Nom du GPU | GF108 | GA106 |
---|---|---|
GPU variant | GF108-400-A1 | GA106-140-A1 |
Architecture | Fermi | Ampere |
Fonderie | TSMC | Samsung |
Taille du processus | 40 nm | 8 nm |
Transistors | 585 million | 13,250 million |
Taille des matrices | 116 mm² | 276 mm² |
Date de publication | Jun 18th, 2014 | 2022 |
---|---|---|
Génération | GeForce 700 | GeForce 30 |
Production | End-of-life | Unreleased |
Bus interface | PCIe 2.0 x16 | PCIe 4.0 x16 |
Prédécesseur | GeForce 600 | GeForce 20 |
Successeur | GeForce 900 | — |
gpu.details.availability | — | 2022 |
GPU clock | 700 MHz | — |
---|---|---|
Shader clock | 1400 MHz | — |
Mémoire clock | 800 MHz 1600 Mbps effective | 1750 MHz 14 Gbps effective |
Base clock | — | 1545 MHz |
Boost clock | — | 1740 MHz |
Taille de la mémoire | 2 GB | 4 GB |
---|---|---|
Type de mémoire | DDR3 | GDDR6 |
Bus mémoire | 128 bit | 128 bit |
Bande passante | 25.60 GB/s | 224.0 GB/s |
Shading units | 96 | 2304 |
---|---|---|
TMUs | 16 | 72 |
ROPs | 4 | 40 |
Comptage SM | 2 | 18 |
Cache L1 | 64 KB (per SM) | 128 KB (per SM) |
Cache L2 | 256 KB | 2 MB |
Tensor cores | — | 72 |
RT cores | — | 18 |
Taux de pixel | 2.800 GPixel/s | 69.60 GPixel/s |
---|---|---|
Taux de texture | 11.20 GTexel/s | 125.3 GTexel/s |
FP32 (float) performance | 268.8 GFLOPS | 8.018 TFLOPS |
FP64 (double) performance | 22.40 GFLOPS (1:12) | 125.3 GFLOPS (1:64) |
FP16 (half) performance | — | 8.018 TFLOPS (1:1) |
Taille Slot | Single-slot | Dual-slot |
---|---|---|
Longueur | 145 mm 5.7 inches | 242 mm 9.5 inches |
TDP | 49 W | 90 W |
Suggestions pour le PSU | 200 W | 250 W |
Sorties | 1x DVI1x HDMI1x VGA | 1x HDMI3x DisplayPort |
Connecteurs de puissance | None | 1x 6-pin |
Board number | P1070, P1071 | — |
Taille | — | 112 mm 4.4 inches |
DirectX | 12 (11_0) | 12 Ultimate (12_2) |
---|---|---|
OpenGL | 4.6 | 4.6 |
OpenCL | 1.1 | 3.0 |
Vulkan | — | 1.2 |
CUDA | 2.1 | 8.6 |
Shader model | 5.1 | 6.6 |