Intel Iris Xe Graphics G4 vs NVIDIA Jetson AGX Xavier GPU
Processeur graphique
| Nom du GPU | Tiger Lake GT2 | GV10B |
|---|---|---|
| Architecture | Generation 12.1 | Volta |
| Fonderie | Intel | TSMC |
| Taille du processus | 10 nm | 12 nm |
| Transistors | unknown | 9,000 million |
| Taille des matrices | 146 mm² | 350 mm² |
Carte graphique intégré
| Date de publication | Sep 2nd, 2020 | Oct 2018 |
|---|---|---|
| Génération | Iris Xe Graphics-M (Tiger Lake) | Tegra |
| Production | Active | End-of-life |
| Bus interface | Ring Bus | IGP |
| Critiques | 1 in our database | — |
| Prix de lancement | — | 899 USD |
Vitesse clock
| Base clock | 300 MHz | 854 MHz |
|---|---|---|
| Boost clock | 1100 MHz | 1377 MHz |
| Mémoire clock | System Shared | System Shared |
Mémoire
| Taille de la mémoire | System Shared | System Shared |
|---|---|---|
| Type de mémoire | System Shared | System Shared |
| Bus mémoire | System Shared | System Shared |
| Bande passante | System Dependent | System Dependent |
Render Config
| Shading units | 384 | 512 |
|---|---|---|
| TMUs | 24 | 32 |
| ROPs | 12 | 16 |
| Unités d'exécution | 48 | — |
| gpu.details.l3-cache | 3.75 MB | — |
| Comptage SM | — | 8 |
| Tensor cores | — | 64 |
| Cache L1 | — | 128 KB (per SM) |
| Cache L2 | — | 512 KB |
Performance théorique
| Taux de pixel | 13.20 GPixel/s | 22.03 GPixel/s |
|---|---|---|
| Taux de texture | 26.40 GTexel/s | 44.06 GTexel/s |
| FP16 (half) performance | 1.690 TFLOPS (2:1) | 2.820 TFLOPS (2:1) |
| FP32 (float) performance | 844.8 GFLOPS | 1,410 GFLOPS |
| FP64 (double) performance | 211.2 GFLOPS (1:4) | 705.0 GFLOPS (1:2) |
Board Design
| Taille Slot | IGP | IGP |
|---|---|---|
| TDP | 15 W | 30 W |
| Sorties | No outputs | No outputs |
| Longueur | — | 100 mm 3.9 inches |
| Taille | — | 87 mm 3.4 inches |
Caractéristiques graphiques
| DirectX | 12 (12_1) | 12 (12_1) |
|---|---|---|
| OpenGL | 4.6 | 4.6 |
| OpenCL | 3.0 | 1.2 |
| Vulkan | 1.2 | 1.2 |
| Shader model | 6.4 | 6.4 |
| CUDA | — | 7.2 |
