GPU Name | Polaris 21 | TU102 |
---|---|---|
GPU Variante | Polaris 21M XT | TU102-875-A1 |
Architektur | GCN 4.0 | Turing |
Hersteller | GlobalFoundries | TSMC |
Prozessgröße | 14 nm | 12 nm |
Transistoren | 3,000 million | 18,600 million |
Chipgröße | 123 mm² | 754 mm² |
Veröffentlichungsdatum | Apr 11th, 2018 | — |
---|---|---|
Generation | Mobility Radeon (RX M500X) | — |
Produktion | Active | — |
Businterface | MXM-B (3.0) | — |
Vorgänger | Crystal System | — |
GPU-Taktung | 1275 MHz | — |
---|---|---|
Speicher-Taktung | 1450 MHz 5.8 Gbps effective | 1575 MHz 12.6 Gbps effective |
Basis-Taktfrequenz | — | 1065 MHz |
Boost-Takt | — | 1395 MHz |
Speichergröße | 4 GB | 16 GB |
---|---|---|
Speichertyp | GDDR5 | GDDR6 |
Speicherbus | 128 bit | 384 bit |
Bandbreite | 92.80 GB/s | 604.8 GB/s |
Shading-Einheiten | 1024 | 4608 |
---|---|---|
TMUs | 64 | 288 |
ROPs | 16 | 96 |
Compute Units | 16 | — |
Cache L1 | 16 KB (per CU) | 64 KB (per SM) |
Cache L2 | 1024 KB | 6 MB |
Anzahl SM | — | 72 |
Tensor Cores | — | 576 |
RT-Kerne | — | 72 |
Pixelrate | 20.40 GPixel/s | 133.9 GPixel/s |
---|---|---|
Texturrate | 81.60 GTexel/s | 401.8 GTexel/s |
FP16 (half) Leistung | 2.611 TFLOPS (1:1) | 25.71 TFLOPS (2:1) |
FP32 (float) Leistung | 2.611 TFLOPS | 12.86 TFLOPS |
FP64 (double) Leistung | 163.2 GFLOPS (1:16) | 401.8 GFLOPS (1:32) |
Steckplatzbreite | MXM Module | Dual-slot |
---|---|---|
TDP | 65 W | 260 W |
Ausgabe | No outputs | No outputs |
Stromverbindung | None | 1x 6-pin + 1x 8-pin |
Länge | — | 267 mm 10.5 inches |
Vorgeschlagenes Netzteil | — | 600 W |
Board-Nummer | — | PG150 SKU 215 |
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 | 6.6 |
CUDA | — | 7.5 |
Veröffentlichungsdatum | — | Unknown |
---|---|---|
Generation | — | GRID |
Produktion | — | Active |
Businterface | — | PCIe 3.0 x16 |