NVIDIA GeForce GTX 860M vs NVIDIA GeForce RTX 3050 Ti Mobile
Grafikprozessor
| GPU Name | GK104 | GA106 |
|---|---|---|
| GPU Variante | N15P-GX-A1 | — |
| Architektur | Kepler | Ampere |
| Hersteller | TSMC | Samsung |
| Prozessgröße | 28 nm | 8 nm |
| Transistoren | 3,540 million | 13,250 million |
| Chipgröße | 294 mm² | 276 mm² |
Mobile Grafik
| Veröffentlichungsdatum | Mar 10th, 2014 | May 11th, 2021 |
|---|---|---|
| Generation | GeForce 800M | GeForce 30 Mobile |
| Produktion | End-of-life | Active |
| Businterface | PCIe 3.0 x16 | PCIe 4.0 x16 |
| Vorgänger | GeForce 700M | GeForce 20 Mobile |
| Nachfolger | GeForce 900M | — |
| gpu.details.availability | — | 2021 |
Takt-Geschwindigkeiten
| Basis-Taktfrequenz | 797 MHz | 735 MHz |
|---|---|---|
| Boost-Takt | 915 MHz | 1035 MHz |
| Speicher-Taktung | 1250 MHz 5 Gbps effective | 1500 MHz 12 Gbps effective |
Speicher
| Speichergröße | 2 GB | 4 GB |
|---|---|---|
| Speichertyp | GDDR5 | GDDR6 |
| Speicherbus | 128 bit | 128 bit |
| Bandbreite | 80.00 GB/s | 192.0 GB/s |
Render-Konfiguration
| Shading-Einheiten | 1152 | 2560 |
|---|---|---|
| TMUs | 96 | 80 |
| ROPs | 16 | 48 |
| Anzahl SMX | 6 | — |
| Cache L1 | 16 KB (per SMX) | 128 KB (per SM) |
| Cache L2 | 256 KB | 2 MB |
| Anzahl SM | — | 20 |
| Tensor Cores | — | 80 |
| RT-Kerne | — | 20 |
Theoretische Leistung
| Pixelrate | 21.96 GPixel/s | 49.68 GPixel/s |
|---|---|---|
| Texturrate | 87.84 GTexel/s | 82.80 GTexel/s |
| FP32 (float) Leistung | 2.108 TFLOPS | 5.299 TFLOPS |
| FP64 (double) Leistung | 87.84 GFLOPS (1:24) | 82.80 GFLOPS (1:64) |
| FP16 (half) Leistung | — | 5.299 TFLOPS (1:1) |
Boarddesign
| Steckplatzbreite | IGP | IGP |
|---|---|---|
| TDP | 75 W | 75 W |
| Ausgabe | No outputs | No outputs |
| Stromverbindung | None | None |
| Board-Nummer | P2053 SKU 4 | — |
Grafik-Features
| DirectX | 12 (11_0) | 12 Ultimate (12_2) |
|---|---|---|
| OpenGL | 4.6 | 4.6 |
| OpenCL | 3.0 | 3.0 |
| Vulkan | 1.1 | 1.2 |
| CUDA | 3.0 | 8.6 |
| Shader-Modell | 5.1 | 6.6 |
