GPU Name | TU106 | GP104 |
---|---|---|
GPU Variante | N18E-G1-KD-A1 | — |
Architektur | Turing | Pascal |
Hersteller | TSMC | TSMC |
Prozessgröße | 12 nm | 16 nm |
Transistoren | 10,800 million | 7,200 million |
Chipgröße | 445 mm² | 314 mm² |
Veröffentlichungsdatum | Jan 29th, 2019 | Feb 21st, 2018 |
---|---|---|
Generation | GeForce 20 Mobile | Quadro Mobile (Px200) |
Produktion | Active | End-of-life |
Businterface | PCIe 3.0 x16 | MXM-B (3.0) |
Vorgänger | GeForce 10 Mobile | — |
Nachfolger | GeForce 30 Mobile | — |
Basis-Taktfrequenz | 960 MHz | 1557 MHz |
---|---|---|
Boost-Takt | 1200 MHz | 1747 MHz |
Speicher-Taktung | 1750 MHz 14 Gbps effective | 1804 MHz 7.2 Gbps effective |
Speichergröße | 6 GB | 16 GB |
---|---|---|
Speichertyp | GDDR6 | GDDR5 |
Speicherbus | 192 bit | 256 bit |
Bandbreite | 336.0 GB/s | 230.9 GB/s |
Shading-Einheiten | 1920 | 2560 |
---|---|---|
TMUs | 120 | 160 |
ROPs | 48 | 64 |
Anzahl SM | 30 | 20 |
Tensor Cores | 240 | — |
RT-Kerne | 30 | — |
Cache L1 | 64 KB (per SM) | 48 KB (per SM) |
Cache L2 | 3 MB | 2 MB |
Pixelrate | 57.60 GPixel/s | 111.8 GPixel/s |
---|---|---|
Texturrate | 144.0 GTexel/s | 279.5 GTexel/s |
FP16 (half) Leistung | 9.216 TFLOPS (2:1) | 139.8 GFLOPS (1:64) |
FP32 (float) Leistung | 4.608 TFLOPS | 8.945 TFLOPS |
FP64 (double) Leistung | 144.0 GFLOPS (1:32) | 279.5 GFLOPS (1:32) |
Steckplatzbreite | MXM Module | MXM Module |
---|---|---|
TDP | 115 W | 100 W |
Ausgabe | No outputs | No outputs |
Stromverbindung | None | None |
Board-Nummer | E4914 SKU 30 | — |
DirectX | 12 Ultimate (12_2) | 12 (12_1) |
---|---|---|
OpenGL | 4.6 | 4.6 |
OpenCL | 3.0 | 3.0 |
Vulkan | 1.2 | 1.2 |
CUDA | 7.5 | 6.1 |
Shader-Modell | 6.6 | 6.4 |