I'd be real careful about stopping at legendary quality 2. For getting the initial batch of modules I have no argument because there's a 60 second energy cost on a t3 module line which makes it non-trivial.h.q.droid wrote: Wed Jun 11, 2025 2:03 pm My 2 cents: legendary quality 2s are much cheaper and provide better values than less-than-legendary quality 3s. My path was to first roll asteroids with normal quality 2, then get a bunch of legendary quality 2s and roll asteroids with that. And only upcycle EM plants once I got enough legendary quality 2s to fill everything.
I have to have something better than a markov matrix to explain this because I don't know if people have that background universally and my bachelors degree never really addressed that well. I can use a recycler and the state transition from epic to legendary or uncommon to rare to show why lowering quality is dicey at the point of recycling.
When people are doing two distinct experiments, at some point they're asked to draw a tree of the outcomes and count them. You divide the number of events of interest by the number of outcomes to come to a probability between 0 and 1. And after doing that a few times you realize that if there's n and k possible outcomes for experiments 1 and 2, get n*k total outcomes. And if there's some number i of successes you have i/n*k. Some creativity is involved in finding i because I can play games like say "you succeed on a 2 and six when you see 3,4,5 on a two and 1,2 on a six".
If you're a junior in an engineering field, the above is because I have to keep it simple for myself and maybe other folks.
So the recycler part of the chain will say "1 in four come back". That's the first experiment and no amount of effort will change it. 4 t2's add up to 20%, a .2 or 1 in 5 chance of getting the next best item. In that case I can say 1 when I get one item from a recycler in four and then one in five from a module for 1 in 20 to upgrade on tier.
If I use t3's that's 24.6%, close to 25%. So 1 in four. If I where to compare those as a ratio, my chance of getting any upgrade is 4/5 by using t2's.
Say I get 16 items together. That's nearly a yellow belt. If I cut the belt with recycler, it's like I divide a square of trash four times and then keep a good section. Then repeat that and get another section, then I'm done. But each time I do that, I have the choice of rolling a d5 and a d4 and if I I'm rolling against a heap of items going 1 a second, so it suggests I'm not taking an 80% loss over time but something much worse. The only reason I'd suffer that is if I attack the problem with a lot of numbers, and asteroid mining effectively gives you that. I'll throw 200 em plants at a module and ship a unique part to avoid that loss being experienced by my factory, collectively.
I said 18% because if I have an unused foundry on a space platform those rocks are being thrown out anyways. And a 125% bonus on a quality drill is going to inflict a penalty but a mine which isn't being used is 0 productivity by definition. And when it counts it's going to be 1 in 16 for an upgrade cycle and I need to to that five times by recycling w/out productivity. I'll actually do stuff like throw empty beacon on a build to tune the rate of a speed 3 module down; because they have the best power to weight ratio. But just using speed 2's and speed 1's at legendary quality is another option if you want to try and fine tune for a rate with minimal modules without a regard for waste. Don't know how much it helps and you need to manually test each time; and probably in an environment where you can see an hour in a minute if you can't do great math.
One thing people might not have noticed: asteroid rerolling is 3 rolls as I read it, one for a 2/5, one for 1/5 and one for a 1/5. A reroller station only takes 2 modules, so that's 12.5% or so or 1/8 at best. So you have a 2 in 40, one in 40 and one in 40 chance being thrown collectively off one rock. And the cycle time is 2 seconds of energy. If the machines weren't memoryless I'd say that's a 1 in 10 chance. What it seems like is I'm getting 1 in 20 of a material I'm interested in, 1 in 20 of a material I'm interested in, then losing a 1 in 5 chance of calcite which is so valuable I may as well as keep it!
How long does it take a 1 in 16 recycler to recycle an asteroid?
. I don't see any bad foot here from me, i try to correct what i understand being wrong in messages that i felt were adressed to me due to the quotation .... i may be wrong myself indeed, but it doesn't appear to have been the case, which is easier when you don't pretend something is "the best" but try to use your brain first to consider the different situation and that different tools are appropriate to serve different purpose making it a futile quest of seeking for "the best setup" since it depend on the situation and your objectives.