AMD Radeon HD 6550A vs NVIDIA GRID A100B
Processeur graphique
| Nom du GPU | Pinewood | GA100 |
|---|---|---|
| GPU variant | Pinewood PRO | — |
| Architecture | TeraScale 2 | Ampere |
| Fonderie | TSMC | TSMC |
| Taille du processus | 40 nm | 7 nm |
| Transistors | 627 million | 54,200 million |
| Taille des matrices | 104 mm² | 826 mm² |
Graphiques mobiles
| Date de publication | Feb 7th, 2011 | — |
|---|---|---|
| Génération | All-In-One (HD 6000) | — |
| Production | End-of-life | — |
| Bus interface | MXM-A (3.0) | — |
Vitesse clock
| GPU clock | 550 MHz | — |
|---|---|---|
| Mémoire clock | 800 MHz 1600 Mbps effective | 1215 MHz 2.4 Gbps effective |
| Base clock | — | 900 MHz |
| Boost clock | — | 1005 MHz |
Mémoire
| Taille de la mémoire | 2 GB | 48 GB |
|---|---|---|
| Type de mémoire | GDDR3 | HBM2E |
| Bus mémoire | 128 bit | 6144 bit |
| Bande passante | 25.60 GB/s | 1,866 GB/s |
Render Config
| Shading units | 480 | 6912 |
|---|---|---|
| TMUs | 24 | 432 |
| ROPs | 8 | 192 |
| Compute units | 6 | — |
| Cache L1 | 8 KB (per CU) | 192 KB (per SM) |
| Cache L2 | 256 KB | 48 MB |
| Comptage SM | — | 108 |
| Tensor cores | — | 432 |
Performance théorique
| Taux de pixel | 4.400 GPixel/s | 193.0 GPixel/s |
|---|---|---|
| Taux de texture | 13.20 GTexel/s | 434.2 GTexel/s |
| FP32 (float) performance | 528.0 GFLOPS | 13.89 TFLOPS |
| FP16 (half) performance | — | 55.57 TFLOPS (4:1) |
| FP64 (double) performance | — | 6.947 TFLOPS (1:2) |
Board Design
| Taille Slot | MXM Module | IGP |
|---|---|---|
| TDP | 35 W | 400 W |
| Sorties | No outputs | No outputs |
| Suggestions pour le PSU | — | 800 W |
| Connecteurs de puissance | — | None |
Caractéristiques graphiques
| DirectX | 11.2 (11_0) | — |
|---|---|---|
| OpenGL | 4.4 | — |
| OpenCL | 1.2 | 3.0 |
| Vulkan | — | — |
| Shader model | 5.0 | — |
| CUDA | — | 8.0 |
AUTRES CARACTÉRISTIQUES
Carte graphique
| Date de publication | — | May 14th, 2020 |
|---|---|---|
| Génération | — | GRID |
| Production | — | Active |
| Bus interface | — | PCIe 4.0 x16 |
