Open ahicks92 opened 1 month ago
LevFendi
commented
1 month ago You can start by fixing rthe obvious issues like missing sections and incorrect keybinds. That will provide high value quickly while the repeated discussions will have less urgency
ahicks92
commented
1 month ago I'm sorta backburnering this for what it's worth. While the discussions are repeated we both have a long backlog. There's also a case to be made to wait until 2.0 anyway for new rails.
But yes a pass for the keybinds is probably worth it.
I dunno if I'd say we have missing sections. It's more missing info. I'll think about that though. I'm sure there's a section I could think to add.
The one we have on the wiki isn't bad considering historical context, but it doesn't explain why, doesn't cover that you can manually place signals and train stops, doesn't cover why you might want to manually do these things. Someone needs to review this.
The biggest thing is signals. I've had to explain them to people a lot. I don't think we even like clearly call out the whole train stops on the right thing. We don't explain important concepts like LHD/RHD. We don't explain regular vs chain signals. The mod even warns if you try to do anything but chain, and indeed the wiki flat out just says don't bother beyond that it's not safe. There are also minor things; we haven't for example updated this for the new keystroke remap. Also where are the dimensions of corners? We don't even tell people "put rails 6 tiles apart and you can merge them".
Dropping some modular blueprints probably wouldn't go amiss. I recently proposed an (unfinished) intersection. We could do that for various levels and kinds of play. For example:
Basically based on discord, I don't think people "get" trains. And I don't think we provide enough for them to have a chance at it either. The current document is, I guess in a way out of date. It's hard to articulate. It reads like someone's college notes. Some specific procedures and some general "if you already know this is how accessible it is".
Also, like for another thing there's (for example) a nice 6-chest tileable load/unload pattern where medium electric poles can cover 3 and you get pole chest x6 pole... and it just fits together. Or, going into the somewhat not obvious thing about how you need buffers because the train is going to leave and fill up and if you don't everything will stop. I don't know how far to take it to be honest--but further than we do. I'm not sure we even explain transport belt 900/min throughput considerations at the moment--good luck working out that's why your train station is always full. Honestly good luck working out how to even get both lanes filled up, given that all the "obvious" "simple" designs won't do that and people often have troubles getting how inserters work.
But...I don't have a plan. Getting something really good here is complicated to say the least. I'm not sure what we can do. It's been proposed I do a stream like the demo however that's very hard to reference after, and even so trying to explain that in realtime with how complex tracks get is...I dunno. It's a hard problem in general. We need like a good analogy or something to build around--a "core" to make it fit together. Maybe we can present it as a case study? It's unclear to me.