GPU Name | Navi 21 | GA104 |
---|---|---|
Architektur | RDNA 2.0 | Ampere |
Hersteller | TSMC | Samsung |
Prozessgröße | 7 nm | 8 nm |
Transistoren | 26,800 million | 17,400 million |
Chipgröße | 520 mm² | 392 mm² |
GPU Variante | — | GA104-200-A1 |
Veröffentlichungsdatum | Jun 8th, 2021 | Dec 1st, 2020 |
---|---|---|
Generation | Radeon Pro | GeForce 30 |
Produktion | Active | Active |
Originalpreis | 2,249 USD | 399 USD |
Businterface | PCIe 4.0 x16 | PCIe 4.0 x16 |
gpu.details.availability | — | Dec 2nd, 2020 |
Vorgänger | — | GeForce 20 |
Reviews | — | 64 in our database |
Basis-Taktfrequenz | 2075 MHz | 1410 MHz |
---|---|---|
Boost-Takt | 2320 MHz | 1665 MHz |
Speicher-Taktung | 2000 MHz 16 Gbps effective | 1750 MHz 14 Gbps effective |
Speichergröße | 32 GB | 8 GB |
---|---|---|
Speichertyp | GDDR6 | GDDR6 |
Speicherbus | 256 bit | 256 bit |
Bandbreite | 512.0 GB/s | 448.0 GB/s |
Shading-Einheiten | 3840 | 4864 |
---|---|---|
TMUs | 240 | 152 |
ROPs | 96 | 80 |
Compute Units | 60 | — |
RT-Kerne | 60 | 38 |
gpu.details.l0-cache | 32 KB per WGP | — |
Cache L1 | 128 KB per Array | 128 KB (per SM) |
Cache L2 | 4 MB | 4 MB |
gpu.details.l3-cache | 128 MB | — |
Anzahl SM | — | 38 |
Tensor Cores | — | 152 |
Pixelrate | 222.7 GPixel/s | 133.2 GPixel/s |
---|---|---|
Texturrate | 556.8 GTexel/s | 253.1 GTexel/s |
FP16 (half) Leistung | 35.64 TFLOPS (2:1) | 16.20 TFLOPS (1:1) |
FP32 (float) Leistung | 17.82 TFLOPS | 16.20 TFLOPS |
FP64 (double) Leistung | 1,114 GFLOPS (1:16) | 253.1 GFLOPS (1:64) |
Steckplatzbreite | Dual-slot | Dual-slot |
---|---|---|
Länge | 267 mm 10.5 inches | 242 mm 9.5 inches |
Breite | 120 mm 4.7 inches | 112 mm 4.4 inches |
Höhe | 50 mm 2 inches | — |
TDP | 250 W | 200 W |
Vorgeschlagenes Netzteil | 600 W | 550 W |
Ausgabe | 6x mini-DisplayPort | 1x HDMI3x DisplayPort |
Stromverbindung | 1x 6-pin + 1x 8-pin | 1x 12-pin |
Teile-Nummer | 102-D16487 | — |
Board-Nummer | D164 | PG142 SKU 20 |
DirectX | 12 Ultimate (12_2) | 12 Ultimate (12_2) |
---|---|---|
OpenGL | 4.6 | 4.6 |
OpenCL | 2.1 | 3.0 |
Vulkan | 1.2 | 1.2 |
Shader-Modell | 6.5 | 6.6 |
CUDA | — | 8.6 |