Change agricultural science spoilage

[spacedog]

TLDR: The spoil timer for agricultural science should be much shorter, but in exchange the science value should not decrease over time.

It's cool you guys tried to do something different here, but it's not fun. The spoilage aspect is fine, but the fact that agricultural science packs have essentially random durability when they reach the labs makes balancing the sciences a real pain. You have to overproduce, but you can't calculate by how much so you have to over-overproduce. And then you can't just let the belt back up because spoilage will clog everything up. So the only options are:

• Bots and requester chests. This doesn't scale well to megabase sized builds, and I'm sure you're aware there are a lot of purists who see bots as "cheating" and want to be able to create belt-only solutions.
• Have a continuously circulating belt. This sucks because there's no way to pull the least fresh packs from an entire belt, so lots of science is wasted to spoilage and you have to over-overproduce even more.
• Complex circuit control where a fixed number of packs are sent to the labs on a timer, and the labs are set up to eject whatever remaining agricultural science is left in them simultaneously when loading the next batch of science packs.

This isn't ideal. I like the spoilage mechanic and how it introduces a time-based optimization puzzle. So just make agricultural science spoil after like 15 or 20 minutes, but don't have progressing spoilage impact the science value. It means we still have to solve the puzzle of getting it off Gleba and to its destination quickly, but we don't have to deal with the stuff above.

I ended up doing the third bullet above as my solution. So, sure, "once you solve it once" and "production is free" and all the things. But it doesn't feel like making us jump through those hoops is adding anything fun to the game.

[mmmPI]

What kind of design would your proposal opens up that isn't currently possible ? = how would you write up the 4) 5)? options if your proposal were added.

I think the TL;DR sounds interesting idea for mod because it would constitute a valid alternate set of rules, but i'm not sure i understand the intent, i feel i would still play the same using 1) 2) or 3) maybe because i'm missing those options you want to be using that would be new. I don't want those "new options" to remove the "spoilage puzzle".

[spacedog]

The proposed change would make the amount of science you get from an agricultural science pack deterministic.

Let's say I'm trying to make 1000 SPM of all sciences. With the current state of the game, I first have to figure out roughly how spoiled my agricultural science will be once it gets to the labs. Let's say I think I can get it to the labs with >= 90% freshness. So I can assume worst case that each agricultural science pack is only worth 90% of the other science packs. So I need to overproduce by 11.111% to make sure I have enough agricultural science -- in the worst case.

But I can't just send that off to the labs blindly, because I can never guarantee an exact freshness value due to several factors -- small timing differences, the way freshness gets averaged depending on how stacks got split/combined all along the way, etc. I can only design around a worst case. That means some of the science packs will be more than 90% fresh. Which means they won't get used up as quickly as the ones that are only at 90% fresh.

Now I have a nastier problem to solve, because my top goal is to ensure the labs stay full. So I have to know when it's time to send more agricultural science off to the labs. But I can't read the lab contents on the circuit network, so all I have to go on is elapsed time. If I just blindly send science packs every N ticks, those packs with more than 90% freshness will eventually cause a back-up in the labs that exceeds the automated insertion limit, and so the inserters won't pick up the next batch of packs that come by. But then I'll run out of packs before the next batch after that comes, and the lab will sit idle.

The solutions to this are the bullet points above, but I'm questioning whether this whole mess is really adding anything fun to the game.

If the agricultural science packs provided the full science value regardless of their freshness, then everything would be deterministic. I'd simply make agricultural science packs at the exact same rate as all the other science packs, and it would balance out correctly. All I'd need to worry about is getting it off Gleba and to the labs in time. (Which is why I think a shorter spoil time would make sense, to emphasize the speed-optimization part of that puzzle.)