~15 FPS returning to places i've been before (forest, castle village) #2615

Closed
opened 2026-11-12 02:23:44 +00:00 by jcreveling · 4 comments
jcreveling commented 2026-11-12 02:23:44 +00:00 (Migrated from github.com)

Hookshot version

2.1.0

What kind of problem

Graphics (missing, black or wrong-coloured things, lighting, flicker)

What happened

Since updating to 2.1.0, I've gone back to Kokiri forest and Hyrule castle village, and as soon as i go in there my FPS drops from 60 to 15 or less. I've been in these areas before, and had zero performance issues, only began after i returned to them after spending time in the village w/ the cemetery etc.. can i downgrade to test if it's the update, or if it's because i've returned?

I've saved and quit, and upon relaunching and returning it behaves much more normal (tho i did drop to ~50-55 FPS at a couple points in Kokiri forest)

Where in the game

kokiri forest, hyrule village

How often

Once so far

Graphics card and driver version

RTX 5080 / current driver

CPU, RAM and operating system

9800X3D, 32GB DDR5, Win 11

Changes from the defaults

No response

Other software hooked into the game

No, nothing hooked in

hookshot_log.txt

hookshot_log.1.txt

hookshot_log.3.txt
hookshot_log.4.txt
hookshot_log.txt

Screenshots or video

No response

Before you send it

  • I'm on the newest release, or the problem is still there on it.
  • I searched the open issues for this problem.
  • I'm not attaching game files, keys, hookshot.exe, the .bin caches or saves.
### Hookshot version 2.1.0 ### What kind of problem Graphics (missing, black or wrong-coloured things, lighting, flicker) ### What happened Since updating to 2.1.0, I've gone back to Kokiri forest and Hyrule castle village, and as soon as i go in there my FPS drops from 60 to 15 or less. I've been in these areas before, and had zero performance issues, only began after i returned to them after spending time in the village w/ the cemetery etc.. can i downgrade to test if it's the update, or if it's because i've returned? I've saved and quit, and upon relaunching and returning it behaves much more normal (tho i did drop to ~50-55 FPS at a couple points in Kokiri forest) ### Where in the game kokiri forest, hyrule village ### How often Once so far ### Graphics card and driver version RTX 5080 / current driver ### CPU, RAM and operating system 9800X3D, 32GB DDR5, Win 11 ### Changes from the defaults _No response_ ### Other software hooked into the game No, nothing hooked in ### hookshot_log.txt [hookshot_log.1.txt](https://github.com/user-attachments/files/34723550/hookshot_log.1.txt) <!-- Failed to upload "hookshot_log.2.txt" --> [hookshot_log.3.txt](https://github.com/user-attachments/files/34723549/hookshot_log.3.txt) [hookshot_log.4.txt](https://github.com/user-attachments/files/34723551/hookshot_log.4.txt) [hookshot_log.txt](https://github.com/user-attachments/files/34723552/hookshot_log.txt) ### Screenshots or video _No response_ ### Before you send it - [x] I'm on the newest release, or the problem is still there on it. - [x] I searched the open issues for this problem. - [x] I'm not attaching game files, keys, hookshot.exe, the .bin caches or saves.
jcreveling commented 2026-11-12 02:39:46 +00:00 (Migrated from github.com)

returning to Kakariko Village same thing... as soon as i go in and walk around, FPS drops to single digits.

returning to Kakariko Village same thing... as soon as i go in and walk around, FPS drops to single digits.
NotRexed commented 2026-11-12 03:08:39 +00:00 (Migrated from github.com)

Thanks for the report. Reports of the same kind are collected in #993, which lists this one with your details and links back here, so one fix closes them together. Follow #993 for the fix and add new details there (or here). If this turns out to be separate it will be reopened. Try the new Texture detail row (Medium) once #2557 is in, and the Quality preset Low meanwhile.

Thanks for the report. Reports of the same kind are collected in #993, which lists this one with your details and links back here, so one fix closes them together. Follow #993 for the fix and add new details there (or here). If this turns out to be separate it will be reopened. Try the new Texture detail row (Medium) once #2557 is in, and the Quality preset Low meanwhile.
jcreveling commented 2026-11-12 03:21:17 +00:00 (Migrated from github.com)

is it possible to revert to the previous build? I just did a rebuild from 2.1.0 -> 2.1.0-rc.3-193-g55b976c0 to test that but if i could just revert to the previous build?

is it possible to revert to the previous build? I just did a rebuild from 2.1.0 -> 2.1.0-rc.3-193-g55b976c0 to test that but if i could just revert to the previous build?
Owner

@mooki000 commented on GitHub · 2026-11-12 12:46 UTC · original

Thanks for the logs. Both v2.1.0 sessions show memory pressure: VRAM peaks at 16,413 / 14,903 MB allowed and 16,291 / 14,965 MB, with private memory reaching 40,515 / 40,824 MB. Both use 3840×2160, 3× interpolation and an 11,000 MB texture budget during play. That is a useful lead for #993, but the logs do not mark the exact slow return, so it is not yet a diagnosis.

The 2.1.0-rc.3-193-g55b976c0 build you tried is newer than v2.1.0, not a downgrade, and already contains #2198's compaction fix. Could you attach that build's log and say whether the same outward-and-return route still slows with the VRAM budget on Auto? Keep a backup of your saves before comparing older builds; the existing logs do not provide a matched-route version comparison.

Here, 11 laps over about 2 h 26 min in one process did not reproduce sustained slowdown or persistent corruption: Linux, RTX 4090/NVIDIA 615.78.08, main 09aac439 plus #2292's read-only state stream, headless, 2048×1152, DLAA, 60 cap, no frame generation. These were debug area transitions and entry views with other work running, not a benchmark. The rising target count in that tour is explained too: all 77 samples balance created minus destroyed against live images; the growth is deliberately retained finalized textures, mostly terrain tiles protected by the pre-release fix 7b774bae, not orphaned target images. That does not establish that your Windows/RTX 5080 problem is fixed.

> [@mooki000](https://github.com/mooki000) commented on GitHub · 2026-11-12 12:46 UTC · [original](https://github.com/NotRexed/Hookshot/issues/2615#issuecomment-6823990087) Thanks for the logs. Both v2.1.0 sessions show memory pressure: VRAM peaks at **16,413 / 14,903 MB allowed** and **16,291 / 14,965 MB**, with private memory reaching **40,515 / 40,824 MB**. Both use 3840×2160, 3× interpolation and an 11,000 MB texture budget during play. That is a useful lead for #993, but the logs do not mark the exact slow return, so it is not yet a diagnosis. The `2.1.0-rc.3-193-g55b976c0` build you tried is newer than v2.1.0, not a downgrade, and already contains #2198's compaction fix. Could you attach that build's log and say whether the same outward-and-return route still slows with the VRAM budget on Auto? Keep a backup of your saves before comparing older builds; the existing logs do not provide a matched-route version comparison. Here, 11 laps over about 2 h 26 min in one process did not reproduce sustained slowdown or persistent corruption: Linux, RTX 4090/NVIDIA 615.78.08, main `09aac439` plus #2292's read-only state stream, headless, 2048×1152, DLAA, 60 cap, no frame generation. These were debug area transitions and entry views with other work running, not a benchmark. The rising target count in that tour is explained too: all 77 samples balance created minus destroyed against live images; the growth is deliberately retained finalized textures, mostly terrain tiles protected by the pre-release fix [7b774bae](https://git.magicbean.dev/MagicBeanDev/Hookshot/commit/7b774bae6416ddebefab35d833150c21cde01400), not orphaned target images. That does not establish that your Windows/RTX 5080 problem is fixed. <!-- github-sync:c:6823990087 -->
Commenting is not possible because the repository is archived.
No milestone
No assignees
2 participants
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set

Reference
MagicBeanDev/Hookshot#2615
No description provided.