GPU Name | TU102 | GP102 |
---|---|---|
GPU Variante | TU102-875-A1 | GP102-450-A1 |
Architektur | Turing | Pascal |
Hersteller | TSMC | TSMC |
Prozessgröße | 12 nm | 16 nm |
Transistoren | 18,600 million | 11,800 million |
Chipgröße | 754 mm² | 471 mm² |
Veröffentlichungsdatum | Aug 13th, 2018 | Apr 6th, 2017 |
---|---|---|
Generation | Quadro | GeForce 10 |
Produktion | Active | Active |
Originalpreis | 9,999 USD | 1,199 USD |
Businterface | PCIe 3.0 x16 | PCIe 3.0 x16 |
Vorgänger | — | GeForce 900 |
Nachfolger | — | GeForce 20 |
Reviews | — | 9 in our database |
Basis-Taktfrequenz | 1395 MHz | 1405 MHz |
---|---|---|
Boost-Takt | 1770 MHz | 1582 MHz |
Speicher-Taktung | 1750 MHz 14 Gbps effective | 1426 MHz 11408 MHz effective |
Speichergröße | 48 GB | 12 GB |
---|---|---|
Speichertyp | GDDR6 | GDDR5X |
Speicherbus | 384 bit | 384 bit |
Bandbreite | 672.0 GB/s | 547.6 GB/s |
Shading-Einheiten | 4608 | 3840 |
---|---|---|
TMUs | 288 | 240 |
ROPs | 96 | 96 |
Anzahl SM | 72 | 30 |
Tensor Cores | 576 | — |
RT-Kerne | 72 | — |
Cache L1 | 64 KB (per SM) | 48 KB (per SM) |
Cache L2 | 6 MB | 3 MB |
Pixelrate | 169.9 GPixel/s | 151.9 GPixel/s |
---|---|---|
Texturrate | 509.8 GTexel/s | 379.7 GTexel/s |
FP16 (half) Leistung | 32.62 TFLOPS (2:1) | 189.8 GFLOPS (1:64) |
FP32 (float) Leistung | 16.31 TFLOPS | 12.15 TFLOPS |
FP64 (double) Leistung | 509.8 GFLOPS (1:32) | 379.7 GFLOPS (1:32) |
Steckplatzbreite | Dual-slot | Dual-slot |
---|---|---|
Länge | 267 mm 10.5 inches | 267 mm 10.5 inches |
Breite | 111 mm 4.4 inches | 112 mm 4.4 inches |
TDP | 260 W | 250 W |
Vorgeschlagenes Netzteil | 600 W | 600 W |
Ausgabe | 4x DisplayPort1x USB Type-C | 1x HDMI3x DisplayPort |
Stromverbindung | 1x 6-pin + 1x 8-pin | 1x 6-pin + 1x 8-pin |
Höhe | — | 40 mm 1.6 inches |
Board-Nummer | — | PG611 SKU 30 |
DirectX | 12 Ultimate (12_2) | 12 (12_1) |
---|---|---|
OpenGL | 4.6 | 4.6 |
OpenCL | 3.0 | 1.2 |
Vulkan | 1.2 | 1.2.131 |
CUDA | 7.5 | 6.1 |
Shader-Modell | 6.6 | 6.4 |