AMD Radeon RX 6700M vs NVIDIA GeForce RTX 3080 Ti Mobile
Processeur graphique
| Nom du GPU | Navi 22 | GA103 |
|---|---|---|
| GPU variant | Navi 22 | GA103S |
| Architecture | RDNA 2.0 | Ampere |
| Fonderie | TSMC | Samsung |
| Taille du processus | 7 nm | 8 nm |
| Transistors | 17,200 million | unknown |
| Taille des matrices | 335 mm² | unknown |
Graphiques mobiles
| Date de publication | Unknown | 2022 |
|---|---|---|
| Génération | Mobility Radeon (Navi II) | GeForce 30 Mobile |
| Production | Active | Unreleased |
| Bus interface | PCIe 4.0 x16 | PCIe 4.0 x16 |
| gpu.details.availability | — | — |
| Prédécesseur | — | GeForce 20 Mobile |
Vitesse clock
| Base clock | 1489 MHz | 975 MHz |
|---|---|---|
| Boost clock | 2250 MHz | 1395 MHz |
| Mémoire clock | 2000 MHz 16 Gbps effective | 2000 MHz 16 Gbps effective |
Mémoire
| Taille de la mémoire | 10 GB | 16 GB |
|---|---|---|
| Type de mémoire | GDDR6 | GDDR6 |
| Bus mémoire | 160 bit | 256 bit |
| Bande passante | 320.0 GB/s | 512.0 GB/s |
Render Config
| Shading units | 2304 | 7424 |
|---|---|---|
| TMUs | 144 | 232 |
| ROPs | 64 | 96 |
| Compute units | 36 | — |
| RT cores | 36 | 58 |
| gpu.details.l0-cache | 32 KB per WGP | — |
| Cache L1 | 128 KB per Array | 128 KB (per SM) |
| Cache L2 | 3 MB | 4 MB |
| gpu.details.l3-cache | 96 MB | — |
| Comptage SM | — | 58 |
| Tensor cores | — | 232 |
Performance théorique
| Taux de pixel | 144.0 GPixel/s | 133.9 GPixel/s |
|---|---|---|
| Taux de texture | 324.0 GTexel/s | 323.6 GTexel/s |
| FP16 (half) performance | 20.74 TFLOPS (2:1) | 20.71 TFLOPS (1:1) |
| FP32 (float) performance | 10.37 TFLOPS | 20.71 TFLOPS |
| FP64 (double) performance | 648.0 GFLOPS (1:16) | 323.6 GFLOPS (1:64) |
Board Design
| Taille Slot | IGP | — |
|---|---|---|
| TDP | 180 W | 175 W |
| Sorties | No outputs | No outputs |
| Connecteurs de puissance | None | None |
Caractéristiques graphiques
| DirectX | 12 Ultimate (12_2) | 12 Ultimate (12_2) |
|---|---|---|
| OpenGL | 4.6 | 4.6 |
| OpenCL | 2.1 | 3.0 |
| Vulkan | 1.2 | 1.2 |
| Shader model | 6.5 | 6.5 |
| CUDA | — | 8.6 |
