Does LuaSurface::can_place_entity() work for you? https://lua-api.factorio.com/latest/classes/LuaSurface.html#can_place_entity


I'm already using that but the issue is that the coordinates that are used to determine placement aren't necessarily the ones passed in. And if you don't know the ...

Yes, the LuaPlayer->build_from_cursor() function. In the UI, as you move the cursor it's constantly refreshed showing the exact location the build will happen and whether it will succeed. In the API though there doesn't seem to be any way to get that information that I can find.

Is there a lua function that will tell you the coordinates where build_from_cursor() will try and build something? In the game, you can visually see the cursor snap to the possible placements. But via the API I don't see a way to do this. In my MCP code, a successful placement results in a callback ...

Interesting. How far Claude went into playing the game and building a factory, in your testing?


Not very far. I've got a couple of blog posts with some random thoughts but in general it's not very good. It burns through tokens so I can't let it just run for days on end. In my experience it will ...

I'm not sure where this fits in terms of the Boards here. It's not actually a mod and it's only sort of a tool. I've been working on a Factorio MCP server which lets you wire up Factorio to a language model and let it try and play Factorio. I'll be honest up front and say that I haven't had much ...

(which just happens naturally when you still build something or there is not enough demand yet)

My super simple technique for this case is to have a circuit disable the agri towers when I'm working. I have a combinator wired up to them and the towers only activate when I have the combinator put ...

This seems like a mystery. Clearly your VM has some level of working DNS and reachability to auth.factorio.com. Where did the systemd service definitions come from? Those aren't part of the official distribution. It seems like you'd almost have to go out of your way to craft a systemd startup file ...

I'll add my vote for an Linux ARM64 build. It would let us run a server in a VM on a Mac. I would guess that all of the x86-isms have already been removed to support the Mac client so it might be a pretty small lift to get it compiled. In a perfect world it wouldn't be any more work than just ...

Can you show the full log? journalctl -u factorio.service




Looking there I see:
Error HttpSharedState.cpp:175: CURL failed: code:6, 6; Could not resolve host: auth.factorio.com

Error ServerMultiplayerManager.cpp:84: MultiplayerManager failed: Failed to reach auth server. Code: 520. Error ...

It works fine. I've only played SA on my deck. I haven't yet made it to Aquillo but so far no issues. Other than maybe the screen is a bit small so sometimes I have to turn off the minimap to see the whole tooltip for some things. Having a minimap toggle control would be nice instead of having to go ...

Another possibility would be to have blueprints/planners be something a requester chest could ask for. Then the logistic bots could gather them up in one place. I've run into similar issues when I have to get rid of planners in order to travel to another planet and if you aren't careful it's easy to ...

PssX wrote: Wed Dec 18, 2024 10:54 pm It's like 10th thread about it. They should just merge these two technologies together.

Or at least add the calcite requirement to the foundry page. If it has two craft requirements then they should both be listed.

not sure what I'm doing wrong here, i copy and pasted the blueprint as-is, then I swapped the fast inserter out for a bulk. Then I cut/paste the constant combinators to move them into a position that I could fit them on the platform. That kept all the wires connected it. I double checked that all ...

Also, maybe 8 yumako trees isn't enough (or is too much)?


8 is definitely too much, since I don't run out of yumiko fruit and I have to mash it and burn the mash. But 7 was almost too few.

I settled at 8 yumiko trees and 6 jellynut trees. Its not ideal, as it results in more pollution than ...

Kirk's factorio calculator says 1 biolab making science needs 25 yamuko and 10 jellynut per minute (although I don't think it's taking nutrient needs into account). That matches my experience. I have 1 tower for each fruit and it keeps between 1 and 2 biolabs busy. And the towers are disabled ...

Kirk's factorio calculator says 1 biolab making science needs 25 yamuko and 10 jellynut per minute (although I don't think it's taking nutrient needs into account). That matches my experience. I have 1 tower for each fruit and it keeps between 1 and 2 biolabs busy. And the towers are disabled ...

If you only have one biolab making science, you only need a single tree of each type. Less than a single tree. Overproducing fruit so it backs up and loses freshness feels a lot less optimal than disabling the towers. Granted my Gleba experience is limited, but I think two cardinal rules are don ...

Rather than burning fruits, process them (even in assemblers), to get back seeds, then burn mush /jelly. Otherwise you might run out or seeds (not too likely, but possible). But also have a way of burning seeds if there is too much of them.
I made constant flow fruit bus, then some single ...

The FFF isn't clear to me. Are the combat balancing changes only in the expansion? Or will they be in the base game as well? In the expansion I trust that having played it the devs know that the PLD changes are needed in the expansion since you can have legendary power armor filled with legendary ...

Very cool! Something not mentioned that I think would really be an additional improvement on top of interrupts would be the ability to temporarily disable interrupts, or fully disable interrupts and only check them at specific points in a schedule.

For example, if you know that the refueling ...