Nom du GPU | TU116 | GA106 |
---|---|---|
GPU variant | TU116-250-KA-A1 | GA106-140-A1 |
Architecture | Turing | Ampere |
Fonderie | TSMC | Samsung |
Taille du processus | 12 nm | 8 nm |
Transistors | 6,600 million | 13,250 million |
Taille des matrices | 284 mm² | 276 mm² |
Date de publication | Nov 22nd, 2019 | 2022 |
---|---|---|
Génération | GeForce 16 | GeForce 30 |
Prédécesseur | GeForce 10 | GeForce 20 |
Successeur | GeForce 20 | — |
Production | Active | Unreleased |
Bus interface | PCIe 3.0 x16 | PCIe 4.0 x16 |
Critiques | 27 in our database | — |
Prix de lancement | 159 USD | — |
gpu.details.availability | — | 2022 |
Base clock | 1530 MHz | 1545 MHz |
---|---|---|
Boost clock | 1725 MHz | 1740 MHz |
Mémoire clock | 1500 MHz 12 Gbps effective | 1750 MHz 14 Gbps effective |
Taille de la mémoire | 4 GB | 4 GB |
---|---|---|
Type de mémoire | GDDR6 | GDDR6 |
Bus mémoire | 128 bit | 128 bit |
Bande passante | 192.0 GB/s | 224.0 GB/s |
Shading units | 1280 | 2304 |
---|---|---|
TMUs | 80 | 72 |
ROPs | 32 | 40 |
Comptage SM | 20 | 18 |
Cache L1 | 64 KB (per SM) | 128 KB (per SM) |
Cache L2 | 1024 KB | 2 MB |
Tensor cores | — | 72 |
RT cores | — | 18 |
Taux de pixel | 55.20 GPixel/s | 69.60 GPixel/s |
---|---|---|
Taux de texture | 138.0 GTexel/s | 125.3 GTexel/s |
FP16 (half) performance | 8.832 TFLOPS (2:1) | 8.018 TFLOPS (1:1) |
FP32 (float) performance | 4.416 TFLOPS | 8.018 TFLOPS |
FP64 (double) performance | 138.0 GFLOPS (1:32) | 125.3 GFLOPS (1:64) |
Taille Slot | Dual-slot | Dual-slot |
---|---|---|
Longueur | 229 mm 9 inches | 242 mm 9.5 inches |
Taille | 111 mm 4.4 inches | 112 mm 4.4 inches |
Taille | 35 mm 1.4 inches | — |
TDP | 100 W | 90 W |
Suggestions pour le PSU | 300 W | 250 W |
Sorties | 1x DVI1x HDMI1x DisplayPort | 1x HDMI3x DisplayPort |
Connecteurs de puissance | 1x 6-pin | 1x 6-pin |
Board number | PG166 SKU 24 | — |
DirectX | 12 (12_1) | 12 Ultimate (12_2) |
---|---|---|
OpenGL | 4.6 | 4.6 |
OpenCL | 3.0 | 3.0 |
Vulkan | 1.2 | 1.2 |
CUDA | 7.5 | 8.6 |
Shader model | 6.6 | 6.6 |