The “albedo bit” is related to glossy reflections. Here’s an example: 1st shot is the cache without jitter (exposing the tiling), 2nd shot is with jitter, and you can see how the details disappear… Finally, 3rd shot divides the radiance by the average cell albedo and re-multiply with per-ray albedo.

That’s a great article, thanks for sharing! I’ll try to answer all your questions.
1. Yes, that’s definitely a hard one, balancing between stability and responsiveness is really difficult. I’ve ended up with tracking temporally smoothed delta of luminance values (you can tweak all values of course):

New blog post! Behind the scenes of some of the techniques involved in making our last PC demo 💫 gboisse.github.io/posts/this-i...

Performance is quite decent even at the highest quality settings; 120us for specular (and ~20us for diffuse):

Been wanting to implement that fast GGX-based convolution for glossy reflections probes for quite a while and finally made some time for it:) research.activision.com/publications...

I did this, combined with some efficient light culling, to get some real-time dynamic GI on PS4 some nearly 10 years ago now:)

Quick behind the scenes of our latest PC demo production, here’s the node graph for the part from 02:14 to 03:00: youtu.be/W7Om9rf0qKc?... Might print out more graphs from other scenes if there’s some interest:)

Perks of the job:) #rdna4

2nd place 🎉 thanks @revision-party.bsky.social!

Looking forward to tonight’s compos! @revision-party.bsky.social #demoscene

Now drawn as a mesh 🎉

Testing the fluid sim. with object collision:)

Re-wrote my acceleration structure for neighbourhood search based on some insight from @vassvik.bsky.social and got significant speedups on large fluid simulations:) Here’s ~1/2 million particles ✨