Nom du GPU | Amethyst | G80 |
---|---|---|
GPU variant | E8950 | G80-100-K0-A2 |
Architecture | GCN 3.0 | Tesla |
Fonderie | TSMC | TSMC |
Taille du processus | 28 nm | 90 nm |
Transistors | 5,000 million | 681 million |
Taille des matrices | 366 mm² | 484 mm² |
Date de publication | Sep 29th, 2015 | — |
---|---|---|
Génération | Embedded (8000) | — |
Production | End-of-life | — |
Bus interface | MXM-B (3.0) | — |
Critiques | 3 in our database | — |
Base clock | 735 MHz | — |
---|---|---|
Boost clock | 1000 MHz | — |
Mémoire clock | 1500 MHz 6 Gbps effective | 792 MHz 1584 Mbps effective |
GPU clock | — | 513 MHz |
Shader clock | — | 1188 MHz |
Taille de la mémoire | 8 GB | 320 MB |
---|---|---|
Type de mémoire | GDDR5 | GDDR3 |
Bus mémoire | 256 bit | 320 bit |
Bande passante | 192.0 GB/s | 63.36 GB/s |
Shading units | 2048 | 96 |
---|---|---|
TMUs | 128 | 24 |
ROPs | 32 | 20 |
Compute units | 32 | — |
Cache L1 | 16 KB (per CU) | — |
Cache L2 | 512 KB | 80 KB |
Comptage SM | — | 12 |
Taux de pixel | 32.00 GPixel/s | 10.26 GPixel/s |
---|---|---|
Taux de texture | 128.0 GTexel/s | 24.62 GTexel/s |
FP16 (half) performance | 4.096 TFLOPS (1:1) | — |
FP32 (float) performance | 4.096 TFLOPS | 228.1 GFLOPS |
FP64 (double) performance | 256.0 GFLOPS (1:16) | — |
Taille Slot | MXM Module | Dual-slot |
---|---|---|
TDP | 95 W | 143 W |
Sorties | No outputs | 2x DVI1x S-Video |
Connecteurs de puissance | None | 1x 6-pin |
Board number | C769 | P356 |
Longueur | — | 267 mm 10.5 inches |
Suggestions pour le PSU | — | 300 W |
DirectX | 12 (12_0) | 11.1 (10_0) |
---|---|---|
OpenGL | 4.6 | 3.3 |
OpenCL | 2.0 | 1.1 (1.0) |
Vulkan | 1.2 | — |
Shader model | 6.3 | 4.0 |
CUDA | — | 1.0 |
Date de publication | — | Feb 12th, 2007 |
---|---|---|
Génération | — | GeForce 8 |
Production | — | End-of-life |
Prix de lancement | — | 269 USD |
Bus interface | — | PCIe 1.0 x16 |
Critiques | — | 403 in our database |
Prédécesseur | — | GeForce 7 PCIe |
Successeur | — | GeForce 9 |