Little side note. I've been very impressed with Github Copilot. I checked out a few AIs asking them the same general questions about bit reversal routines. (non paid versions) Github Copilot basically completed my tests and was a joy to bounce ideas back and forth with. After that I paid for a...
Realize. This is different in the way it's outputting the data to an array. I didn't see the code before. Sorry about that. (note: Not my coding style just told it what to do)
// 3dpm3.cpp : Defines the entry point for the application.
//
#define _USE_MATH_DEFINES
#include <stdio.h>
#include...
I did some vibe coding. (github copilot) and for the most part on my 9950x3d is memory bound. Using a std::mersenne_twister_engine for random values.
[Standard] Generated 134217728 random normalized vectors in 177.939 ms.
Vector 0: (0.453678, 0.710586, 0.537815, 1)
Vector 1: (0.121839...
Good time to watch the Gamers Nexus video with Amit Mehra and Wendell.
Amit tell some good stories on what he has and what he currently designs in the chips. There is one part where he says something to the effect. A lot of pieces can run faster but you basically end up running at the speed...
Ok since this is back in my head I will expand on why my algorithm was designed as such.
The original algorithm is flawed. The whole point of it is to produce random vectors to simulate a particle moving randomly. In post #4 a simple explanation of it is given.
This is a mapping problem that...
The irony. The place where you quoted all these graphs just did a followup basically agreeing with Steve. Steve will not have to do an investigation.
https://www.techspot.com/review/3008-radeon-9070-xt-amd-finewine/
Edit: In a strange turn of fate, the techspot review is done by a Steven. Well...
I have a monitor that has a bad display port. For the longest time, I swapped out different cables for varying results, sometimes random artifacts, videos freezing, couple crashes. Didn't think about trying the other port. Guess what. No more issues.
Happened with both nVidia and AMD cards...
If I knew then what I know now I could of expressed my thoughts better.
I still think that a half angle table and a quasi quaternion slurp would avoid a lot of expensive trigonometry calculations.
Although GPU operations are often referred to as SIMD they are more classified as SIMT (single instruction multiple thread). While the implementation is left up to the compute unit, there is no dispatch unit that can pre-group data into a SMID avx type format. Basically there is no front end to...
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.