Nom du GPU | Blackcomb | GP107 |
---|---|---|
GPU variant | Blackcomb XT (216-0811000) | — |
Architecture | TeraScale 2 | Pascal |
Fonderie | TSMC | Samsung |
Taille du processus | 40 nm | 14 nm |
Transistors | 1,700 million | 3,300 million |
Taille des matrices | 212 mm² | 132 mm² |
Date de publication | Aug 25th, 2011 | Feb 23rd, 2019 |
---|---|---|
Génération | Vancouver (HD 6900M) | GeForce MX (1xx) |
Production | End-of-life | End-of-life |
Bus interface | MXM-B (3.0) | PCIe 3.0 x4 |
Prédécesseur | Manhattan | — |
Successeur | London | — |
GPU clock | 680 MHz | — |
---|---|---|
Mémoire clock | 900 MHz 3.6 Gbps effective | 1502 MHz 6 Gbps effective |
Base clock | — | 1469 MHz |
Boost clock | — | 1532 MHz |
Taille de la mémoire | 2 GB | 2 GB |
---|---|---|
Type de mémoire | GDDR5 | GDDR5 |
Bus mémoire | 256 bit | 64 bit |
Bande passante | 115.2 GB/s | 48.06 GB/s |
Shading units | 960 | 384 |
---|---|---|
TMUs | 48 | 24 |
ROPs | 32 | 16 |
Compute units | 12 | — |
Cache L1 | 8 KB (per CU) | 48 KB (per SM) |
Cache L2 | 512 KB | 512 KB |
Comptage SM | — | 3 |
Taux de pixel | 21.76 GPixel/s | 24.51 GPixel/s |
---|---|---|
Taux de texture | 32.64 GTexel/s | 36.77 GTexel/s |
FP32 (float) performance | 1,306 GFLOPS | 1,177 GFLOPS |
FP16 (half) performance | — | 18.38 GFLOPS (1:64) |
FP64 (double) performance | — | 36.77 GFLOPS (1:32) |
Taille Slot | MXM Module | IGP |
---|---|---|
TDP | 75 W | 25 W |
Sorties | No outputs | No outputs |
Connecteurs de puissance | None | None |
Board number | C296 | — |
DirectX | 11.2 (11_0) | 12 (12_1) |
---|---|---|
OpenGL | 4.4 | 4.6 |
OpenCL | 1.2 | 3.0 |
Vulkan | — | 1.2 |
Shader model | 5.0 | 6.4 |
CUDA | — | 6.1 |