GPU Name | TU117 | GA104 |
---|---|---|
GPU Variante | TU117-300-A1 | GA104-300-A1 |
Architektur | Turing | Ampere |
Hersteller | TSMC | Samsung |
Prozessgröße | 12 nm | 8 nm |
Transistoren | 4,700 million | 17,400 million |
Chipgröße | 200 mm² | 392 mm² |
Veröffentlichungsdatum | Apr 23rd, 2019 | Sep 1st, 2020 |
---|---|---|
Generation | GeForce 16 | GeForce 30 |
Vorgänger | GeForce 10 | GeForce 20 |
Nachfolger | GeForce 20 | — |
Produktion | Active | Active |
Originalpreis | 149 USD | 499 USD |
Businterface | PCIe 3.0 x16 | PCIe 4.0 x16 |
Reviews | 45 in our database | 78 in our database |
gpu.details.availability | — | Oct 29th, 2020 |
Basis-Taktfrequenz | 1485 MHz | 1500 MHz |
---|---|---|
Boost-Takt | 1665 MHz | 1725 MHz |
Speicher-Taktung | 2001 MHz 8 Gbps effective | 1750 MHz 14 Gbps effective |
Speichergröße | 4 GB | 8 GB |
---|---|---|
Speichertyp | GDDR5 | GDDR6 |
Speicherbus | 128 bit | 256 bit |
Bandbreite | 128.1 GB/s | 448.0 GB/s |
Shading-Einheiten | 896 | 5888 |
---|---|---|
TMUs | 56 | 184 |
ROPs | 32 | 96 |
Anzahl SM | 14 | 46 |
Cache L1 | 64 KB (per SM) | 128 KB (per SM) |
Cache L2 | 1024 KB | 4 MB |
Tensor Cores | — | 184 |
RT-Kerne | — | 46 |
Pixelrate | 53.28 GPixel/s | 165.6 GPixel/s |
---|---|---|
Texturrate | 93.24 GTexel/s | 317.4 GTexel/s |
FP16 (half) Leistung | 5.967 TFLOPS (2:1) | 20.31 TFLOPS (1:1) |
FP32 (float) Leistung | 2.984 TFLOPS | 20.31 TFLOPS |
FP64 (double) Leistung | 93.24 GFLOPS (1:32) | 317.4 GFLOPS (1:64) |
Steckplatzbreite | Dual-slot | Dual-slot |
---|---|---|
Länge | 229 mm 9 inches | 242 mm 9.5 inches |
Breite | 111 mm 4.4 inches | 112 mm 4.4 inches |
Höhe | 35 mm 1.4 inches | — |
TDP | 75 W | 220 W |
Vorgeschlagenes Netzteil | 250 W | 550 W |
Ausgabe | 1x DVI1x HDMI1x DisplayPort | 1x HDMI3x DisplayPort |
Stromverbindung | None | 1x 12-pin |
Board-Nummer | PG174 SKU 0 | PG142 SKU 10 |
DirectX | 12 (12_1) | 12 Ultimate (12_2) |
---|---|---|
OpenGL | 4.6 | 4.6 |
OpenCL | 3.0 | 3.0 |
Vulkan | 1.2 | 1.2 |
CUDA | 7.5 | 8.6 |
Shader-Modell | 6.6 | 6.6 |