Benchmarking GPUs for Mixed Precision Training with Deep Learning
NVIDIAが語るVoltaとTuring、最新GPUはこう使おう：GTC Japan 2018（3/4 ページ） - EE Times Japan
AMD FidelityFX Super Resolution FP32 fallback tested, native FP16 is 7% faster - VideoCardz.com
Hopperアーキテクチャ｜NTTPCのGPU+｜NVIDIA Eliteパートナー
Testing AMD Radeon VII Double-Precision Scientific And Financial Performance – Techgage
TensorFlow Performance with 1-4 GPUs -- RTX Titan, 2080Ti, 2080, 2070, GTX 1660Ti, 1070, 1080Ti, and Titan V | Puget Systems
TensorFlowでディープラーニング性能をGPU別にベンチマーク比較 | パソコン工房 NEXMAG
FP64, FP32, FP16, BFLOAT16, TF32, and other members of the ZOO | by Grigory Sapunov | Medium
NVIDIA RTX 2080 Ti Benchmarks for Deep Learning with TensorFlow: Updated with XLA & FP16 | Exxact Blog
Using Tensor Cores for Mixed-Precision Scientific Computing | NVIDIA Technical Blog
AMD FidelityFX Super Resolution FP32 fallback tested, native FP16 is 7% faster - VideoCardz.com
FP16 Throughput on GP104: Good for Compatibility (and Not Much Else) - The NVIDIA GeForce GTX 1080 & GTX 1070 Founders Editions Review: Kicking Off the FinFET Generation
Accelerating TensorFlow on NVIDIA A100 GPUs - Edge AI and Vision Alliance
Harnessing GPU Tensor Cores for Fast FP16 Arithmetic to Speed up Mixed-Precision Iterative Refinement Solvers
株式会社HPCテック | NVIDIA A100
TITAN RTX Benchmarks for Deep Learning in TensorFlow 2019: XLA, FP16, FP32, & NVLink | Exxact Blog
More In-Depth Details of Floating Point Precision - NVIDIA CUDA - PyTorch Dev Discussions
TensowflowでFP16をいい感じに使って自動で高速化してくれるAutomatic Mixed Precision | cocoinit23
Choose FP16, FP32 or int8 for Deep Learning Models
Arm NN for GPU inference FP16 and FastMath - AI and ML blog - Arm Community blogs - Arm Community
Nvidia Unveils Pascal Tesla P100 With Over 20 TFLOPS Of FP16 Performance - Powered By GP100 GPU With 15 Billion Transistors & 16GB Of HBM2
Hardware for Deep Learning. Part 3: GPU | by Grigory Sapunov | Intento
Mixed-Precision Programming with CUDA 8 | NVIDIA Technical Blog
Dell Precision T7920 Dual Intel Xeon Workstation Review - Page 5 of 9 - ServeTheHome
FP16 Throughput on GP104: Good for Compatibility (and Not Much Else) - The NVIDIA GeForce GTX 1080 & GTX 1070 Founders Editions Review: Kicking Off the FinFET Generation
NVIDIA Tesla T4 ResNet 50 Training FP16 - ServeTheHome
