Come all ye self-proclaimed alchemists, let's study the ways of old and share the fruits of our labour. What have you managed to create? How efficient are your contraptions? Care to lend a hand to those in need? Let's get down to business and work our way through the latest Zachtronics puzzler.
I'm currently consumed by this game. I designed solutions to puzzles in head when I should have been sleeping. I'm currently just going through the story, but biasing for speed in my solutions. I'll go back and do cheapest / smallest area solutions later.
Boy have I stabilized some water! 50:7 / cost:area in one solution and 20 cycles in another.
Unicycle, I did a double-take with your lead-gold machine, it's the 180 degree rotation of mine. Are you left-handed?
I'll probably grab this today so I can get in on the alchemical competition with y'all.
Let's see if linking through google photos works:
Refined Gold:
Waterproof Sealant:
Sword Alloy
Lead separation:
Some things to point out:
a) like all Zachtronics games, the graphs show your best over all puzzles, so you can optimise for individual things. I'm clearly burning money and area to get speed here.
b) There's a speedup you can get if you unwrap the loops rather than having repeats. That's what I had to do to reach polite shouting distance of MooQuack's score on the refined gold puzzle
c) I tend to unwrap my own moves instead of using the C command
d) shift scroll does a horizontal scroll of the command track
e) you can have lots of grabbers, clearly
f) I find tracks longer than 2-3 spaces really expensive in time.
Can't wait to get this game - but too many games on the need to buy soon list. Not sure how I should prioritize some of them. Loved SpaceChem
When working out a low-cycle-count solution, the limiting factor is how quickly you can pull out reagents. Everything else is just getting items where they need to be as quickly as possible.
The magic number, most of the time, is four: This is the number of cycles it takes to grab something, rotate it once, then cycle back to the start. Meanwhile, another magic number is two: This is how often, in cycles, you can grab a reagent off a source. Thus, two arms may keep a single reagent source fully occupied. Your job with the rest of the solution is to keep them fully occupied, by making a solution which is capable of utilizing those elements quickly enough.
I think a good example of this is my 28-cycle Precision Machine Oil solution:
Notice how the tracks in that solution are performing precisely the same action, just offset in time from one another. This is a technique I use often: A single track can't operate quickly enough to keep up with the elements that are coming in, so I add a second (or even a third) one, doing exactly the same thing, just operating with slightly different timing. Something has to be there to pick up that element when it's time for it to move. It just has to be there in time.
Another important technique is watching the solution step-by-step, and verifying that there aren't any wasted cycles. Can something be moved on the same cycle that something else is happening? Sometimes you can squeeze out an extra cycle by kicking off an arm just slightly earlier. This kind of detail work isn't always obvious, and can require just staring at the solution for a while, looking for things that aren't moving when they should.
I can't see your solution there Mooquack. Your 4 / 2 -cycle concept is awesome.
The other thing I'd say looking at your propellant solution Unicycle is that I tend to not worry about making stuff into salt too early (unless I have an animasus thing to feed which requires salt). I just put the elemtntal atoms in the right place. Since you are always moving the completed molecule around anyway, there are always places you can put the calcification glyphs. Your machine would be quicker if it just fed fire atoms, something else bonded them and moved them around so they bond in the right configuration and then something else pulled it out of the way past some glyphs. That avoids the long salt tracks.
I can't see your solution there Mooquack. Your 4 / 2 -cycle concept is awesome.
Here it is on imgur. I'm not sure if we can embed those.
You can, indeed, embed imgur images. Normally you just add a 'gif' to the end of a image URL to get the direct bitstream instead of the HTML container, but it wasn't working in this case. I had to figure out what the actual link was by downloading it, using the download link info to reconstruct the image's actual HTML link, and then adding a gif to THAT. It's not normally that difficult. Normally, you just add gif. (or, for static images, png or jpg.)
Result:
I'd appreciate it if people could spoiler their machine gifs and specific advice, maybe with the name of the challenge showing, so we could decide whether to look at them or not.
Malor, I was able to click on the imgur image, then select "Original GIF link" and use that with the img tag here. Results in spoiler below.
I picked this one to compare to yours and Unicycle's solutions, because I was somewhat optimizing for space and parts, and it shows how different solutions can be when they are balanced differently. I like that I only needed one bonder to get 3 elements joined in a row. I think I might be able to get it more efficient, but my head hurts a bit now.
Boy have I stabilized some water! 50:7 / cost:area in one solution and 20 cycles in another.
Thanks to working overtime in my dreams, I've gotten Prologue - Stabilized Water down to 40g - 15cy - 7a. I think that might be as low as it can go.
I really want to take a quick look - just the tiniest of peeks - at the next puzzle but that would be really dangerous so close to bed time.
Thanks Unicycle, it's frustrating to have stuff handed to me on a platter lol.
I've learned from this game, more than Zach's others, that my early training in APL and Assembler has strongly biased me towards reuse of parts, reduced cost and simplicity over speed. It's always good to get insight into *why* you think the way you do, and the bias became starkly clear to me looking at other people's solutions.
I have to force myself to optimize for speed, that's the hardest one for me because it removes the nice, neat, tight solutions.
With Shenzhen and TIS-100 I strongly favoured readability over speed, but with the visual ones I find it easier to ignore my years of writing boring, stupid, clear code and hack like a new maths graduate.
I will go back and optimise for cost and area (I think I only have 4-5 puzzles left to solve, though they will be bastards). And then look at the internet's solutions and do it all again when I learn the black arts they're using to be best in the world at it.
Glad someone made a thread for this! I need some friends playing this that aren't MightyMooquack, whose scores make me feel inadequate. I'll send out some requests to people in the thread, feel free to send any my way if I miss you. If I do miss you, it's because I'm lazy and no comment on your character, I promise!
Spoilers for Surrender Flare:
Despite this being a fairly awful design in terms of score - especially cycles - I kinda like how it turned out. It's majestic, in its own way! I get the feeling I'm using the 6 arm wheel fairly incorrectly here.
Edit: I just realised I almost certainly don't need to strip all the salt off that metal, just enough to fit around the Glyph of Projection. Ahh well.
Glad someone made a thread for this! I need some friends playing this that aren't MightyMooquack, whose scores make me feel inadequate.
What can I say? I have some very competitive people on my friends list, and I get the most fun out of these games from beating or at least tying their scores. I did a first pass on a lot of the puzzles in the game, just solving the puzzles any which-way, and now I'm going back and methodically topping the list on everything from the beginning. I actually haven't even finished the game yet, since I've been so engrossed with optimizing the earlier puzzles.
At this point I'm on the top of my friends list (tied or not) for everything from the prologue through the end of chapter 2.
Thanks Unicycle!
Here's a few more of my favourites so far, spoilers for Rocket Propellant, Mist of Incapacitation & Health Tonic.
I definitely try to keep things compact where I can.
Redwing wrote:Glad someone made a thread for this! I need some friends playing this that aren't MightyMooquack, whose scores make me feel inadequate.
What can I say? I have some very competitive people on my friends list, and I get the most fun out of these games from beating or at least tying their scores. I did a first pass on a lot of the puzzles in the game, just solving the puzzles any which-way, and now I'm going back and methodically topping the list on everything from the beginning. I actually haven't even finished the game yet, since I've been so engrossed with optimizing the earlier puzzles.
At this point I'm on the top of my friends list (tied or not) for everything from the prologue through the end of chapter 2.
I'll probably do another pass myself once I'm done. I seem to remember having a better score that you in a select few categories on a couple of puzzles, but you've since blitzed me. Some of your scores are frankly incredible.
I spent something like 4 hours on a single puzzle in section I yesterday, building it one way, then tearing it down and rebuilding with some other central focus. I'm almost done that section but taking it slowly.
Pages