WaitForPresent

Making Code go brrrrrrrrr at Keen Games. Previously Shapefarm, Ubisoft.

@jon_valdes This is so frustrating! We had something similar like this recently at work as well which led me to implement a renderdoc style texture debugger (in my hobby project at least for now). Because obviously the issue *never* reproduced in renderdoc 🙃

WaitForPresent boosted:
2025-06-12

meshoptimizer 0.24 is out! Featuring new meshlet builder for raytracing workloads, improvements to cluster partitioning, new functions for reindexing, simplification, clustering and coverage computation, many internal improvements and, of course, VS2013 support!

Thanks to Valve for sponsoring most of the work on the core library in this release!

Boost / GitHub stars appreciated as always :)

github.com/zeux/meshoptimizer/

WaitForPresent boosted:
Josh Simmonsdotstdy
2025-06-10

@aras @zeux allocations >:)

WaitForPresent boosted:
2025-06-07

@dotstdy @jon_valdes @RYStorm yeah settlers was on snowdrop but i doubt there is another one coming anytime soon :D. Anno still has its own thing running, same as BGE2 afaik

@dotstdy @jon_valdes @RYStorm is dunia officially done now?

WaitForPresent boosted:
2025-05-27

Some more great lighting screenshots

WaitForPresent boosted:

“Rendering Assassin’s Creed Shadows” slides are available on gdcvault.com/play/1035526/Rend

WaitForPresent boosted:

@xoofx My 'binary response' of trashing 'AI' isn't due to limited technical thinking, but instead taking into account wider issues.

Huge financial and resources investments are being pushed into the development of a tool which has not yet shown itself to be economically beneficial. At the same time companies are using this tool as leverage against skilled employees. Companies are going back on their renewable plans in order to grow the required datacentres. None of this is good.

WaitForPresent boosted:

Giving a talk about game industry insights next week together with Jan Klose (game Hessen) at Technische Universität Darmstadt. Feel free to drop by if you want to listen to me promoting in-house engines for half an hour.
May 13th, 09:50 - 11:30 in S101/A03.

@MouseByTheSea if you are on windows, i can wholeheartedly recommend 10x as a code editor I use it both for work and private projects. Sadly no linux port yet though :(

@MouseByTheSea yeah its really unfortunate because otherwise zed is a really good editor. :/

@mjp I wouldn't recommend doing this to anyone as it is still super inefficient, but we were weeks from launch and this saved us close to 2ms in the shadow pass 😅

@mjp The bounds of the shadow casters are usually very long and thin and pretty bad for a traditional occlusion culling test. So I just cleared the shadow maps to the FAR value and used the main-views downsampled depth buffer (I think it was mip 5 or 6) to generate frustum slabs that I rendered into the shadowmap with depth overwrite set to NEAR. So if you had good occluders in the main view you'd also mark a good portion of the shadowmaps for much cheaper early z culling.

@mjp Great blog post! Early-Z is really great if you can leverage it. One kindof unorthodox way I have (ab)used this was for shadow "occlusion culling". Usually for occlusion culling you'd use the previous frame depth buffer and project your draws into the previous frames space and check for visibility. But for shadows you can additionally use the main view for visibility information (i.e. if the object and its shadow is not on screen it does not need to be rendered to the shadow map).

@breakin @dotstdy That is actually a good point, it is a very similar problem!

@lritter @dotstdy I was talking about rendering resources from the very beginning (mostly as an example but that is my concrete use case also )

@lritter @dotstdy True but for GPU Allocations you don't have direct access to virtual memory

@dotstdy @lritter It works fine yes, yet I can almost always manually write a better allocation scheme - and thus use less total memory - because I know all the allocations upfront. It's sad to waste that precious memory :(

@lritter not at all 🙃

Client Info

Server: https://mastodon.social
Version: 2025.04
Repository: https://github.com/cyevgeniy/lmst