Performance issues (micro-stuttering) and character lighting bug #34
Labels
No labels
accessibility
autorelease: pending
autorelease: tagged
bug
dependencies
documentation
duplicate
enhancement
github_actions
good first issue
help wanted
invalid
question
testing wanted
wontfix
No milestone
No assignees
1 participant
Due date
No due date set.
Dependencies
No dependencies set
Reference
MagicBeanDev/Hookshot#34
Loading…
Reference in a new issue
No description provided.
Delete branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Device & System Specifications
Issue Description
I wanted to share a report on how the game runs on my hardware setup. I have noticed a few graphical bugs, performance drops, and missing settings:
1. Graphical & Lighting Bug
2. Performance & Stuttering
3. Missing Settings / Constraints
Media & Attachments
(Please see the attached screenshots showing the lighting bug, current FPS counter, and the stuttering/frametime spikes)
CPU: AMD Ryzen 5 7600X
RAM: 32.0 GB
GPU: AMD RX 9070 XT
Same graphical issues, performance looks good and stutters seems to align to normal shader cache compilation.
Update on testing and performance:
• Language: Text languages can now be changed successfully.
• Performance & FPS: The game no longer suffers from freezing and maintains a steady 60 FPS most of the time.
• Stuttering: There is significantly less stuttering during the Deku Tree introduction scene.
• Lighting bug: The only remaining issue is the lighting bug. I suspect that the intense lighting in those areas might actually be what's causing the lingering stuttering.
Thanks for confirming performance and the language work. The dark characters were fixed in
lighting-fix. Closing; reopen with a log from the newest release if anything is still off.2026-11-10: porting NotRexed/Hookshot@9928d74 (BC images with colour attachment usage, raw BC views sized in blocks, OoT's noise volume pass 53509044401d78d2 no longer skipped) aborts the Metal driver on macOS: SIGABRT in AGX findOrCreateDriverProgramVariant (the attachment load program) from
MVKCommandEncoder::beginMetalRenderPass, 2 of 2 forest starts. MoltenVK also refuses colour attachment usage on 2D BC images (-11). Reverted onfix/upstream-mesh-fallback; the pass stays skipped on macOS, so the noise volumes stay zero and the foliage's LOD crossfade is a hard cut (Hookshot #1151). Next: render the noise into an uncompressed volume of its own and sample that.🤖 Generated with Claude Code