ATI All-In-Wonder 9000 PRO vs NVIDIA P102-100
Grafikprozessor
| GPU Name | RV250 | GP102 |
|---|---|---|
| Architektur | Rage 7 | Pascal |
| Hersteller | TSMC | TSMC |
| Prozessgröße | 150 nm | 16 nm |
| Transistoren | 36 million | 11,800 million |
| Chipgröße | 97 mm² | 471 mm² |
| GPU Variante | — | GP102-100-A1 |
Grafikkarte
| Veröffentlichungsdatum | Mar 31st, 2003 | Feb 12th, 2018 |
|---|---|---|
| Generation | All-In-Wonder | Mining GPUs |
| Produktion | End-of-life | End-of-life |
| Businterface | AGP 8x | PCIe 3.0 x4 |
Takt-Geschwindigkeiten
| GPU-Taktung | 275 MHz | — |
|---|---|---|
| Speicher-Taktung | 270 MHz 540 Mbps effective | 1376 MHz 11 Gbps effective |
| Basis-Taktfrequenz | — | 1582 MHz |
| Boost-Takt | — | 1683 MHz |
Speicher
| Speichergröße | 64 MB | 5 GB |
|---|---|---|
| Speichertyp | DDR | GDDR5X |
| Speicherbus | 128 bit | 320 bit |
| Bandbreite | 8.640 GB/s | 440.3 GB/s |
Render-Konfiguration
| Pixelshader | 4 | — |
|---|---|---|
| Vertex-Shader | 1 | — |
| TMUs | 4 | 200 |
| ROPs | 4 | 80 |
| Shading-Einheiten | — | 3200 |
| Anzahl SM | — | 25 |
| Cache L1 | — | 48 KB (per SM) |
| Cache L2 | — | 2.5 MB |
Theoretische Leistung
| Pixelrate | 1.100 GPixel/s | 134.6 GPixel/s |
|---|---|---|
| Vertex-Rate | 68.75 MVertices/s | — |
| Texturrate | 1.100 GTexel/s | 336.6 GTexel/s |
| FP16 (half) Leistung | — | 168.3 GFLOPS (1:64) |
| FP32 (float) Leistung | — | 10.77 TFLOPS |
| FP64 (double) Leistung | — | 336.6 GFLOPS (1:32) |
Boarddesign
| Steckplatzbreite | Single-slot | Dual-slot |
|---|---|---|
| TDP | unknown | 250 W |
| Vorgeschlagenes Netzteil | 200 W | 600 W |
| Ausgabe | 1x DVI2x S-Video | No outputs |
| Stromverbindung | None | 2x 8-pin |
| Board-Nummer | 959-00 | PG611 SKU 100 |
| Länge | — | 267 mm 10.5 inches |
Grafik-Features
| DirectX | 8.1 | 12 (12_1) |
|---|---|---|
| OpenGL | 1.4 | 4.6 |
| OpenCL | — | 3.0 |
| Vulkan | — | 1.2 |
| Pixelshader | 1.4 | — |
| Vertex-Shader | 1.1 | — |
| CUDA | — | 6.1 |
| Shader-Modell | — | 6.4 |
