https://github.com/pikachuxxxx/tlou2_optimizations_deepdive
My notes on the YT video Low-level Optimization for TLOU2 by parikshit to explain myself better
https://github.com/pikachuxxxx/tlou2_optimizations_deepdive
Last synced: 2 months ago
JSON representation
My notes on the YT video Low-level Optimization for TLOU2 by parikshit to explain myself better
- Host: GitHub
- URL: https://github.com/pikachuxxxx/tlou2_optimizations_deepdive
- Owner: Pikachuxxxx
- Created: 2024-12-20T10:57:26.000Z (10 months ago)
- Default Branch: master
- Last Pushed: 2025-04-14T17:50:42.000Z (6 months ago)
- Last Synced: 2025-07-29T12:44:23.237Z (2 months ago)
- Language: TeX
- Size: 387 KB
- Stars: 1
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# TLOU2_Optimizations_DeepDive
https://www.youtube.com/watch?v=CvS6rX-67wA
# What is this?
I came across this [YT video a SIGGAPH presentation on TLOU2 optimizations by parikshit](https://www.youtube.com/watch?v=CvS6rX-67wA), I couldn't get my head 😶🌫️ around any of it the first time I listened throught, so I decided to write some deep notes on it. This repo is an attempt to eplain the topics he mentioned in the video in a much deeper sense, I've tried to make assumptions and intuitively explain why they did what they did. **There might be some mistakes so please open an issues to correct them.** or reach out to me at phani.s2909@gmail.com to help me out.
In the first draft I just make conceptual explanations of the points they mentioned in the video. y next goal is to provide all my conclusions with small snippets of code and GPU captures to understand GPU architexture better so expect a V2 sometime later next year.
I've also tried to add as many defitnions to glossary as possible and I will tryo add some notes on GPU architecture especiall y how WGPs, TDS/GDS are stored and how caches work behind the scenes.
[PDF Notes](https://github.com/Pikachuxxxx/TLOU2_Optimizations_DeepDive/blob/master/TLOU2OptimizationsYTVideoNotes.pdf)
## V1
The V1 just contains my notes in .tex and .pdf format.
- I'll also try to add more references to any kind of glossary used and why I made certain intutions
## V2
- Provide example snippets to emulate the scnarions and profile them using NSigh/AMD tools.
- Use these experiments to validate the theory and intuition made
## Contributing
If you feel something doesn't sound right please open an issues addressing your conerns. I'll be happy to correct and learn from you. Thanksss! ✌🏻