GPU Name | Polaris 23 | GP108 |
---|---|---|
GPU Variante | Polaris 23 MXT | GP108-650-A1 (N17S-G1-A1) |
Architektur | GCN 4.0 | Pascal |
Hersteller | GlobalFoundries | Samsung |
Prozessgröße | 14 nm | 14 nm |
Transistoren | 2,200 million | 1,800 million |
Chipgröße | 103 mm² | 74 mm² |
Veröffentlichungsdatum | Apr 11th, 2018 | May 17th, 2017 |
---|---|---|
Generation | Mobility Radeon (M500X) | GeForce MX (1xx) |
Produktion | Active | End-of-life |
Businterface | PCIe 3.0 x8 | PCIe 3.0 x4 |
Vorgänger | Crystal System | — |
Basis-Taktfrequenz | 1100 MHz | 1469 MHz |
---|---|---|
Boost-Takt | 1287 MHz | 1532 MHz |
Speicher-Taktung | 1500 MHz 6 Gbps effective | 1502 MHz 6 Gbps effective |
Speichergröße | 4 GB | 2 GB |
---|---|---|
Speichertyp | GDDR5 | GDDR5 |
Speicherbus | 64 bit | 64 bit |
Bandbreite | 48.00 GB/s | 48.06 GB/s |
Shading-Einheiten | 640 | 384 |
---|---|---|
TMUs | 40 | 24 |
ROPs | 16 | 16 |
Compute Units | 10 | — |
Cache L1 | 16 KB (per CU) | 48 KB (per SM) |
Cache L2 | 256 KB | 512 KB |
Anzahl SM | — | 3 |
Pixelrate | 20.59 GPixel/s | 24.51 GPixel/s |
---|---|---|
Texturrate | 51.48 GTexel/s | 36.77 GTexel/s |
FP16 (half) Leistung | 1.647 TFLOPS (1:1) | 18.38 GFLOPS (1:64) |
FP32 (float) Leistung | 1.647 TFLOPS | 1,177 GFLOPS |
FP64 (double) Leistung | 103.0 GFLOPS (1:16) | 36.77 GFLOPS (1:32) |
Steckplatzbreite | IGP | IGP |
---|---|---|
TDP | 50 W | 25 W |
Ausgabe | No outputs | No outputs |
Stromverbindung | None | None |
Board-Nummer | — | E2902 SKU 0 |
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 |