Nom du GPU | DG1 | GM20B |
---|---|---|
Architecture | Generation 12.1 | Maxwell 2.0 |
Fonderie | Intel | TSMC |
Taille du processus | 10 nm | 20 nm |
Transistors | unknown | 2,000 million |
Taille des matrices | 95 mm² | 118 mm² |
GPU variant | — | ODNX02-A2 |
Date de publication | Oct 31st, 2020 | — |
---|---|---|
Génération | Iris Xe Graphics-M (Tiger Lake) | — |
Production | Active | — |
Bus interface | PCIe 4.0 x4 | — |
Critiques | 1 in our database | — |
Base clock | 300 MHz | 384 MHz |
---|---|---|
Boost clock | 1650 MHz | 768 MHz |
Mémoire clock | 2133 MHz 4.3 Gbps effective | 1600 MHz 3.2 Gbps effective |
Taille de la mémoire | 4 GB | 4 GB |
---|---|---|
Type de mémoire | LPDDR4X | DDR4 |
Bus mémoire | 128 bit | 64 bit |
Bande passante | 68.26 GB/s | 25.60 GB/s |
Shading units | 768 | 256 |
---|---|---|
TMUs | 48 | 16 |
ROPs | 24 | 16 |
Unités d'exécution | 96 | — |
Cache L2 | 1024 KB | — |
gpu.details.l3-cache | 16 MB | — |
Comptage SMM | — | 2 |
Taux de pixel | 39.60 GPixel/s | 12.29 GPixel/s |
---|---|---|
Taux de texture | 79.20 GTexel/s | 12.29 GTexel/s |
FP16 (half) performance | 5.069 TFLOPS (2:1) | 786.4 GFLOPS (2:1) |
FP32 (float) performance | 2.534 TFLOPS | 393.2 GFLOPS |
FP64 (double) performance | 633.6 GFLOPS (1:4) | 12.29 GFLOPS (1:32) |
Taille Slot | IGP | IGP |
---|---|---|
TDP | 25 W | 15 W |
Sorties | No outputs | No outputs |
Longueur | — | 239 mm 9.4 inches |
Taille | — | 101 mm 4 inches |
Taille | — | 28 mm 1.1 inches |
Poids | — | 0.4 kg (0.88 lbs) |
Suggestions pour le PSU | — | 350 W |
Connecteurs de puissance | — | None |
Board number | — | HAC-001 |
DirectX | 12 (12_1) | 12 (12_1) |
---|---|---|
OpenGL | 4.6 | 4.6 |
OpenCL | 3.0 | 1.2 |
Vulkan | 1.2 | 1.1 |
Shader model | 6.4 | 6.4 |
CUDA | — | 5.3 |
Date de publication | — | Mar 17th, 2017 |
---|---|---|
Génération | — | Console GPU |
Production | — | End-of-life |
Bus interface | — | IGP |
Prix de lancement | — | 299 USD |