NVIDIA CMP 50HX vs NVIDIA CMP 90HX
Grafikprozessor
| GPU Name | TU102 | GA102 |
|---|---|---|
| GPU Variante | TU102-100-A1 | GA102-100-A1 |
| Architektur | Turing | Ampere |
| Hersteller | TSMC | Samsung |
| Prozessgröße | 12 nm | 8 nm |
| Transistoren | 18,600 million | 28,300 million |
| Chipgröße | 754 mm² | 628 mm² |
Grafikkarte
| Veröffentlichungsdatum | Jun 24th, 2021 | Jul 28th, 2021 |
|---|---|---|
| Generation | Mining GPUs | Mining GPUs |
| Produktion | Active | Active |
| Businterface | PCIe 3.0 x16 | PCIe 4.0 x16 |
Takt-Geschwindigkeiten
| Basis-Taktfrequenz | 1350 MHz | 1500 MHz |
|---|---|---|
| Boost-Takt | 1545 MHz | 1710 MHz |
| Speicher-Taktung | 1750 MHz 14 Gbps effective | 1188 MHz 19 Gbps effective |
Speicher
| Speichergröße | 10 GB | 10 GB |
|---|---|---|
| Speichertyp | GDDR6 | GDDR6X |
| Speicherbus | 320 bit | 320 bit |
| Bandbreite | 560.0 GB/s | 760.3 GB/s |
Render-Konfiguration
| Shading-Einheiten | 3584 | 6400 |
|---|---|---|
| TMUs | 192 | 200 |
| ROPs | 80 | 80 |
| Anzahl SM | 56 | 50 |
| Tensor Cores | 448 | 200 |
| RT-Kerne | 56 | 50 |
| Cache L1 | 64 KB (per SM) | 128 KB (per SM) |
| Cache L2 | 5 MB | 5 MB |
Theoretische Leistung
| Pixelrate | 123.6 GPixel/s | 136.8 GPixel/s |
|---|---|---|
| Texturrate | 296.6 GTexel/s | 342.0 GTexel/s |
| FP16 (half) Leistung | 22.15 TFLOPS (2:1) | 21.89 TFLOPS (1:1) |
| FP32 (float) Leistung | 11.07 TFLOPS | 21.89 TFLOPS |
| FP64 (double) Leistung | 346.1 GFLOPS (1:32) | 342.0 GFLOPS (1:64) |
Boarddesign
| Steckplatzbreite | Dual-slot | Dual-slot |
|---|---|---|
| Länge | 267 mm 10.5 inches | 285 mm 11.2 inches |
| Breite | 116 mm 4.6 inches | 112 mm 4.4 inches |
| Höhe | 35 mm 1.4 inches | — |
| TDP | 250 W | 320 W |
| Vorgeschlagenes Netzteil | 600 W | 700 W |
| Ausgabe | No outputs | No outputs |
| Stromverbindung | 2x 8-pin | 2x 8-pin |
| Board-Nummer | PG150 SKU 100 | PG132 SKU 100 |
Grafik-Features
| DirectX | 12 Ultimate (12_2) | 12 Ultimate (12_2) |
|---|---|---|
| OpenGL | 4.6 | 4.6 |
| OpenCL | 3.0 | 3.0 |
| Vulkan | 1.2 | 1.2 |
| CUDA | 7.5 | 8.6 |
| Shader-Modell | 6.6 | 6.6 |
