GPU Name | GF100 | TU117 |
---|---|---|
GPU Variante | GF100-825-A3 | — |
Architektur | Fermi | Turing |
Hersteller | TSMC | TSMC |
Prozessgröße | 40 nm | 12 nm |
Transistoren | 3,100 million | 4,700 million |
Chipgröße | 529 mm² | 200 mm² |
Veröffentlichungsdatum | Nov 2nd, 2010 | May 6th, 2021 |
---|---|---|
Generation | Quadro | Quadro |
Produktion | End-of-life | Active |
Originalpreis | 1,199 USD | — |
Businterface | PCIe 2.0 x16 | PCIe 3.0 x16 |
Reviews | 26 in our database | — |
GPU-Taktung | 475 MHz | — |
---|---|---|
Shader-Taktung | 950 MHz | — |
Speicher-Taktung | 702 MHz 2.8 Gbps effective | 1250 MHz 10 Gbps effective |
Basis-Taktfrequenz | — | 1065 MHz |
Boost-Takt | — | 1395 MHz |
Speichergröße | 2 GB | 8 GB |
---|---|---|
Speichertyp | GDDR5 | GDDR6 |
Speicherbus | 256 bit | 128 bit |
Bandbreite | 89.86 GB/s | 160.0 GB/s |
Shading-Einheiten | 256 | 896 |
---|---|---|
TMUs | 32 | 56 |
ROPs | 32 | 32 |
Anzahl SM | 8 | 14 |
Cache L1 | 64 KB (per SM) | 64 KB (per SM) |
Cache L2 | 512 KB | 1024 KB |
Pixelrate | 7.600 GPixel/s | 44.64 GPixel/s |
---|---|---|
Texturrate | 15.20 GTexel/s | 78.12 GTexel/s |
FP32 (float) Leistung | 486.4 GFLOPS | 2.500 TFLOPS |
FP64 (double) Leistung | 243.2 GFLOPS (1:2) | 78.12 GFLOPS (1:32) |
FP16 (half) Leistung | — | 5.000 TFLOPS (2:1) |
Steckplatzbreite | Single-slot | Single-slot |
---|---|---|
Länge | 241 mm 9.5 inches | — |
Breite | 111 mm 4.4 inches | — |
TDP | 142 W | 50 W |
Vorgeschlagenes Netzteil | 300 W | 250 W |
Ausgabe | 1x DVI2x DisplayPort | 4x mini-DisplayPort |
Stromverbindung | 1x 6-pin | None |
Board-Nummer | P1031 | — |
DirectX | 12 (11_0) | 12 (12_1) |
---|---|---|
OpenGL | 4.6 | 4.6 |
OpenCL | 1.1 | 3.0 |
Vulkan | — | 1.2 |
CUDA | 2.0 | 7.5 |
Shader-Modell | 5.1 | 6.6 |