NVIDIA CMP 40HX vs NVIDIA P102-101
Processeur graphique
| Nom du GPU | TU106 | GP102 |
|---|---|---|
| GPU variant | TU106-100-A1 | GP102-101-A1 |
| Architecture | Turing | Pascal |
| Fonderie | TSMC | TSMC |
| Taille du processus | 12 nm | 16 nm |
| Transistors | 10,800 million | 11,800 million |
| Taille des matrices | 445 mm² | 471 mm² |
Carte graphique
| Date de publication | Feb 25th, 2021 | 2018 |
|---|---|---|
| Génération | Mining GPUs | Mining GPUs |
| Production | Active | End-of-life |
| Prix de lancement | 699 USD | — |
| Bus interface | PCIe 3.0 x16 | PCIe 3.0 x4 |
Vitesse clock
| Base clock | 1470 MHz | 1557 MHz |
|---|---|---|
| Boost clock | 1650 MHz | 1670 MHz |
| Mémoire clock | 1750 MHz 14 Gbps effective | 2002 MHz 8 Gbps effective |
Mémoire
| Taille de la mémoire | 8 GB | 10 GB |
|---|---|---|
| Type de mémoire | GDDR6 | GDDR5 |
| Bus mémoire | 256 bit | 320 bit |
| Bande passante | 448.0 GB/s | 320.3 GB/s |
Render Config
| Shading units | 2304 | 3200 |
|---|---|---|
| TMUs | 144 | 200 |
| ROPs | 64 | 80 |
| Comptage SM | 36 | 25 |
| Tensor cores | 288 | — |
| RT cores | 36 | — |
| Cache L1 | 64 KB (per SM) | 48 KB (per SM) |
| Cache L2 | 4 MB | 2.5 MB |
Performance théorique
| Taux de pixel | 105.6 GPixel/s | 133.6 GPixel/s |
|---|---|---|
| Taux de texture | 237.6 GTexel/s | 334.0 GTexel/s |
| FP16 (half) performance | 15.21 TFLOPS (2:1) | 167.0 GFLOPS (1:64) |
| FP32 (float) performance | 7.603 TFLOPS | 10.69 TFLOPS |
| FP64 (double) performance | 237.6 GFLOPS (1:32) | 334.0 GFLOPS (1:32) |
Board Design
| Taille Slot | Dual-slot | Dual-slot |
|---|---|---|
| Longueur | 229 mm 9 inches | 267 mm 10.5 inches |
| Taille | 111 mm 4.4 inches | — |
| Taille | 35 mm 1.4 inches | — |
| TDP | 185 W | 250 W |
| Suggestions pour le PSU | 450 W | 600 W |
| Sorties | No outputs | No outputs |
| Connecteurs de puissance | 1x 8-pin | 2x 8-pin |
| Board number | PG161 SKU 100 | PG611 SKU 101 |
Caractéristiques graphiques
| DirectX | 12 Ultimate (12_2) | 12 (12_1) |
|---|---|---|
| OpenGL | 4.6 | 4.6 |
| OpenCL | 3.0 | 3.0 |
| Vulkan | 1.2 | 1.2 |
| CUDA | 7.5 | 6.1 |
| Shader model | 6.6 | 6.4 |
