To the people calling this Tyrannobuilder... I don't think it is... It's
You must be registered to see the links
. (Decoding tool in
You must be registered to see the links
).
Edit: No need to decode chapter 3... it hasn't been encoded.
The engine is crap. [...] I am still hoping the author switches over to renpy or somebody creates an unauthorized version because this is almost unplayable on some systems.
The main issue is that NVM handle variables in a really odd way. I'm sure in it's own editor... it's perfectly sensible. But in the json files (once they're decoded)... I've found it almost impossible. It's pretty easy to get a list of variables, but figuring out if it's doing +1 / -1 / set True / set False ... Well, if it's possible... I'm clearly missing something fundamental. Anyone trying to convert the game would need to be able to play-test the game enough to guess how the variables are being updated... or alternatively, the author would need to collaborate.
Edit: I'm getting closer. Turns out variables are stored in generic arrays. Each variable has an associated "index" which tells it which array element to use. So one of the reasons I couldn't follow things was that I was looking for the variable name and not it's index. There appears to be 3 arrays (integers, strings and booleans). Each array has what it calls a domain, which appears to be a unique name. I'm not sure if that's changeable because I've only looked at this one game. Anyway... this is just my way of putting an explanation out there, if anyone goes deep delving into VNM code.
Based on 3rd hand information... I do know the author has found using VNM a lot easier to write with than RenPy. And whilst I too wish he'd migrate back to RenPy, my general advice to any developer is to write your game for yourself and ignore those whining gits on the internet. Realistically, swapping back to RenPy would be a lot of work - though given the huge dip in his Patreon numbers... maybe it would be worth it.