Ray Tracing [C++ & SDL2] Faster Linear Algebra (Episode 25)

Ray Tracing [C++ & SDL2] Faster Linear Algebra (Episode 25)

In the last of my three-part look at improving the performance of our ray tracer, I focus on further improvements to the linear algebra library involving implementing new version of the matrix class that don't use memory from the heap. I also go over various other changes I have made to the code to make better use of the new vector and matrix classes. This results in a huge improvement in performance, taking the rendering time for our test scene down to just 3.6 seconds on my system. Considering that when we started look at performance the rendering time was 86 seconds for the same scene, I think this is a significant improvement!

As with all of my videos, I show the code that I have developed to solve the particular problems that I need to solve, with the hope that it may provide sufficient background and inspiration to allow you to go on to develop your own code if you wish to. My code may not be the most appropriate solution to the problem you are trying to solve, so I urge you to consider the problem carefully and decide for yourself on the most appropriate solution. I make every effort to ensure that my code works as it should and is free of bugs, but of course I cannot provide any guarantees. If you use my code as shown, I strongly encourage you to make sure that you test it thoroughly to ensure that it works as you need it to. If you find a bug, do please let me know in the comments!

The Github repository for this series can be found here:
https://github.com/QuantitativeBytes/...

The Github repository for my linear algebra library can be found here:
https://github.com/QuantitativeBytes/...

#raytracing #cplusplus #rendering #pathtracing #qbray