This week I (Dark_Brick) continued worked in a team with transcendentalguy, although we ended up mostly going our separate ways (I did this rate solve, he did the amarms) and giving some minor feedback to each other here and there. Thanks to this arrangement, I had more time to do other productive things, like mining bobsidian in minecrap.
Metric review: The gimmick is pretty easy to figure out, you need to dispose of 1.33 quicksilver every single cycle so you use the provided lead to dispose of higher metals instead. That's it, that's the whole puzzle. Once you figure that out, its just arm geometry all the way down. I'm gonna have to give this metric a B- grade, as making a rate solve is pretty fun, and the gimmick is pretty unique, but after you get min rate the secondary isn't very exciting to optimize. For me, the best kind of puzzle is one where the min primary is just barely possible to reach so you feel acomplished when you get to it (like during thermite tape cost). If the primary is impossible to perfect (like a sum puzzle) or too easy (like non-critellium rate) then you don't get the satisfaction from solving it.
Solve explanation: Ima keep it real with you chief, I have no idea what I was cooking when I made this. The ideal best way to dissasemble an input atom like this would be to move it onto a debonder while grabbing the air, then pivot it twice, grabbing 1 quilver each rotation. Of course this is rate we're talking about, so you can't do that efficiently without a trillion pipelines. The next best thing is to rotate it using a hexarm and have 3 debonders, grabbing 1 quicksilver after each hexarm rotation. This approach of course introduces 2 major problems: 1) you need to duplicate the setup twice, as each hexarm handles half the inputs and 2) the quilvers need to be grabbed from 3 very different spots, and all moved onto the same projection. If you were smart, you'd use a different setup instead of a hexarm, but I am well uhhhhhhh this solve is good enough for me. Cool things this solve has that are worth mentioning: Look at those trackloops going around the hexarms, that's so silly!!! Those arms are having so much fun going round and round. Another cool thing is the one lonely arm 2 right of the output whose only job is to grab 1 air to descync the 2 sides of the machine more so they don't output at the same time. You could probably get rid of it if you reprogrammed the machine in a clever way but that caused lead to be grabbed at the same time and I don't wanna think about it I don't wanna think about it there's too much going on in this machine already and because its mirrored editing it takes twice as much time. Cost is a tertiary so its not that big of a deal, not like you're gonna lose a placement over it. One last thing is the atrocious metal management on the right. The lead gets grabbed and moved by a 3 long arm, that's a good start, and then after being promoted it gets grabbed by 3 different arms to get to the output, so that's a pretty big waste of area. Maybe i shouldn't have mentioned this to you, so this puzzle would look more efficient and less flawed? Doesn't really matter that much, this is all for fun.
It's getting late for me, so I'm ending the notes here before I write down something stupid. Good luck everyone!