Intel UHD Graphics 605 vs NVIDIA GeForce RTX 3090 Ti
Processeur graphique
| Nom du GPU | Gemini Lake GT1.5 | GA102 |
|---|---|---|
| Architecture | Generation 9.5 | Ampere |
| Fonderie | Intel | Samsung |
| Taille du processus | 14 nm | 8 nm |
| Transistors | unknown | 28,300 million |
| Taille des matrices | unknown | 628 mm² |
| GPU variant | — | GA102-350-A1 |
Carte graphique intégré
| Date de publication | Dec 11th, 2017 | — |
|---|---|---|
| Génération | HD Graphics-T (Goldmont Plus) | — |
| Production | Active | — |
| Bus interface | Ring Bus | — |
| Critiques | 1 in our database | — |
Vitesse clock
| Base clock | 200 MHz | 1560 MHz |
|---|---|---|
| Boost clock | 750 MHz | 1860 MHz |
| Mémoire clock | System Shared | 1325 MHz 21.2 Gbps effective |
Mémoire
| Taille de la mémoire | System Shared | 24 GB |
|---|---|---|
| Type de mémoire | System Shared | GDDR6X |
| Bus mémoire | System Shared | 384 bit |
| Bande passante | System Dependent | 1,018 GB/s |
Render Config
| Shading units | 144 | 10752 |
|---|---|---|
| TMUs | 18 | 336 |
| ROPs | 3 | 112 |
| Unités d'exécution | 18 | — |
| Comptage SM | — | 84 |
| Tensor cores | — | 336 |
| RT cores | — | 84 |
| Cache L1 | — | 128 KB (per SM) |
| Cache L2 | — | 6 MB |
Performance théorique
| Taux de pixel | 2.250 GPixel/s | 208.3 GPixel/s |
|---|---|---|
| Taux de texture | 13.50 GTexel/s | 625.0 GTexel/s |
| FP16 (half) performance | 432.0 GFLOPS (2:1) | 40.00 TFLOPS (1:1) |
| FP32 (float) performance | 216.0 GFLOPS | 40.00 TFLOPS |
| FP64 (double) performance | 54.00 GFLOPS (1:4) | 625.0 GFLOPS (1:64) |
Board Design
| Taille Slot | IGP | Triple-slot |
|---|---|---|
| TDP | 5 W | 450 W |
| Sorties | No outputs | 1x HDMI3x DisplayPort |
| Longueur | — | 336 mm 13.2 inches |
| Taille | — | 140 mm 5.5 inches |
| Taille | — | 61 mm 2.4 inches |
| Suggestions pour le PSU | — | 850 W |
| Connecteurs de puissance | — | 1x 12-pin |
| Board number | — | PG132 SKU |
Caractéristiques graphiques
| DirectX | 12 (12_1) | 12 Ultimate (12_2) |
|---|---|---|
| OpenGL | 4.6 | 4.6 |
| OpenCL | 3.0 | 3.0 |
| Vulkan | 1.2 | 1.2 |
| Shader model | 6.4 | 6.6 |
| CUDA | — | 8.6 |
AUTRES CARACTÉRISTIQUES
Carte graphique
| Date de publication | — | Unknown |
|---|---|---|
| gpu.details.availability | — | 2022 |
| Génération | — | GeForce 30 |
| Prédécesseur | — | GeForce 20 |
| Production | — | Unreleased |
| Prix de lancement | — | 1,699 USD |
| Bus interface | — | PCIe 4.0 x16 |
| Critiques | — | 55 in our database |
