GPU Name | TU116 | GA107S |
---|---|---|
Architektur | Turing | Ampere |
Hersteller | TSMC | Samsung |
Prozessgröße | 12 nm | 8 nm |
Transistoren | 6,600 million | unknown |
Chipgröße | 284 mm² | unknown |
Veröffentlichungsdatum | Apr 23rd, 2019 | 2022 |
---|---|---|
Generation | GeForce 16 Mobile | GeForce 20 Mobile |
Produktion | Active | Unreleased |
Businterface | PCIe 3.0 x16 | PCIe 3.0 x16 |
Vorgänger | GeForce 10 Mobile | GeForce 10 Mobile |
Nachfolger | GeForce 20 Mobile | GeForce 30 Mobile |
Basis-Taktfrequenz | 1140 MHz | 1155 MHz |
---|---|---|
Boost-Takt | 1335 MHz | 1477 MHz |
Speicher-Taktung | 1500 MHz 12 Gbps effective | 1750 MHz 14 Gbps effective |
Speichergröße | 6 GB | 4 GB |
---|---|---|
Speichertyp | GDDR6 | GDDR6 |
Speicherbus | 192 bit | 64 bit |
Bandbreite | 288.0 GB/s | 112.0 GB/s |
Shading-Einheiten | 1536 | 2048 |
---|---|---|
TMUs | 96 | 128 |
ROPs | 48 | 48 |
Anzahl SM | 24 | 32 |
Cache L1 | 64 KB (per SM) | 64 KB (per SM) |
Cache L2 | 1536 KB | 2 MB |
Tensor Cores | — | 256 |
RT-Kerne | — | 32 |
Pixelrate | 64.08 GPixel/s | 70.90 GPixel/s |
---|---|---|
Texturrate | 128.2 GTexel/s | 189.1 GTexel/s |
FP16 (half) Leistung | 8.202 TFLOPS (2:1) | 12.10 TFLOPS (2:1) |
FP32 (float) Leistung | 4.101 TFLOPS | 6.050 TFLOPS |
FP64 (double) Leistung | 128.2 GFLOPS (1:32) | 189.1 GFLOPS (1:32) |
Steckplatzbreite | MXM Module | Dual-slot |
---|---|---|
TDP | unknown | 45 W |
Ausgabe | No outputs | 1x DVI1x HDMI2x DisplayPort |
Stromverbindung | None | 1x 6-pin |
Board-Nummer | E4914 SKU 50 | — |
Länge | — | 229 mm 9 inches |
Breite | — | 113 mm 4.4 inches |
Höhe | — | 35 mm 1.4 inches |
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 |