![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
(no subject)
Oct. 30th, 2016 12:16 pmIt's not terribly original, but here's Piper's Halloween costume for the party that Danae and I are going to today.
I might do more preparation next year and make it actually fold open and have a silly little poem about her in it. (It occurs to me that I'm already thinking about having her around next year. I have no idea whether that will be the case or not. I really like her, but I don't know if she's the dog I'd choose to have permanently if I was to choose a dog. And honestly, I don't think we could afford her medical bills on our own either. *sighs*)
Anyway, I've been doing a lot of Factorio. It's a wonderful distraction. Before sitting down to write this, I was lying on the couch with graph paper and trying to figure out how to graph multivariate (is that the right term?) functions with either 4 or 8 inputs and one output for arbitrary inputs so I can make the supply system that feeds my electronic circuit production line faster and more efficient. I've never done this kind of math. Google to the rescue! Though it may be that, as long as the splits and merges are done correctly so that all outputs are equal, knowing how input asymmetry effects the system's total throughput isn't critical. (It kind of looks like, from a quick look around, this kind of thing isn't exactly plotable in the same way that a single variable function would be. I wonder if it would help to plot it once for each input variable while holding the others constant?)
I've never cared much about achievements in games - they usually annoy me if I bother thinking about them at all - but I've never actually gotten every achievement for a game before, and with Factorio they give me fun goals to work toward while I play around with designs. I've been really enjoying doing things like finishing the game in under 8 hours, or doing it without use of laser turrets. I'm working on the last achievement I haven't gotten; making 20 million electronic circuits. It's the least commonly attained one in the game. Working on it, I can see why; the scale of this project is far larger than anything I've done before. Even at around 5000 a minute (a rate that I haven't managed to keep consistent yet, but am working on), it takes over three hours to produce a million circuits. I keep tweaking things to improve production, but I go through cycles of maximizing my current design and having to start from scratch with a higher capacity layout. In proper doses, this has been a lot of fun!
ETA:
Ok, I did some graphs where I held some inputs constant and increased other inputs synchronously and I figured out what I need to know. In the system in consideration, each increase of 1 to input on any line results in an increase of .25 to each output line, up to individual capacity of input/output lines. I guess I simply proved that my four-way merge/split is doing its job. Which is good! Now on to design an 8-to-8 split/merge. Four full belts feeding the green circuit makers is not enough throughput.
I might do more preparation next year and make it actually fold open and have a silly little poem about her in it. (It occurs to me that I'm already thinking about having her around next year. I have no idea whether that will be the case or not. I really like her, but I don't know if she's the dog I'd choose to have permanently if I was to choose a dog. And honestly, I don't think we could afford her medical bills on our own either. *sighs*)
Anyway, I've been doing a lot of Factorio. It's a wonderful distraction. Before sitting down to write this, I was lying on the couch with graph paper and trying to figure out how to graph multivariate (is that the right term?) functions with either 4 or 8 inputs and one output for arbitrary inputs so I can make the supply system that feeds my electronic circuit production line faster and more efficient. I've never done this kind of math. Google to the rescue! Though it may be that, as long as the splits and merges are done correctly so that all outputs are equal, knowing how input asymmetry effects the system's total throughput isn't critical. (It kind of looks like, from a quick look around, this kind of thing isn't exactly plotable in the same way that a single variable function would be. I wonder if it would help to plot it once for each input variable while holding the others constant?)
I've never cared much about achievements in games - they usually annoy me if I bother thinking about them at all - but I've never actually gotten every achievement for a game before, and with Factorio they give me fun goals to work toward while I play around with designs. I've been really enjoying doing things like finishing the game in under 8 hours, or doing it without use of laser turrets. I'm working on the last achievement I haven't gotten; making 20 million electronic circuits. It's the least commonly attained one in the game. Working on it, I can see why; the scale of this project is far larger than anything I've done before. Even at around 5000 a minute (a rate that I haven't managed to keep consistent yet, but am working on), it takes over three hours to produce a million circuits. I keep tweaking things to improve production, but I go through cycles of maximizing my current design and having to start from scratch with a higher capacity layout. In proper doses, this has been a lot of fun!
ETA:
Ok, I did some graphs where I held some inputs constant and increased other inputs synchronously and I figured out what I need to know. In the system in consideration, each increase of 1 to input on any line results in an increase of .25 to each output line, up to individual capacity of input/output lines. I guess I simply proved that my four-way merge/split is doing its job. Which is good! Now on to design an 8-to-8 split/merge. Four full belts feeding the green circuit makers is not enough throughput.
