GPU Name | Wimbledon | Baffin |
---|---|---|
GPU Variante | Wimbledon XT (216-0836036) | Baffin XT |
Architektur | GCN 1.0 | GCN 4.0 |
Hersteller | TSMC | GlobalFoundries |
Prozessgröße | 28 nm | 14 nm |
Transistoren | 2,800 million | 3,000 million |
Chipgröße | 212 mm² | 123 mm² |
Veröffentlichungsdatum | Apr 24th, 2012 | Jan 5th, 2017 |
---|---|---|
Generation | London (HD 7900M) | Mobility Radeon (RX M500) |
Produktion | End-of-life | Active |
Businterface | MXM-B (3.0) | MXM-B (3.0) |
Vorgänger | Vancouver | Crystal System |
Nachfolger | Solar System | — |
GPU-Taktung | 850 MHz | — |
---|---|---|
Speicher-Taktung | 1200 MHz 4.8 Gbps effective | 1500 MHz 6 Gbps effective |
Basis-Taktfrequenz | — | 1090 MHz |
Boost-Takt | — | 1202 MHz |
Speichergröße | 2 GB | 4 GB |
---|---|---|
Speichertyp | GDDR5 | GDDR5 |
Speicherbus | 256 bit | 128 bit |
Bandbreite | 153.6 GB/s | 96.00 GB/s |
Shading-Einheiten | 1280 | 1024 |
---|---|---|
TMUs | 80 | 64 |
ROPs | 32 | 16 |
Compute Units | 20 | 16 |
Cache L1 | 16 KB (per CU) | 16 KB (per CU) |
Cache L2 | 512 KB | 1024 KB |
Pixelrate | 27.20 GPixel/s | 19.23 GPixel/s |
---|---|---|
Texturrate | 68.00 GTexel/s | 76.93 GTexel/s |
FP32 (float) Leistung | 2.176 TFLOPS | 2.462 TFLOPS |
FP64 (double) Leistung | 136.0 GFLOPS (1:16) | 153.9 GFLOPS (1:16) |
FP16 (half) Leistung | — | 2.462 TFLOPS (1:1) |
Steckplatzbreite | MXM Module | MXM Module |
---|---|---|
TDP | 100 W | 65 W |
Ausgabe | No outputs | No outputs |
Stromverbindung | None | None |
Board-Nummer | C429 | — |
DirectX | 12 (11_1) | 12 (12_0) |
---|---|---|
OpenGL | 4.6 | 4.6 |
OpenCL | 1.2 | 2.1 |
Vulkan | 1.2 | 1.2 |
Shader-Modell | 5.1 | 6.4 |