NVIDIA Quadro RTX 4000 vs NVIDIA T1000
Grafikprozessor
| GPU Name | TU104 | TU117 |
|---|---|---|
| GPU Variante | TU104-850-A1 | — |
| Architektur | Turing | Turing |
| Hersteller | TSMC | TSMC |
| Prozessgröße | 12 nm | 12 nm |
| Transistoren | 13,600 million | 4,700 million |
| Chipgröße | 545 mm² | 200 mm² |
Grafikkarte
| Veröffentlichungsdatum | Nov 13th, 2018 | May 6th, 2021 |
|---|---|---|
| Generation | Quadro | Quadro |
| Produktion | Active | Active |
| Originalpreis | 899 USD | — |
| Businterface | PCIe 3.0 x16 | PCIe 3.0 x16 |
Takt-Geschwindigkeiten
| Basis-Taktfrequenz | 1005 MHz | 1065 MHz |
|---|---|---|
| Boost-Takt | 1545 MHz | 1395 MHz |
| Speicher-Taktung | 1625 MHz 13 Gbps effective | 1250 MHz 10 Gbps effective |
Speicher
| Speichergröße | 8 GB | 4 GB |
|---|---|---|
| Speichertyp | GDDR6 | GDDR6 |
| Speicherbus | 256 bit | 128 bit |
| Bandbreite | 416.0 GB/s | 160.0 GB/s |
Render-Konfiguration
| Shading-Einheiten | 2304 | 896 |
|---|---|---|
| TMUs | 144 | 56 |
| ROPs | 64 | 32 |
| Anzahl SM | 36 | 14 |
| Tensor Cores | 288 | — |
| RT-Kerne | 36 | — |
| Cache L1 | 64 KB (per SM) | 64 KB (per SM) |
| Cache L2 | 4 MB | 1024 KB |
Theoretische Leistung
| Pixelrate | 98.88 GPixel/s | 44.64 GPixel/s |
|---|---|---|
| Texturrate | 222.5 GTexel/s | 78.12 GTexel/s |
| FP16 (half) Leistung | 14.24 TFLOPS (2:1) | 5.000 TFLOPS (2:1) |
| FP32 (float) Leistung | 7.119 TFLOPS | 2.500 TFLOPS |
| FP64 (double) Leistung | 222.5 GFLOPS (1:32) | 78.12 GFLOPS (1:32) |
Boarddesign
| Steckplatzbreite | Single-slot | Single-slot |
|---|---|---|
| Länge | 241 mm 9.5 inches | — |
| Breite | 111 mm 4.4 inches | — |
| TDP | 160 W | 50 W |
| Vorgeschlagenes Netzteil | 450 W | 250 W |
| Ausgabe | 3x DisplayPort1x USB Type-C | 4x mini-DisplayPort |
| Stromverbindung | 1x 8-pin | None |
Grafik-Features
| DirectX | 12 Ultimate (12_2) | 12 (12_1) |
|---|---|---|
| OpenGL | 4.6 | 4.6 |
| OpenCL | 3.0 | 3.0 |
| Vulkan | 1.2 | 1.2 |
| CUDA | 7.5 | 7.5 |
| Shader-Modell | 6.6 | 6.6 |
