GPU Name | Polaris 21 | GP107 |
---|---|---|
GPU Variante | Polaris 21M XT | — |
Architektur | GCN 4.0 | Pascal |
Hersteller | GlobalFoundries | Samsung |
Prozessgröße | 14 nm | 14 nm |
Transistoren | 3,000 million | 3,300 million |
Chipgröße | 123 mm² | 132 mm² |
Veröffentlichungsdatum | Apr 11th, 2018 | Feb 7th, 2017 |
---|---|---|
Generation | Mobility Radeon (RX M500X) | Quadro Mobile (Px000) |
Produktion | Active | End-of-life |
Businterface | MXM-B (3.0) | PCIe 3.0 x16 |
Vorgänger | Crystal System | — |
GPU-Taktung | 1275 MHz | — |
---|---|---|
Speicher-Taktung | 1450 MHz 5.8 Gbps effective | 1253 MHz 5 Gbps effective |
Basis-Taktfrequenz | — | 1430 MHz |
Boost-Takt | — | 1557 MHz |
Speichergröße | 4 GB | 4 GB |
---|---|---|
Speichertyp | GDDR5 | GDDR5 |
Speicherbus | 128 bit | 128 bit |
Bandbreite | 92.80 GB/s | 80.19 GB/s |
Shading-Einheiten | 1024 | 384 |
---|---|---|
TMUs | 64 | 24 |
ROPs | 16 | 16 |
Compute Units | 16 | — |
Cache L1 | 16 KB (per CU) | 48 KB (per SM) |
Cache L2 | 1024 KB | 1024 KB |
Anzahl SM | — | 3 |
Pixelrate | 20.40 GPixel/s | 24.91 GPixel/s |
---|---|---|
Texturrate | 81.60 GTexel/s | 37.37 GTexel/s |
FP16 (half) Leistung | 2.611 TFLOPS (1:1) | 18.68 GFLOPS (1:64) |
FP32 (float) Leistung | 2.611 TFLOPS | 1,196 GFLOPS |
FP64 (double) Leistung | 163.2 GFLOPS (1:16) | 37.37 GFLOPS (1:32) |
Steckplatzbreite | MXM Module | — |
---|---|---|
TDP | 65 W | 25 W |
Ausgabe | No outputs | No outputs |
Stromverbindung | None | None |
DirectX | 12 (12_0) | 12 (12_1) |
---|---|---|
OpenGL | 4.6 | 4.6 |
OpenCL | 2.1 | 3.0 |
Vulkan | 1.2 | 1.2 |
Shader-Modell | 6.4 | 6.4 |
CUDA | — | 6.1 |