This course aims to compare the performance of Nvidia RTX 3090 and RTX 3060 graphics cards in stable diffusion, machine learning, artificial intelligence, and video rendering tasks. The course covers topics such as OpenAI's Whisper, DaVinci Resolve, and FFmpeg benchmarks. Students will learn about GPU installation, driver updates, benchmarking methodologies, and optimizing hardware for specific tasks. The intended audience for this course includes individuals interested in understanding the capabilities of different graphics cards for machine learning and video rendering applications.
RTX 3090 vs RTX 3060 Ultimate Showdown for Stable Diffusion, ML, AI & Video Rendering Performance
Software Engineering Courses - SE Courses via YouTube
Overview
Syllabus
Box opening of Gainward #RTX3090 Ti and Cougar #GEX1050
Installation of RTX 3090 GPU and Cougar GEX1050 PSU into the computer case
Final view of the installed case
CPU, Ram and other hardware overview of the used PC
Used gist file explanation in this video
How to install latest Nvidia GeForce driver
What is the difference between Nvidia Game Ready drivers and Studio drivers?
OpenAI Whisper speech to text transcription benchmarks
How to verify installed and used PyTorch, CUDA and cuDNN versions via my custom script
How to update Whisper to latest version
Testing command used for Whisper
Demo of Whisper transcription benchmarks
How to install Torch version 2 on main Python installation
How to install cuNDD latest DLL files
Benchmark results of all Whisper tests
When RTX 3090 and #RTX3060 transcribing speech at the same time
4K Video rendering tests in Davinci Resolve
How to change rendering GPU in Davinci Resolve
Rendering results of Davinci Resolve benchmarks
Bug in Davinci Resolve, RTX 3060 is not used
Where to download FFmpeg with hardware acceleration - CUDA and GPU support
How to set default FFmpeg via environment variables path
Testing setup of the FFmpeg 8k video rendering
Demo demonstration of FFmpeg benchmark
Final results of FFmpeg benchmarks on both RTX 3060 and RTX 3090
Starting to benchmark Stable Diffusion via Automatic1111 Web UI
How to see used Torch, CUDA and cuDNN DLL version of your Web UI
How to update Web UI xFormers version
it/s iteration per second testing
Demo of testing methodologies that will be used for Stable Diffusion benchmarks
Starting result analysis of Stable Diffusion benchmarks Torch 1.13
Used DreamBooth training settings for benchmarking
Stable Diffusion benchmarks with Torch 2.0
How to make sure that Web UI uses second device in all cases
opt-sdp-attention benchmark results with Stable Diffusion
The discovery I made about optimizers used in Stable Diffusion Web UI
Solution for Stable Diffusion NansException : A Tensor with all NaNs in Unet
Taught by
Software Engineering Courses - SE Courses
