Skip to content

Conversation

@Pepper0ni
Copy link
Contributor

@Pepper0ni Pepper0ni commented May 13, 2025

The "Entryway" regions in dungeons to filter between Vanilla and MQ dungeons did not have any condition for going back into the boss room, despite the fact they spawned you in front of a locked door, potentially causing impossible seeds. This addresses the problem by moving the key and trials checks to the entryway region. Note that this may make boss key skips more difficult to implement in future, either a better way to handle entrance randomiser of 2 separate regions for entering and exiting the boss room will be needed to handle this.

Build Artifacts

@Malkierian
Copy link
Contributor

So, for entrance shuffle, are the Entryways what you spawn in when entering the dungeon proper from the boss door entrance?

@Pepper0ni
Copy link
Contributor Author

Entryways are connection regions that exist so entrance shuffle can properly interact with the split between MQ and Vanilla dungeons. They do not represent any real space in the game, they simply serve to direct the player to the correct version of the dungeon or to the shuffled entrance, depending on what direction they enter from.

The issue is that, because they were still real regions, they also allow you to immediately turn around and go back the way you came in logic, which is a different place in decoupled. The real fix to this is to refactor entrance shuffle to be ok with multiple references to the same shuffled entrance, but that's a while off, so I patched the immediate problem.

@Malkierian
Copy link
Contributor

I wasn't asking about real space in the game. I was asking which side of the boss entrance is which.

@Pepper0ni
Copy link
Contributor Author

Oh, in that case yes they are on the "Dungeon" side as the Entryway to Boss Room entrance is what gets shuffled.

@Pepper0ni
Copy link
Contributor Author

I noticed that the child dungeon doors have non-key requirements which also needed addressing and could vary by quest, so I decided to apply a "proper" filter to them.

Some secondary testing would be appreciated as I currently need to merge #5467 to build and that makes applying fixes clunky.

@Malkierian
Copy link
Contributor

If you wanted to test that yourself now that the gcc15 issues are resolved, we should probably get this in.

@Pepper0ni
Copy link
Contributor Author

Pepper0ni commented May 18, 2025

Gave it a look and it seems good, though it's hard to test deku while #5514 makes it hard to remove shields from the equation

@Malkierian
Copy link
Contributor

Just a quick question, the fact that the pairs of pairs don't map inversely 1:1 doesn't cause issues?

@Pepper0ni
Copy link
Contributor Author

I'm not an expert on entrance shuffle but pairs are declared explicitly, so there shouldn't be an issue

@Malkierian Malkierian merged commit 5bf3761 into HarbourMasters:develop-blair May 21, 2025
6 checks passed
ReddestDream pushed a commit to ReddestDream/Shipwright that referenced this pull request May 21, 2025
…ourMasters#5484)

* Fix boss and tower entryways in decoupled

* Add patches to child boss room entryways

* Apply fixes

* remember to set ALL the new exits
ReddestDream pushed a commit to ReddestDream/Shipwright that referenced this pull request May 23, 2025
…ourMasters#5484)

* Fix boss and tower entryways in decoupled

* Add patches to child boss room entryways

* Apply fixes

* remember to set ALL the new exits
krazyjakee pushed a commit to krazyjakee/OOT that referenced this pull request Sep 6, 2025
…ourMasters#5484)

* Fix boss and tower entryways in decoupled

* Add patches to child boss room entryways

* Apply fixes

* remember to set ALL the new exits
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants