HUGE VRAM and RAM leaks #2395

Closed
opened 2026-11-11 17:11:06 +00:00 by WheatleyOSv2 · 4 comments
WheatleyOSv2 commented 2026-11-11 17:11:06 +00:00 (Migrated from github.com)

Hookshot version

v2.1.0

What kind of problem

Low fps or stutter

What happened

Incredible VRAM and RAM consumption, leading to a critical drop in performance.

image

Where in the game

After several transitions between locations.

How often

Every time

Graphics card and driver version

RTX 5070, 621.14

CPU, RAM and operating system

Ryzen 5 3600, 32 GB

Changes from the defaults

ReShade

Other software hooked into the game

Yes (named under "Changes from the defaults")

hookshot_log.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 v2.1.0 ### What kind of problem Low fps or stutter ### What happened Incredible VRAM and RAM consumption, leading to a critical drop in performance. <img width="884" height="438" alt="image" src="https://github.com/user-attachments/assets/14e3f5cb-8e2b-4e0f-b347-1aab2aa26a24" /> ### Where in the game After several transitions between locations. ### How often Every time ### Graphics card and driver version RTX 5070, 621.14 ### CPU, RAM and operating system Ryzen 5 3600, 32 GB ### Changes from the defaults ReShade ### Other software hooked into the game Yes (named under "Changes from the defaults") ### hookshot_log.txt [hookshot_log.txt](https://github.com/user-attachments/files/34698404/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.
Heliumrich commented 2026-11-11 18:10:57 +00:00 (Migrated from github.com)

This AI slop stuff is stupid.
It either load ASTC textures as uncompressed (huge vram usage) or as BC3 (which is stupidely inefficient and low quality ; it should use BC4 for normal maps, BC7 for albedo/diffuse, etc.)

It's coming from yuzu logic, which was less stupid because it was a general emulator. But here considering it's tailored to a specific game, and looking at the game files, it's easy to know what each texture type is and use the proper format that result in lower VRAM usage and way higher quality (especially normal maps)
Just BC4 instead of BC3 for normal maps, you would shave off multiple GB of VRAM usage, and increase quality a lot.

This AI slop stuff is stupid. It either load ASTC textures as uncompressed (huge vram usage) or as BC3 (which is stupidely inefficient and low quality ; it should use BC4 for normal maps, BC7 for albedo/diffuse, etc.) It's coming from yuzu logic, which was less stupid because it was a general emulator. But here considering it's tailored to a specific game, and looking at the game files, it's easy to know what each texture type is and use the proper format that result in lower VRAM usage and way higher quality (especially normal maps) Just BC4 instead of BC3 for normal maps, you would shave off multiple GB of VRAM usage, and increase quality a lot.
codefaux commented 2026-11-11 19:06:09 +00:00 (Migrated from github.com)

@WheatleyOSv2 - This has not provided evidence of a memory leak.

Leaks inflate memory over time as new memory is allocated and old memory is not released, until memory is exhausted and crashing. Using lots of memory is not a memory leak. Please clarify; are you upset with total memory usage, or are you seeing steadily ballooning memory usage which results in an OOM crash?

Does this issue still occur when you aren't using an unsupported external graphical modification/hook program?

Just ...

@Heliumrich - Pull requests welcome. This project literally could not exist within a year of the game's release without AI, but the project welcomes contributions from humans too. It sure sounds a lot like you know everything required to help out, why not go for it?

@WheatleyOSv2 - This has not provided evidence of a memory leak. Leaks inflate memory over time as new memory is allocated and old memory is not released, until memory is exhausted and crashing. Using lots of memory is not a memory leak. Please clarify; are you upset with total memory usage, or are you seeing steadily ballooning memory usage which results in an OOM crash? Does this issue still occur when you aren't using an unsupported external graphical modification/hook program? > Just ... @Heliumrich - Pull requests welcome. This project literally could not exist within a year of the game's release without AI, but the project welcomes contributions from humans too. It sure sounds a lot like you know everything required to help out, why not go for it?
WheatleyOSv2 commented 2026-11-11 19:11:38 +00:00 (Migrated from github.com)

@codefaux
The problem occurs even without ReShade; a few transitions between locations are enough to fill up the VRAM and start offloading data to RAM.

In another project — NX2Navi — even with ReShade and DLSS5 enabled, VRAM consumption does not exceed 8 GB.

Also, in the previous version of Hookshot there were no such problems, neither with nor without ReShade.

@codefaux The problem occurs even without ReShade; a few transitions between locations are enough to fill up the VRAM and start offloading data to RAM. In another project — NX2Navi — even with ReShade and DLSS5 enabled, VRAM consumption does not exceed 8 GB. Also, in the previous version of Hookshot there were no such problems, neither with nor without ReShade.
NotRexed commented 2026-11-11 21:29:51 +00:00 (Migrated from github.com)

Thanks for the report. Reports of the same problem are now collected in #993, which lists this one with your details and links back here, so one fix closes them together and the open list stays readable. Follow #993 for the fix, and add new details there (or here). If this turns out to be a separate problem it will be reopened.

Thanks for the report. Reports of the same problem are now collected in #993, which lists this one with your details and links back here, so one fix closes them together and the open list stays readable. Follow #993 for the fix, and add new details there (or here). If this turns out to be a separate problem it will be reopened.
Commenting is not possible because the repository is archived.
No milestone
No assignees
1 participant
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#2395
No description provided.