17 May 2026

Lights Collaboration

I should have read John-K's dreamsmith project a bit closer. This format.md shows a clear scheme for light interpretation. A closer look will highly-likely solve my misalignment problem I've been running into. Chopping these .LTSDM files for editing seems to be quite a chore. I do want to edit them by hand as I am not super confident in my ability to computationally stratify them properly yet. If I can hand-stitch a file together, and it works, then I would feel more comfortable moving forward with a computational pipeline. I just need to meticulously confirm how to inject new content in prior to investing so much time making an app to automatically insert data. I will attempt to recruit generative AI to help with the highlighting in ImHex as this would be a major life improvement.

Analysis

When I diff the injection-complete.bin and BSLSBSBaseF.bin I can see that the size of Region 1 is different (40 88 00 00 vs 42 88 00 00 respectively). This may be the culprit for the light seizure. I have changed injection-complete.bin's data to the right size. It seems all other data (the light data) is cut correctly, and its length preserved. If it works with just that one byte adjustment, then we are in business! Once I have access to my flasher, I'll upload it and play it to see if we succeed. Here is the initial attempt.

⟵ Older	Table of Contents	Newer ⟶