GPU Name | Gemini Lake GT1 | GA100 |
---|---|---|
Architektur | Generation 9.5 | Ampere |
Hersteller | Intel | TSMC |
Prozessgröße | 14 nm | 7 nm |
Transistoren | unknown | 54,200 million |
Chipgröße | unknown | 826 mm² |
Veröffentlichungsdatum | Dec 11th, 2017 | — |
---|---|---|
Generation | HD Graphics-T (Goldmont Plus) | — |
Produktion | Active | — |
Businterface | Ring Bus | — |
Reviews | 1 in our database | — |
Basis-Taktfrequenz | 200 MHz | 765 MHz |
---|---|---|
Boost-Takt | 650 MHz | 1410 MHz |
Speicher-Taktung | System Shared | 1215 MHz 2.4 Gbps effective |
GPU-Taktung | — | 1410 MHz |
Speichergröße | System Shared | 40 GB |
---|---|---|
Speichertyp | System Shared | HBM2E |
Speicherbus | System Shared | 5120 bit |
Bandbreite | System Dependent | 1,555 GB/s |
Shading-Einheiten | 96 | 6912 |
---|---|---|
TMUs | 12 | 432 |
ROPs | 2 | 160 |
Execution Units | 12 | — |
Anzahl SM | — | 108 |
Tensor Cores | — | 432 |
Cache L1 | — | 192 KB (per SM) |
Cache L2 | — | 40 MB |
Pixelrate | 1.300 GPixel/s | 225.6 GPixel/s |
---|---|---|
Texturrate | 7.800 GTexel/s | 609.1 GTexel/s |
FP16 (half) Leistung | 249.6 GFLOPS (2:1) | 77.97 TFLOPS (4:1) |
FP32 (float) Leistung | 124.8 GFLOPS | 19.49 TFLOPS |
FP64 (double) Leistung | 31.20 GFLOPS (1:4) | 9.746 TFLOPS (1:2) |
Steckplatzbreite | IGP | Dual-slot |
---|---|---|
TDP | 5 W | 250 W |
Ausgabe | No outputs | No outputs |
Länge | — | 267 mm 10.5 inches |
Vorgeschlagenes Netzteil | — | 600 W |
Stromverbindung | — | 8-pin EPS |
Board-Nummer | — | P1001 SKU 200 |
DirectX | 12 (12_1) | — |
---|---|---|
OpenGL | 4.6 | — |
OpenCL | 3.0 | 3.0 |
Vulkan | 1.2 | — |
Shader-Modell | 6.4 | — |
CUDA | — | 8.0 |
Veröffentlichungsdatum | — | Jun 22nd, 2020 |
---|---|---|
Generation | — | Tesla |
Produktion | — | Active |
Businterface | — | PCIe 4.0 x16 |