AMD Radeon HD 6870 1600SP Edition vs NVIDIA Tesla P100 PCIe 12 GB
Grafikprozessor
| GPU Name | Cypress | GP100 |
|---|---|---|
| GPU Variante | Cypress XT (215-0735033) | — |
| Architektur | TeraScale 2 | Pascal |
| Hersteller | TSMC | TSMC |
| Prozessgröße | 40 nm | 16 nm |
| Transistoren | 2,154 million | 15,300 million |
| Chipgröße | 334 mm² | 610 mm² |
Grafikkarte
| Veröffentlichungsdatum | Jan 14th, 2013 | Jun 20th, 2016 |
|---|---|---|
| Generation | Northern Islands | Tesla |
| Produktion | End-of-life | End-of-life |
| Businterface | PCIe 2.0 x16 | PCIe 3.0 x16 |
| Reviews | 170 in our database | — |
| Vorgänger | Evergreen | — |
| Nachfolger | Southern Islands | — |
| Originalpreis | — | 4,599 USD |
Takt-Geschwindigkeiten
| GPU-Taktung | 850 MHz | — |
|---|---|---|
| Speicher-Taktung | 1050 MHz 4.2 Gbps effective | 715 MHz 1430 Mbps effective |
| Basis-Taktfrequenz | — | 1190 MHz |
| Boost-Takt | — | 1329 MHz |
Speicher
| Speichergröße | 1024 MB | 12 GB |
|---|---|---|
| Speichertyp | GDDR5 | HBM2 |
| Speicherbus | 256 bit | 3072 bit |
| Bandbreite | 134.4 GB/s | 549.1 GB/s |
Render-Konfiguration
| Shading-Einheiten | 1600 | 3584 |
|---|---|---|
| TMUs | 80 | 224 |
| ROPs | 32 | 96 |
| Compute Units | 20 | — |
| Cache L1 | 8 KB (per CU) | 24 KB (per SM) |
| Cache L2 | 512 KB | 3 MB |
| Anzahl SM | — | 56 |
Theoretische Leistung
| Pixelrate | 27.20 GPixel/s | 127.6 GPixel/s |
|---|---|---|
| Texturrate | 68.00 GTexel/s | 297.7 GTexel/s |
| FP32 (float) Leistung | 2.720 TFLOPS | 9.526 TFLOPS |
| FP64 (double) Leistung | 544.0 GFLOPS (1:5) | 4.763 TFLOPS (1:2) |
| FP16 (half) Leistung | — | 19.05 TFLOPS (2:1) |
Boarddesign
| Steckplatzbreite | Dual-slot | Dual-slot |
|---|---|---|
| Länge | 282 mm 11.1 inches | 267 mm 10.5 inches |
| TDP | 175 W | 250 W |
| Vorgeschlagenes Netzteil | 450 W | 600 W |
| Ausgabe | 2x DVI1x HDMI2x mini-DisplayPort | No outputs |
| Stromverbindung | 2x 6-pin | 1x 8-pin |
| Board-Nummer | C008, E140 | — |
Grafik-Features
| DirectX | 11.2 (11_0) | 12 (12_1) |
|---|---|---|
| OpenGL | 4.4 | 4.6 |
| OpenCL | 1.2 | 3.0 |
| Vulkan | — | 1.2 |
| Shader-Modell | 5.0 | 6.4 |
| CUDA | — | 6.0 |
