- Aug 12, 2017
- 3,746
- 3,106
- Sep 30, 2017
- 82
- 106
I'm still around btw. One of the issues I've had with developing is less to do with the complexities of doing the work and more to do with the fact the technology is moving too quickly. I've been trying to get Lama3 working but I'm having difficulties, if I could get it working it might improve the workflow a bit and result in having to make less manual changes to the outputs.
On top of this I've learned some new programming tricks which require a full rewrite.
I expect to make 2 protoytypes, 1 for story progression and another for the sandbox mechanics which would be devoid of story but show the players progression. The two would then be merged where the story progression would then be fleshed out for the sandbox.
On top of this I've learned some new programming tricks which require a full rewrite.
I expect to make 2 protoytypes, 1 for story progression and another for the sandbox mechanics which would be devoid of story but show the players progression. The two would then be merged where the story progression would then be fleshed out for the sandbox.
- Sep 30, 2017
- 82
- 106
Hey, I've been messing around with Civitai, KREA AI and Leonardo AI.
And I know a bit about Twine.
Hit me up if you need help with your AI Avatar.
Not using Twine and for good reasons. I can make this a lot less difficult for people to use compared to Twine. That way if you did want to contribute you could do so without any issue. Twine introduces the issue of spaghetti code. I'd rather most people just shove the content into the game, set some parameters and be done with it.
Every programming language does it. The problem lies between the keyboard and the chair, not in the language.
- Dec 9, 2018
- 4,389
- 5,540
Maybe set up a dev thread for your 'ai' garbage rather than infesting another game's thread with it.
- Oct 8, 2016
- 59
- 59
Sounds like you want to make a whole new game engine, using AI, that will allow people to contribute at a story level without the creation of complex code? If you can solve for that, you should probably aim a bit higher than remaking a 8 year old twine porn game.
There lies the problem right, that this is totally a waste of time. Besides PE's community have been gullible fools for the past 8 years so they don't deserve shit.
Irony since you're infesting this thread for a dead game with your absolutely pointless comment.
To an extent however I've found Twines performance with PE and Secretary to be sub-optimal because its not really designed for such heavy use of graphics and UI this is why I don't like Twine. Better instead to use an engine that is built for graphics first. Twine does have a very nice editor though for planning layout. But its not really designed to make the types of games PE and Secretary were trying to make which is why they've taken so long to make.
It's not the UI that slows the game down, at least in PE's case. It's the (depth of) history stack combined with the size of game state data. Correct me if i'm wrong, but you've only started to dabble with this couple months ago, so you're trying to speak authoritatively about something you have very little actual knowledge of..?
On a porn game forum for a dead game. Nevermind. Enjoy your crap.
He's right though. PE's performance tanks (and most other complex twine games) because as the game goes on and more variables are tracked, they're tracked between history states as well. Page transitions get longer and longer in a save because of that. I think PE's history states is like 10-15, which is about the same as other games like Secretary and The Company, both of which have the same problem (The Company slows to a fuckin CRAWL because of this). Changing the history state to like 5 or so brings things snappy again like it was when the game was started. You can try this for yourself with the command "Sugarcube.Config.history.maxstates" (might be iffy on the capitalization, but the console usually autocompletes for me), on a game that's got shitty performance like that, and lower the history state to see it in action.
It's not so much a problem with the UI or graphics, as web browsers themselves are more than optimized to display those things, but because people really like to stretch Twine out to do things it wasn't really designed in the first place to do. For tracking a handful of variables, it's fine, but when you start getting in to the hundreds or thousands, all of which have to update every page transition, well, it's obvious where the performance hit is coming from.
I don't really have a horse in this argument (whether it's a waste of time or not), and am intrigued to see what you come out with, but I thought I'd clear the air a bit.
sure but this only proves that Twine is not a good tool for scalability. If it can't cope with a constantly expanding history needed for the sheer fact that as a game develops it gets more complex, then it was a poor initial choice for an engine because it just cannot allow for the number of variables and constant global variable tracking needed for a sandbox, that's why RAGS always worked better because it was built for that exact purpose.
I just honestly wished that games like PE weren't such a hassle to make which is why I wanted a framework combined with AI so that games could be made in weeks rather than years. Most of the games on F95Zone take too long. Most of the games on TFGamesite are scams. So it'd be nice if we could just stop all that shit and just get back to having fun. That was the whole reason I wanted to do this.
- Dec 21, 2017
- 498
- 364
Random question. Why wouldn't you just grab the last version before lots of content was removed, import that into twine, add your own content,ending, and avatar, and call it a day? Seems so much simpler to do it that way.
PE specifically was affected by a bug (mentioned earlier in the thread iirc) that was shipping a build with history stack set to something like 50, for debug purposes. Setting it to 10 or so it's supposed to have makes things work at normal/acceptable pace.
I think a lot of that boils down to inexperience for most people. You pick an engine because it's either easy to pick up, or you know a bit of it already, and some time down the line if you keep at it and get better it bites you in the ass as you run up against its limitations. It's the same for a lot of engines, not just Twine.
Hell, it's a common thing in all kinds of facets of life. Like buying a camera because you have a interest in photography, and you start getting decent at it. Then you start running up against issues with that camera, like maybe the lens isn't detachable and it has fixed focal length, or you can't adjust the exposure, or focus like you'd want. So now you're hindered in what you can do, there's all kinds of things that you'd like to do, but you can't.
I guess the difference there is that you can always buy another camera once you've saved up enough, but until then, you just trudge through shit and kick yourself in the ass because you didn't buy a better camera when you started (or chose a better game engine), but there's not really any way you could have known in the first place until you got to that point.
Of course, someone exceptionally more skilled than you could probably run circles around you with that camera/engine, and tell you that it's perfectly fine if you know what you're doing, but not everyone is that guy. Likewise, some other guy, could go back after the fact with a great camera and take all the same photos you did, "but better", and criticize you for not picking the good camera in the first place.
I think I'll stop with the camera analogy, but you get the point. Hopefully... The point is that hindsight is always 20:20. You don't know what you don't know, until you know it.
Because even that was still bad. You'd still run into the exact same problems the PE team was faced with because that version of the game was never designed to go beyond the ending where the parent turns evil, and you chose to either make the bully pay or not. But the therapist was still under developed and the punishment system was still at that point very broken.
Plus the game is extremely linear anyway which sucks. But from all the pushback I'm getting I seriously cbf. I'd rather just go make something new then. No loss on my end, I only spent like a day on the systems and those can be reused. I've been playing a few management games on here and that's scratching a particular itch instead.
- Jun 3, 2019
- 349
- 323
Didn't you already change a Lotta things? Enough so it may as well be new.
I think 1.3 was the last before it went patreon rules nuked and there was a GitHub setup with the src's, quick google should find it for you & then you can scale it to your hearts desire, I was modding it on and off for fun once the game went bad
Game wasn't built in twine by the end (like free cities) I believe people helped dev move beyond to speed up the making.
Game wasn't built in twine by the end (like free cities) I believe people helped dev move beyond to speed up the making.
Last edited:
- Dec 21, 2017
- 498
- 364
Well, if you ever need help figuring out the twine side of things, let me know. I can probably point you in the right direction.