GPU Name | Vega 10 | GP107 |
---|---|---|
GPU Variante | Vega 10 XT (215-0894200) | N17P-G0-A1 |
Architektur | GCN 5.0 | Pascal |
Hersteller | GlobalFoundries | Samsung |
Prozessgröße | 14 nm | 14 nm |
Transistoren | 12,500 million | 3,300 million |
Chipgröße | 495 mm² | 132 mm² |
Veröffentlichungsdatum | Jun 27th, 2017 | Feb 1st, 2019 |
---|---|---|
Generation | Radeon Pro Mac (Vega Series) | GeForce 10 Mobile |
Produktion | Active | End-of-life |
Businterface | PCIe 3.0 x16 | PCIe 3.0 x16 |
Reviews | — | 15 in our database |
Vorgänger | — | GeForce 900M |
Nachfolger | — | GeForce 20 Mobile |
Basis-Taktfrequenz | 1250 MHz | 1366 MHz |
---|---|---|
Boost-Takt | 1350 MHz | 1442 MHz |
Speicher-Taktung | 786 MHz 1572 Mbps effective | 1752 MHz 7 Gbps effective |
Speichergröße | 16 GB | 3 GB |
---|---|---|
Speichertyp | HBM2 | GDDR5 |
Speicherbus | 2048 bit | 96 bit |
Bandbreite | 402.4 GB/s | 84.10 GB/s |
Shading-Einheiten | 4096 | 768 |
---|---|---|
TMUs | 256 | 48 |
ROPs | 64 | 24 |
Compute Units | 64 | — |
Cache L1 | 16 KB (per CU) | 48 KB (per SM) |
Cache L2 | 4 MB | 768 KB |
Anzahl SM | — | 6 |
Pixelrate | 86.40 GPixel/s | 34.61 GPixel/s |
---|---|---|
Texturrate | 345.6 GTexel/s | 69.22 GTexel/s |
FP16 (half) Leistung | 22.12 TFLOPS (2:1) | 34.61 GFLOPS (1:64) |
FP32 (float) Leistung | 11.06 TFLOPS | 2.215 TFLOPS |
FP64 (double) Leistung | 691.2 GFLOPS (1:16) | 69.22 GFLOPS (1:32) |
Steckplatzbreite | IGP | — |
---|---|---|
TDP | 250 W | 75 W |
Ausgabe | No outputs | No outputs |
Stromverbindung | None | None |
DirectX | 12 (12_1) | 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 |