That begs the question. What engine is designed for it? I'm not saying it's a bad game. At least up until it had issues with patreon, which is when I stopped following it. But I've noticed a serious sluggishness in games like this, Life at University, The Company. Good games, but, big delays moving screen to screen. Lot of lagging, etc. Makes it hard to play.
Too many variables for a browser engine to handle. When they set out to make these games, there wasn't much documentation or proof they would ever be a problem (how could a 16 core 4Ghz CPU ever struggle with a browser game right?) There are methods now to make them more efficient, like unsetting variables from memory when not needed, but it would be a big hassle to go back and figure out when they can be. And the devs generally feel their best use of time is writing new content, not optimizing.
Wasn't that what you were doing in the first place? As I understood it, you were making your own, better, PE (with blackjack, and hookers), not taking what's there, putting it in a new engine and adding on top.
PE being more of a jumping off point, using similar themes, but making your own thing and leveraging generated content.
Pretty much.
This is why I said there was a need for a framework because Godot doesn't really have a system in place for handling text based games like this, its more designed for visual games. Unity had similar problems though, but was worse becaues of how unperformant it was.
The Unity prototype I had featured the entire wardrobe system with custom art - however it was placeholder, it was very barebones in story. The godot prototype I have is more fleshed out but missing the wardrobe and at this stage I'm unsure how I'd handle the avatar now that AI can do most of the work. I did consider converting the art I already had and interpreting it using AI so I may go with that. The artstyle was reminsent of Good Girl Gone Bad so using AI to move away from that would be ideal.
Not a browser based engine that's for sure. I've built prototypes in unity and godot. The unity one sucked because unity's rich text is seriously lacking and Unity has fallen out of favor due to the company being run by idiots. So I've chosen godot. The fact that it has support for bbcode with images and that link signaling was very easy to get working, plus in compatibility mode it was insanely fast and ran on everything I threw at it. And Godot is really good at handling global variables - especially custom classes which would be necessary for the sandbox elements.
This is why I said there was a need for a framework because Godot doesn't really have a system in place for handling text based games like this, its more designed for visual games. Unity had similar problems though, but was worse becaues of how unperformant it was.
The Unity prototype I had featured the entire wardrobe system with custom art - however it was placeholder, it was very barebones in story. The godot prototype I have is more fleshed out but missing the wardrobe and at this stage I'm unsure how I'd handle the avatar now that AI can do most of the work. I did consider converting the art I already had and interpreting it using AI so I may go with that. The artstyle was reminsent of Good Girl Gone Bad so using AI to move away from that would be ideal.
I don't want to sound rude but you might want to make your own development thread. What you guys are talking about is off-topic and mods might just remove all of this if they so choose. After making the thread put a link in your signature and post something here for all of us.
It almost seems like there's a market for someone to design a new engine. But I wouldn't even know how that would start.
If you hang out in the Twine community you might recognise my alias, and that I have some experience regarding what can and can't be done with Twine.
Project's like this one and Secretary basically use Graphics in one of three ways:
- they display images and videos within the textual content of a Passage.
- they display images within Character / Show inventories.
- they layer images within a set area to implement "Paper Dolls"
- "Passage" meta-data
- Custom JavaScript
- and the selected Story Format's JavaScript based runtime engine.
- the amount of stateful data the Author wants to store within each Moment of Progress History
- the maximum number of Moments the Author configures that Progress History to have.
- extensive looping through arrays of data.
- not optimising the size of the images and videos files being shown.
But that's just my opinion...
The Story Format that this project is built on originally set the default size of the (what you're calling) History Stack to 100 Moments. In a later release of that Story Format the default size was reduced to 40 Moments. A project's Author has always been able to change that size to be whatever they wanted, including being able to change it to only track the current Moment. eg. size equals 1.
No offense taken. I intend to do so in time once I have something tangible.
My knowledge on all this stuff is very limited but just wanted to highlight as I mentioned a tad earlier this game is on github with version 1.3 (the last before the patreon hoo haa) & its in an un compiled state with the SRC's etc. I don't think they were putting through twine by the end (at least not to compile/publish).
As I said my knowledge is pretty basic but figured I'd let you know incase it helps in some way.
As I said my knowledge is pretty basic but figured I'd let you know incase it helps in some way.
Yes, it was fixed. PE1.3200, the one with poor performance, had it at:
Code:
Config.history.maxStates=50; //TODO Set this back to 5 after testingThey used Tweego. It lets you make passages as small snippets on their own, outside of Twine's editor, and then it compiles it into Twine/Sugarcube. Basically, once a Twine game reaches a certain point, you don't have a choice but to use something like that as the Twine editor interface becomes a fucking nightmare. Not that it wasn't horrible to use in the first place.
- Jul 25, 2022
- 9
- 18
Very useful stuff. I had played PE and gotten some slow down issues but never really thought about what was causing them. Nice explanation. <3
- Oct 8, 2016
- 60
- 59
I remember that bug, I may or may not have been the cause of it
...IIRC there was a version of the game with the cut content merged back in, that was released as a proper release, not a mod. It might be worth trawling through the repo to see what was checked in, also check out some of the other branches, there was a bit of a habit of running parallel branches with different content for a while there.
I think Twine/Twigo works "fine" for this style of game, but if I had a choice to start from scratch, I'd probably look at RenPy, as it seems to have good support for these story types of games, but also good state tracking and capabilities to extend. There was a lot of custom work put into this game, which was made more difficult than it needed to be by using Twine. I don't have enough experience with RenPy to say if it would be easier or harder with that engine, but I've seen other games that seem to handle this kind of gameplay well (eg, Headmaster).
At the end of the day, however, we really need to ask why we'd update or change this game at all? Engine wise, changing that would be an absolutely massive task, and you'd end up at the same place we're at now. I think the biggest problems with this game weren't the engine in the end, but rather story progression and endings. If we were to do a remake, I'd start of by re-writing it, then implementing that re-write from scratch.