Astro Hacker News - Show HN: Factorio but with unreliable components

katzenq |next [-]

It doesn't look like a human designed this or wrote any of the text. The layout is broken on a phone. What work did you actually do here?

zug_zug |root |parent [-]

I appreciate those two points of feedback. I'll work on those today.

netsharc |root |parent |next [-]

Hah, output of a politician LLM.

How to not answer the question...

katzenq |root |parent |previous [-]

I'll take that as 'none'. When you say work do you mean telling a chatbot to make a change?

bigyabai |root |parent [-]

I take it you haven't seen what game development looks like in the past half-decade, have you?

katzenq |root |parent [-]

I've been doing it for a lot longer than that, in communities where trying to pass off generated shit as your own work gets you rightfully shamed. Actual game development still values human art and effort.

delichon |next |previous [-]

Factorio already has unreliable components, it just models them as random alien attacks. It fits with Murphy's Law surprising well.

ateesdalejr |next |previous [-]

I played through scenario 1, and thought it was really painful. Which imo as it should be, but why not code a factorio mod with this functionality? I think it would be even more fun, and less of a UX/UI hassle?

ateesdalejr |root |parent [-]

Most of the pain was in the controls just being really nitpicky / not what I'm used to when playing regular factorio. E.g. the keypress shortcuts not always working / weirdly returning to default to place belts, etc. I found myself wanting underground belts a few times.

budro |next |previous [-]

Semi-interesting concept, but it's initially _very_ hard to get into because of how AI slop-coded the intro is. Do your own writing! Also the game was way too manual, you can't just call any game with conveyor belts Factorio. Not once in the game did it feel like I automated any process.

If you can get past all of that, the game does get somewhat interesting. The first level is just a game of APM, replacing your components with higher quality components until 100% and you win. The 2nd level puts reliability behind identification which has a cooldown. This was pretty neat because you discover how to use the markers + feedback loops to actually create a useful reliability comparator. You then use that to only identify candidates with known-higher reliability.

Unfortunately I think that's where the game (and maybe even the concept?) peaks. The 3rd level is basically identical, but with a longer cooldown on identification. That does teach you to use counters as an early source of high quality components to sort against (reliability% = components counted / total components on conveyor) but the winning strategy remains the same. The added Marker-Gate component doesn't really become useful until your reliability is high enough (or not without cascading 5+ or more of them) so the way to beat level 3 is exactly the same as level 2.

It would be cool to see this evolve into something more meaningful, but I have a feeling you'll quickly find out why all automation games have 100% reliable machines.

trylfthsk |root |parent [-]

Same feedback - conditional probability until you eventually straight pipe a 100% reliability sorter to the solution. I think one workaround is to have a max reliability for the level - say 95% - requiring actual infrastructure in the long run.

KomisarzRyba |next |previous [-]

Please stop wasting everyone's time with slop. Learning how to do this yourself would be much, much more rewarding. For both you and us.

mock-possum |previous [-]

Seems like a fun idea, definitely doesn’t play nice with safari on iPhone, every click and drag is a viewport scroll event which is pretty unplayable. I’ll check it out again later when I’m on a real computer.

zug_zug |root |parent [-]

I'll try to make the UI better for mobile browsers. I hadn't intended for it to be on mobile, but let me brainstorm on that.