémoire | 1024 MB | 12 GB |
|---|---|---|
| Type de mémoire | GDDR5 | HBM2 |
| Bus mémoire | 256 bit | 3072 bit |
| Bande passante | 128.0 GB/s | 549.1 GB/s |
Render Config
| Shading units | 1600 | 3584 |
|---|---|---|
| TMUs | 80 | 224 |
| ROPs | 32 | 96 |
| Compute units | 20 | — |
| Cache L1 | 8 KB (per CU) | 24 KB (per SM) |
| Cache L2 | 512 KB | 3 MB |
| Comptage SM | — | 56 |
Performance théorique
| Taux de pixel | 23.20 GPixel/s | 127.6 GPixel/s |
|---|---|---|
| Taux de texture | 58.00 GTexel/s | 297.7 GTexel/s |
| FP32 (float) performance | 2.320 TFLOPS | 9.526 TFLOPS |
| FP64 (double) performance | 464.0 GFLOPS (1:5) | 4.763 TFLOPS (1:2) |
| FP16 (half) performance | — | 19.05 TFLOPS (2:1) |
Board Design
| Taille Slot | Dual-slot | Dual-slot |
|---|---|---|
| Longueur | 311 mm 12.2 inches | 267 mm 10.5 inches |
| Taille | 111 mm 4.4 inches | — |
| Taille | 40 mm 1.6 inches | — |
| TDP | 294 W | 250 W |
| Suggestions pour le PSU | 600 W | 600 W |
| Sorties | 2x DVI1x mini-DisplayPort | No outputs |
| Connecteurs de puissance | 1x 6-pin + 1x 8-pin | 1x 8-pin |
| Board number | C000-37 | — |
Caractéristiques graphiques
| DirectX | 11.2 (11_0) | 12 (12_1) |
|---|---|---|
| OpenGL | 4.4 | 4.6 |
| OpenCL | 1.2 | 3.0 |
| Vulkan | — | 1.2 |
| Shader model | 5.0 | 6.4 |
| CUDA | — | 6.0 |
