GPU Name | Baffin | GM107 |
---|---|---|
GPU Variante | Baffin XT | N16P-GX-A1 |
Architektur | GCN 4.0 | Maxwell |
Hersteller | GlobalFoundries | TSMC |
Prozessgröße | 14 nm | 28 nm |
Transistoren | 3,000 million | 1,870 million |
Chipgröße | 123 mm² | 148 mm² |
Veröffentlichungsdatum | Oct 30th, 2016 | Mar 13th, 2015 |
---|---|---|
Generation | Radeon Pro Mac (400 Series) | GeForce 900A |
Produktion | Active | End-of-life |
Businterface | PCIe 3.0 x8 | MXM-B (3.0) |
Vorgänger | — | GeForce 800A |
Basis-Taktfrequenz | 850 MHz | 1029 MHz |
---|---|---|
Boost-Takt | 907 MHz | 1085 MHz |
Speicher-Taktung | 1270 MHz 5.1 Gbps effective | 1253 MHz 5 Gbps effective |
Speichergröße | 4 GB | 2 GB |
---|---|---|
Speichertyp | GDDR5 | GDDR5 |
Speicherbus | 128 bit | 128 bit |
Bandbreite | 81.28 GB/s | 80.19 GB/s |
Shading-Einheiten | 1024 | 640 |
---|---|---|
TMUs | 64 | 40 |
ROPs | 16 | 16 |
Compute Units | 16 | — |
Cache L1 | 16 KB (per CU) | 64 KB (per SMM) |
Cache L2 | 1024 KB | 2 MB |
Anzahl SMM | — | 5 |
Pixelrate | 14.51 GPixel/s | 17.36 GPixel/s |
---|---|---|
Texturrate | 58.05 GTexel/s | 43.40 GTexel/s |
FP16 (half) Leistung | 1.858 TFLOPS (1:1) | — |
FP32 (float) Leistung | 1.858 TFLOPS | 1,389 GFLOPS |
FP64 (double) Leistung | 116.1 GFLOPS (1:16) | 43.40 GFLOPS (1:32) |
Steckplatzbreite | IGP | MXM Module |
---|---|---|
TDP | 35 W | 75 W |
Ausgabe | No outputs | No outputs |
Stromverbindung | None | — |
DirectX | 12 (12_0) | 12 (11_0) |
---|---|---|
OpenGL | 4.6 | 4.6 |
OpenCL | 2.1 | 3.0 |
Vulkan | 1.2 | 1.1 |
Shader-Modell | 6.4 | 5.1 |
CUDA | — | 5.0 |