GPU Name | GA100 | Kal-El |
---|---|---|
Architektur | Ampere | VLIW Vec4 |
Hersteller | TSMC | TSMC |
Prozessgröße | 7 nm | 40 nm |
Transistoren | 54,200 million | 10 million |
Chipgröße | 826 mm² | 80 mm² |
GPU Variante | — | T30-P-A3 |
Veröffentlichungsdatum | Jun 22nd, 2020 | — |
---|---|---|
Generation | Tesla | — |
Produktion | Active | — |
Businterface | PCIe 4.0 x16 | — |
GPU-Taktung | 1410 MHz | — |
---|---|---|
Speicher-Taktung | 1215 MHz 2.4 Gbps effective | System Shared |
Basis-Taktfrequenz | 765 MHz | 416 MHz |
Boost-Takt | 1410 MHz | 520 MHz |
Speichergröße | 40 GB | System Shared |
---|---|---|
Speichertyp | HBM2E | System Shared |
Speicherbus | 5120 bit | System Shared |
Bandbreite | 1,555 GB/s | System Dependent |
Shading-Einheiten | 6912 | — |
---|---|---|
TMUs | 432 | 8 |
ROPs | 160 | 8 |
Anzahl SM | 108 | — |
Tensor Cores | 432 | — |
Cache L1 | 192 KB (per SM) | — |
Cache L2 | 40 MB | — |
Pixelshader | — | 8 |
Vertex-Shader | — | 4 |
Pixelrate | 225.6 GPixel/s | 4.160 GPixel/s |
---|---|---|
Texturrate | 609.1 GTexel/s | 4.160 GTexel/s |
FP16 (half) Leistung | 77.97 TFLOPS (4:1) | — |
FP32 (float) Leistung | 19.49 TFLOPS | — |
FP64 (double) Leistung | 9.746 TFLOPS (1:2) | — |
Vertex-Rate | — | 416.0 MVertices/s |
Steckplatzbreite | Dual-slot | IGP |
---|---|---|
Länge | 267 mm 10.5 inches | — |
TDP | 250 W | 20 W |
Vorgeschlagenes Netzteil | 600 W | — |
Ausgabe | No outputs | No outputs |
Stromverbindung | 8-pin EPS | — |
Board-Nummer | P1001 SKU 200 | — |
DirectX | — | — |
---|---|---|
OpenGL | — | ES 2.0 |
OpenCL | 3.0 | — |
Vulkan | — | — |
CUDA | 8.0 | — |
Shader-Modell | — | — |
Pixelshader | — | — |
Vertex-Shader | — | — |
Veröffentlichungsdatum | — | Nov 9th, 2011 |
---|---|---|
Generation | — | Tegra |
Produktion | — | End-of-life |
Businterface | — | IGP |