<aside> 💡
These are my regular updates on my progress on Pixel Storm for February 2026! Updates from the previous month are here: January 2026 Pixel Storm Updates
</aside>
February 11, 2026 9:10 PM (PST) Well, a couple of weeks ago I sped my way to the end of Game Physics Engine Development (2nd Ed) so I could get the full picture, be done with the book and code the rest of the physics. But, since then I had to spend a week refactoring my code to remove the shims I had in place to make a particle physics engine work with what I had, and then go and build a contact manifold without the help of the book and the accompanying code base, since it left that as an exercise to the reader to figure out. It was a lot to figure out. Now that I’m finally ready to work on the rigid body collision resolution I realize I forgot everything I had read, and have to re-read the last 150 pages of the book! Well, back to reading I guess. I should have it down though after I re-read the thing.
February 11, 2026 12:12 AM (PST)
Ready for rigid body contact resolution. Physics are working great with the contact manifold.
February 10, 2026 9:41 PM (PST)
I have completed a persistent contact manifold with feature-based caching, recalculation from feature IDs, manifold culling, bounding sphere cleanup, and penetration propagation. I can now finally add angular resolution. Well let me run this thing and see if it even runs.
February 9, 2026 11:16 PM (PST) Ok I implemented recalculating contact information for cached contacts from features. It was really hard. I rewarded myself by playing my game for a bit. I think knocking the ball into some goal will be a way to score points or something. I don’t know.
February 9, 2026 8:48 PM (PST)
Working on contact coherency. It is mostly left unimplemented in the book and it turns out to be really complex.
February 8, 2026 5:16 PM (PST)
Ok the SAT collisions have been implemented. Now I need to generate multiple contacts and begin the work of rigid body collision resolution, including friction. Then, I should finally be done with physics?
February 8, 2026 12:23 AM (PST) I have reviewed the Game Physics Engine Development (2nd Ed) ‘s Cyclone engine code for the SAT boxAndBox function and I feel like I can implement it tomorrow and it will actually be much better than what I currently have both in terms of performance and accuracy. It will probably take me all day, but then I can work on generating multiple contacts, and then, finally I can do the rigid body contact resolution that will let me get ramps I can drive up holy moly.
February 7, 2026 9:20 PM (PST) Sooooo my existing collision test is not good enough for rigid body and I need to implement the SAT (separating axis test) from the book so I can efficiently generate multiple contacts in order to get good rotation data. Cool cool. It’s going to be at least another week until I’m done with this physics stuff. I am so over it, but I need to continue and get it over with. I can’t add any of the features I want until I have rigid body physics.