GPU Name | TU102 | GT200B |
---|---|---|
GPU Variante | TU102-875-A1 | G200-835-B2 |
Architektur | Turing | Tesla 2.0 |
Hersteller | TSMC | TSMC |
Prozessgröße | 12 nm | 55 nm |
Transistoren | 18,600 million | 1,400 million |
Chipgröße | 754 mm² | 470 mm² |
Veröffentlichungsdatum | Unknown | Mar 30th, 2009 |
---|---|---|
Generation | GRID | Quadro FX |
Produktion | Active | End-of-life |
Businterface | PCIe 3.0 x16 | PCIe 2.0 x16 |
Originalpreis | — | 799 USD |
Reviews | — | 26 in our database |
Basis-Taktfrequenz | 1065 MHz | — |
---|---|---|
Boost-Takt | 1395 MHz | — |
Speicher-Taktung | 1750 MHz 14 Gbps effective | 800 MHz 1600 Mbps effective |
GPU-Taktung | — | 600 MHz |
Shader-Taktung | — | 1204 MHz |
Speichergröße | 4 GB | 1024 MB |
---|---|---|
Speichertyp | GDDR6 | GDDR3 |
Speicherbus | 384 bit | 256 bit |
Bandbreite | 672.0 GB/s | 51.20 GB/s |
Shading-Einheiten | 4608 | 192 |
---|---|---|
TMUs | 288 | 64 |
ROPs | 96 | 16 |
Anzahl SM | 72 | 24 |
Tensor Cores | 576 | — |
RT-Kerne | 72 | — |
Cache L1 | 64 KB (per SM) | — |
Cache L2 | 6 MB | 128 KB |
Pixelrate | 133.9 GPixel/s | 9.600 GPixel/s |
---|---|---|
Texturrate | 401.8 GTexel/s | 38.40 GTexel/s |
FP16 (half) Leistung | 25.71 TFLOPS (2:1) | — |
FP32 (float) Leistung | 12.86 TFLOPS | 462.3 GFLOPS |
FP64 (double) Leistung | 401.8 GFLOPS (1:32) | 57.79 GFLOPS (1:8) |
Steckplatzbreite | Dual-slot | Single-slot |
---|---|---|
Länge | 267 mm 10.5 inches | 198 mm 7.8 inches |
TDP | 260 W | 108 W |
Vorgeschlagenes Netzteil | 600 W | 300 W |
Ausgabe | No outputs | 1x DVI2x DisplayPort |
Stromverbindung | 1x 6-pin + 1x 8-pin | 1x 6-pin |
Board-Nummer | PG150 SKU 215 | P755 |
Breite | — | 111 mm 4.4 inches |
DirectX | 12 Ultimate (12_2) | 11.1 (10_0) |
---|---|---|
OpenGL | 4.6 | 3.3 |
OpenCL | 3.0 | 1.1 |
Vulkan | 1.2 | — |
CUDA | 7.5 | 1.3 |
Shader-Modell | 6.6 | 4.0 |