GPU Name | Polaris 20 | GA103 |
---|---|---|
GPU Variante | Polaris 20 XL (215-0910052) | GA103 |
Architektur | GCN 4.0 | Ampere |
Hersteller | GlobalFoundries | Samsung |
Prozessgröße | 14 nm | 8 nm |
Transistoren | 5,700 million | unknown |
Chipgröße | 232 mm² | unknown |
Veröffentlichungsdatum | Unknown | — |
---|---|---|
Generation | Mining GPUs | — |
Produktion | Active | — |
Businterface | PCIe 3.0 x16 | — |
Reviews | 29 in our database | — |
Basis-Taktfrequenz | 1120 MHz | 585 MHz |
---|---|---|
Boost-Takt | 1206 MHz | 1125 MHz |
Speicher-Taktung | 2100 MHz 8.4 Gbps effective | 1500 MHz 12 Gbps effective |
Speichergröße | 8 GB | 16 GB |
---|---|---|
Speichertyp | GDDR5 | GDDR6 |
Speicherbus | 256 bit | 256 bit |
Bandbreite | 268.8 GB/s | 384.0 GB/s |
Shading-Einheiten | 2048 | 7680 |
---|---|---|
TMUs | 128 | 240 |
ROPs | 32 | 96 |
Compute Units | 32 | — |
Cache L1 | 16 KB (per CU) | 128 KB (per SM) |
Cache L2 | 2 MB | 4 MB |
Anzahl SM | — | 60 |
Tensor Cores | — | 240 |
RT-Kerne | — | 60 |
Pixelrate | 38.59 GPixel/s | 108.0 GPixel/s |
---|---|---|
Texturrate | 154.4 GTexel/s | 270.0 GTexel/s |
FP16 (half) Leistung | 4.940 TFLOPS (1:1) | 17.28 TFLOPS (1:1) |
FP32 (float) Leistung | 4.940 TFLOPS | 17.28 TFLOPS |
FP64 (double) Leistung | 308.7 GFLOPS (1:16) | 270.0 GFLOPS (1:64) |
Steckplatzbreite | Dual-slot | — |
---|---|---|
Länge | 241 mm 9.5 inches | — |
TDP | 180 W | 95 W |
Vorgeschlagenes Netzteil | 450 W | — |
Ausgabe | 1x HDMI | No outputs |
Stromverbindung | 2x 8-pin | None |
Board-Nummer | P010 | — |
DirectX | 12 (12_0) | 12 Ultimate (12_2) |
---|---|---|
OpenGL | 4.6 | 4.6 |
OpenCL | 2.1 | 3.0 |
Vulkan | 1.2 | 1.2 |
Shader-Modell | 6.4 | 6.5 |
CUDA | — | 8.6 |
Veröffentlichungsdatum | — | 2022 |
---|---|---|
gpu.details.availability | — | — |
Generation | — | GeForce 30 Mobile |
Vorgänger | — | GeForce 20 Mobile |
Produktion | — | Unreleased |
Businterface | — | PCIe 4.0 x16 |