AMD Radeon HD 8670M vs NVIDIA GeForce RTX 3050 Ti Mobile
Processeur graphique
| Nom du GPU | Mars | GA106 |
|---|---|---|
| GPU variant | Mars XT | — |
| Architecture | GCN 1.0 | Ampere |
| Fonderie | TSMC | Samsung |
| Taille du processus | 28 nm | 8 nm |
| Transistors | 950 million | 13,250 million |
| Taille des matrices | 77 mm² | 276 mm² |
Graphiques mobiles
| Date de publication | Mar 1st, 2013 | May 11th, 2021 |
|---|---|---|
| Génération | Solar System (HD 8600M) | GeForce 30 Mobile |
| Production | End-of-life | Active |
| Bus interface | PCIe 3.0 x8 | PCIe 4.0 x16 |
| Prédécesseur | London | GeForce 20 Mobile |
| Successeur | Crystal System | — |
| gpu.details.availability | — | 2021 |
Vitesse clock
| Base clock | 775 MHz | 735 MHz |
|---|---|---|
| Boost clock | 825 MHz | 1035 MHz |
| Mémoire clock | 900 MHz 1800 Mbps effective | 1500 MHz 12 Gbps effective |
Mémoire
| Taille de la mémoire | 1024 MB | 4 GB |
|---|---|---|
| Type de mémoire | DDR3 | GDDR6 |
| Bus mémoire | 64 bit | 128 bit |
| Bande passante | 14.40 GB/s | 192.0 GB/s |
Render Config
| Shading units | 384 | 2560 |
|---|---|---|
| TMUs | 24 | 80 |
| ROPs | 8 | 48 |
| Compute units | 6 | — |
| Cache L1 | 16 KB (per CU) | 128 KB (per SM) |
| Cache L2 | 128 KB | 2 MB |
| Comptage SM | — | 20 |
| Tensor cores | — | 80 |
| RT cores | — | 20 |
Performance théorique
| Taux de pixel | 6.600 GPixel/s | 49.68 GPixel/s |
|---|---|---|
| Taux de texture | 19.80 GTexel/s | 82.80 GTexel/s |
| FP32 (float) performance | 633.6 GFLOPS | 5.299 TFLOPS |
| FP16 (half) performance | — | 5.299 TFLOPS (1:1) |
| FP64 (double) performance | — | 82.80 GFLOPS (1:64) |
Board Design
| TDP | unknown | 75 W |
|---|---|---|
| Sorties | No outputs | No outputs |
| Taille Slot | — | IGP |
| Connecteurs de puissance | — | None |
Caractéristiques graphiques
| DirectX | 12 (11_1) | 12 Ultimate (12_2) |
|---|---|---|
| OpenGL | 4.6 | 4.6 |
| OpenCL | 1.2 | 3.0 |
| Vulkan | 1.2 | 1.2 |
| Shader model | 5.1 | 6.6 |
| CUDA | — | 8.6 |
