NVIDIA GeForce 8600M GS vs NVIDIA Quadro RTX 3000 Max-Q
Grafikprozessor
| GPU Name | G86 | TU106 |
|---|---|---|
| GPU Variante | G86-770-A2 | — |
| Architektur | Tesla | Turing |
| Hersteller | TSMC | TSMC |
| Prozessgröße | 80 nm | 12 nm |
| Transistoren | 210 million | 10,800 million |
| Chipgröße | 127 mm² | 445 mm² |
Mobile Grafik
| Veröffentlichungsdatum | May 1st, 2007 | May 27th, 2019 |
|---|---|---|
| Generation | GeForce 8M (8000M) | Quadro Mobile (Tx000) |
| Produktion | End-of-life | Active |
| Businterface | MXM-II | PCIe 3.0 x16 |
| Vorgänger | GeForce Go 7 | — |
| Nachfolger | GeForce 9M | — |
Takt-Geschwindigkeiten
| GPU-Taktung | 450 MHz | — |
|---|---|---|
| Shader-Taktung | 900 MHz | — |
| Speicher-Taktung | 400 MHz 800 Mbps effective | 1500 MHz 12 Gbps effective |
| Basis-Taktfrequenz | — | 600 MHz |
| Boost-Takt | — | 1215 MHz |
Speicher
| Speichergröße | 512 MB | 6 GB |
|---|---|---|
| Speichertyp | DDR2 | GDDR6 |
| Speicherbus | 128 bit | 192 bit |
| Bandbreite | 12.80 GB/s | 288.0 GB/s |
Render-Konfiguration
| Shading-Einheiten | 16 | 1920 |
|---|---|---|
| TMUs | 8 | 120 |
| ROPs | 4 | 64 |
| Anzahl SM | 1 | 30 |
| Cache L2 | 32 KB | 3 MB |
| Tensor Cores | — | 240 |
| RT-Kerne | — | 30 |
| Cache L1 | — | 64 KB (per SM) |
Theoretische Leistung
| Pixelrate | 1.800 GPixel/s | 77.76 GPixel/s |
|---|---|---|
| Texturrate | 3.600 GTexel/s | 145.8 GTexel/s |
| FP32 (float) Leistung | 28.80 GFLOPS | 4.666 TFLOPS |
| FP16 (half) Leistung | — | 9.331 TFLOPS (2:1) |
| FP64 (double) Leistung | — | 145.8 GFLOPS (1:32) |
Boarddesign
| Steckplatzbreite | MXM Module | IGP |
|---|---|---|
| TDP | 20 W | 60 W |
| Ausgabe | No outputs | No outputs |
| Stromverbindung | — | None |
Grafik-Features
| DirectX | 11.1 (10_0) | 12 Ultimate (12_2) |
|---|---|---|
| OpenGL | 3.3 | 4.6 |
| OpenCL | 1.1 | 3.0 |
| Vulkan | — | 1.2 |
| CUDA | 1.1 | 7.5 |
| Shader-Modell | 4.0 | 6.6 |
