Mod info request (reworked)

[Tonn013]

The original context of this post met a poor reception and was seemingly unclear. So I've changed the goal from private correspondence to open air discussion. Not really what I was hoping for but it will do. My forum etiquette is admittedly only at beginner levels as my experience in forums is miniscule at best. My apologies for the poor execution before this rework of the post. I hope this is more agreeable.

Idea: Compound subassemblers. The idea is to create a modded assembler that can be "attached" to an adjacent assembler, it can only be set to assemble recipes for the ingredients requested by the parent assembler and place it's output directly into the input of the parent. Effectively eliminating the need for an assembler-to-assembler inserter. The original post that had me thinking about this was wanting something similar for the purpose of lessening UPS demand of the dozens apon dozen of inserters that this would eliminate. His idea was different in a few smalls ways, this version seemed like it might be what he was moving toward and I cant say I hate the idea either, though, not for UPS purposes. The only balance issue I can think of it's the power freed by the missing inserters but that seems easy enough to solve via subassembler power demand.

The question is how viable is the idea and how difficult would something like this be to accomplish for an inexperienced modder?

[Qon]

If you had at least outlined the key features of the mod and described its purpose and things like that so people could discuss it in this thread then it would make it easier for people to help and thus more likely that you would get any. Also I can't see if you have already been helped if it's in PMs. I could tell you if your mod would work, but I don't know anything about it. If I want to help then I first have to PM you and ask about it and then I might get told that you have already been helped. Or I get some unclear description where I have to ask for lots of clarifications because it's not in the open where others could have already asked those questions already. And then it might be a pointless and boring mod idea that I don't want to help with. So why bother PMing you when I might just end up wasting my time?

[eradicator]

TL;DR: In 99.9% of all cases if you want help on the internet you should ask the question you have. Asking meta questions about asking questions is just wasting everyones time. I actually saw this thread hours ago and ignored it because there was no actual question .

[Xorimuth]

If you had at least outlined the key features of the mod and described its purpose and things like that so people could discuss it in this thread then it would make it easier for people to help and thus more likely that you would get any. Also I can't see if you have already been helped if it's in PMs. I could tell you if your mod would work, but I don't know anything about it. If I want to help then I first have to PM you and ask about it and then I might get told that you have already been helped. Or I get some unclear description where I have to ask for lots of clarifications because it's not in the open where others could have already asked those questions already. And then it might be a pointless and boring mod idea that I don't want to help with. So why bother PMing you when I might just end up wasting my time?

Agreed. https://dontasktoask.com/

[eradicator]

The question is how viable is the idea and how difficult would something like this be

You cant get rid of the inserter on a technical level. You can only hide it so the player doesn't see it. You could use a really fast™ inserter ofc which might reduce UPS slightly, but there are already myrads of inserter mods out there so 0 UPS improvement over the existing meta-game.

to accomplish for an inexperienced modder?

Compound entities are 7/10 I'd say. Not something you want to start with without at least a good bunch of programming experience.

[Tonn013]

Can you elaborate "on a technical level" plz? Is that to mean the movement of an item from entity to entity requires the code of an inserter, no exception?

If that is the case, would it be possible to avoid the need to move it by instanciating the new item directly into the parent?

[Xorimuth]

Can you elaborate "on a technical level" plz? Is that to mean the movement of an item from entity to entity requires the code of an inserter, no exception?

You can move an item from an entity to entity without a hidden inserter, there is no simple trigger to get this code to run at the right time for the right assembler-assembler connection so using a hidden inserter would be both easier and more UPS efficient.

In summary, it is possible, but it won't be any more UPS efficient than the base game.

[Tonn013]

Okay. The other guy was the 1 wanting efficiency, I just liked the concept. I'll consider myself satisfied all the same tho. Thank you for your time.

[eradicator]

Can you elaborate "on a technical level" plz? Is that to mean the movement of an item from entity to entity requires the code of an inserter, no exception?

If that is the case, would it be possible to avoid the need to move it by instanciating the new item directly into the parent?

Very broadly speaking in factorio there are two vectors from which you can engage any given problem. Either you abuse native entities like assemblers, inserters, etc. Basically changing only the texture, name and a few numbers like crafting speed. This approach is limited in what it can solve, but has no performance penalty. The second vector is lua control scripting which is versatile, but so ridiculously slow (compared to above) that you can only use it for things that happen infrequenty. Moving items via script at scale would very quickly bring you down into single-digit UPS domain - unplayable. So your problem doesn't require "the code of" an inserter. It requires an actual inserter.