View Full Version : Custom Levels
12-03-2011, 06:11 AM
Ok, researchnet hasn't updated in awhile, so I'm getting some craving for some new levels. I imagine others are, too, so here are a couple of my own.
Haber-Bosch. Different than the one by Cake>Pie, you don't have to make the hydrogen yourself. You will have to figure out some way to sort without a sensor, though.
Water to Lithium. Pretty simple, but you'll need a fancy trick or two.
Breeder Reactor. Probably my magnum opus. It is solvable (solved in 1923/3/99), but you'll need to bring your A game.
12-03-2011, 10:07 AM
I've been meaning to browse through the forums for more levels. I'll check these out, and maybe post one or two of my own at some point.
EDIT: Okay, I solved "Breeder Reactor." Good work with that one. My initial solution was approx. 3K cycles, but a quick edit brought it down to 2K. I also have an idea for a 2-reactor solution.
12-03-2011, 08:56 PM
1492/2/28 on Haber-Bosch, without much optimization, and as a low-symbol solution. Sorting two different molecules by bonds is old hat to the vets by now.
1293/3/70 on Breeder Reactor, which was a really fun level. I had a 2k+ solution with 48 symbols until I changed how the sensor reactor works depending if it detects helium or hydrogen, which sped it up immensely. Still could use some optimization; under 1k is almost certainly possible. Oh yeah, and it's a good thing you only need 40 plutoniums instead of 47.
Water to Lithium I have no clue upon first glance. I'm terrible at fission.
12-03-2011, 09:41 PM
I've been toying around with amino acids and seeing if any would make for a good puzzle. A couple of them have caught my eye, but I've only made one level so far. I'm no good with this sort of level and I haven't figured out a solution yet, so I'd like some help: is this solvable?
Input α: Amino Radical (NH2)
Input β: Propargylene (C3H2)
Output: Tyrosine (C9H11NO3)
8 bonders, a fusion laser, and a sensor. If it can't be solved (or even if it can), some advice on how to improve this would be appreciated.
EDIT: Solved by GuavaMoment (and later myself).
12-04-2011, 12:25 AM
is this solvable?
A solution so hideous you'd think it came from a cake>pie level, but a solution nonetheless.
Edit: Argh I didn't want to optimize but just looking at the uploaded screenshot I figured out how to add a symbol and reduce the time to 1691. Let me sleep now SpaceChem, please.
Edit2: I hate everything. 1537/1/124 (http://cloud.steampowered.com/ugc/612718202457655191/4D50356194F06CD8B7CFA700424F6BD5794A6AAB/)
12-04-2011, 10:59 AM
I do feel a bit hypocritical with this, since I usually despise this sort of level. Either way, I went back and solved it myself: 1495/1/126 (http://cloud.steampowered.com/ugc/614970002291322924/247C80BAB94CC83BD4FC92104E3DF6839EB674CE/)
12-17-2011, 12:04 PM
Since there's been no real activity for nearly two weeks, another level for you all:
Input α: Lanthanum
Input β: Hydrogen
Output ψ: Mischmetal (La/Ce/Pr)
4 bonders and a fusion laser. My initial solutions were 1216/1/53 and 979/1/53; I'm pretty sure those can be beaten. Have at it!
12-22-2011, 09:35 PM
I solved Breeder Reactor, and in a way that would work with an infinite amount of plutonium needed, but it took almost 127,000 cycles.
Did it using binary counters. The first reactor inputs a uranium into alpha, splits it, and sends it out.
Reactor 2 detects hydrogen, and sends it to reactor 3. If it gets anything but hydrogen, it sends it back to reactor 1.
Reactor 1 continues inputting the atoms kicked back from reactor 2 and splits them. One of the waldos is on a binary flip-flop counter, and when it completes, it grabs another uranium and restarts the cycle.
Reactor 3 also uses a binary counter that counts up to 94 (and inputs and fuses with each count) and then outputs the Plutonium.
The pipeline that kicks back atoms from reactor 2 to 1 needs to be of sufficient length to not get backed up, but it completely flushes before a new uranium is taken, so there's no build up over time.
I'm curious what approaches everyone else used to do it in so few cycles, but I'm assuming it involves breaking down just 1 uranium into helium and then fusing them to uranium.
12-22-2011, 10:16 PM
My two-reactor solution was thus:
Reactor 1 is a sensor reactor. The red waldo starts with one uranium, then sends it to output ψ. If it detects hydrogen, it's moved to output ω. After two hydrogen atoms are sent, it signals the blue waldo to send a uranium atom.
Reactor 2 is a nuclear reactor. Each waldo works independently-- one splits atoms, then sends them back to reactor 1 to be picked up by the red waldo again. The other takes three atoms (two hydrogens and one uranium) and fuses it together to make plutonium.
There's just enough pipeline between reactors to reach capacity before it starts emptying faster than filling.
Either way it's a simple, but elegant level. (Also, 94-counter? Ye gods.)
12-23-2011, 01:56 AM
yeah, just a series of flip-flops arrayed in a line, with arrows that send the waldo to upper or lower return tracks which take it back to the beginning of the line. On one track, it hits a sync, and on the other track it doesn't. Each sync sends the other waldo around a loop (where it either splits/outputs or inputs/fuses). The up/down arrows along the flip-flop chain basically act as 1s or 0s in a binary number. When the counter completes, the waldo leaves the end of the chain and does it's thing (in this case, inputting a new uranium, or delivering a finished plutonium). This allows for waldos to alternate by an arbitrary number (blue does something 94 times, then red does something). Not worth using if you just need to do something a few times, but the amount you can count to grows exponentially for the amount of space it takes up. Unfortunately, it's very, very slow :/
Technically, my second reactor (the only sensor) probably isn't necessary. After breaking up uranium into hydrogens, the waldo could probably just cycle 92 more times around another counter to send all the hydrogen to the other output. It'd be tough to fit it in, and go even slower, but a 2-reactor, no-sensor solution would be pretty neat.
10-21-2012, 10:41 AM
It's been... what, 10 months? Here's another level, probably my favorite of all the ones I've come up with so far.
Input is water, output is malic acid (C4O5H6). 8 bonders, fusion/fission lasers, and quantum tunnel. My first solution is 1969/1/109; and I bet it can be done much better.
10-21-2012, 02:39 PM
Here's my Water into Wine solution:
11-02-2012, 05:16 PM
It seems that someone has collected all of Lanky's custom puzzles and uploaded them to Pastebin (http://pastebin.com/M3y8tFxd). I'll have to check those out at some point. (EDIT: solved "Back to Basics" and "Doubleplay". "The Overture" gives me bad flashbacks to building cytosine. "Two-fer" doesn't work properly, the input only gives C=O instead of a mix of C=O and O=C. "Impostor" is just... what.) In the meantime, here are a couple that I made recently. Relatively easy, but might as well share them anyway.
2 of Hearts. α is Antimony, β is Oxygen, output is a crystal of Valentinite. 8 bonders and a sensor, so it shouldn't be much trouble. My solution is 930/42.
Paternò–Büchi Reaction. α is Formaldehyde, β is Ethylene, output is Oxetane. Only 2 bonders. My solution is 526/67.
11-03-2012, 10:01 PM
"The Overture" gives me bad flashbacks to building cytosine.
Sweet, I've damaged someone psychologically! Although I don't think Cytosine was nearly as bad as "Getting Pumped" or "Barrel Marked B" (One that I made after researchnet stopped updating).
11-03-2012, 11:10 PM
Yeah, I'm aware. I downloaded your Breaking Bad levels a day or two ago; solved the first two ("This is Not Meth" ignored the nitric acid and involved a huge garbage pipe for hydrogen), but I haven't taken the time to figure out "Barrel Marked 'B'". The difference between those levels and "Life: Cytosine" is that I've actually beaten Cytosine. On a related note, upon beating "Squaric Acid" and werbad's "Vitamin B3" (not sure why I put that one off for so long), "Getting Pumped" is the only thing standing between me and a completed Journal. ...Yay?
11-12-2012, 10:41 AM
Uh, hi. I learned from cearn that there was some chat about my SpaceChem puzzles (username Lanky). I just wanted to tell you guys that I'm here and to give you an update.
I made another 2 pages worth of puzzles and uploaded the entire collection to my new Pastebin account. You can find it here: http://pastebin.com/eQQVjY4m
Some levels have been edited and even omitted from the collection.
So yeah, I'm Lanky. Add me on Steam if you want to chat about puzzles, since I'm still actively making them.
11-12-2012, 11:16 AM
Oh, neat. I'll have to check those out. (Also, I wish there was a way to rearrange your list of created/imported levels, so you could group them properly.)
11-14-2012, 05:50 PM
Okay, here are a few more that I've made in the past couple of weeks:
Benzene Backbone. Benzene and Butadiyne (C4H2) into a molecule of Chalcone (C15H12O). 4 bonders and a fusion laser. The main problem is moving around the large block; it's not too difficult once you figure it out. My solution is 675/66.
Doubling. Urea and Hydrazine into Biurea and Ammonia. 4 bonders. My solution is 862/101.
Antimony Ores. A production level this time, making Valentinite (ratio of Sb2O3) and Stibnite (Sb2S3). You can use up to 4 reactors, though my solution is 5849/2/101.
Let There Be Light. Formyl Cyanate (C2HNO2) and Acrylonitrile (C3H3N) into Luminol (C8H7N3O2). Erm. This... is probably the hardest level I've made thus far. I... haven't actually beaten this one yet. Some help (is it solvable or not, how could I improve it) would be appreciated.
11-23-2012, 09:20 PM
Here's a set I'm rather proud of:
Cinnamon. Pure CH into Cinnamaldehyde. My best solution is 431/59.
Headaches. A couple of C/H/O molecules into Aspirin. Solved in 541/64.
Riot Control. Benzene and pure C into CS Gas (C10H5N2Cl). Solved in 961/93.
If you're noticing a pattern with these molecules, here's one that doesn't fit it. Chemical Warfare. A production level-- turning lethal compounds into more lethal compounds! Solved in 2604/4/141.
I'm planning to compile a list of some of my better levels and see if I can get a few into the Journal, so critiques/recommendations would be appreciated.
11-26-2012, 10:58 PM
Boom (http://pastebin.com/nWFXzX00). Have fun.
^--Series of assignments that I made.
02-13-2013, 10:10 PM
Okay, I've been putting this off for long enough. 30 levels for your puzzle-solving pleasure (http://pastebin.com/xH4jTZgR), compiled into a single pastebin dump. (Some levels I've already posted on this thread, others have only been seen by one or two other people, some are completely new.) I hope you enjoy, and I welcome your feedback.
vBulletin® v3.8.7, Copyright ©2000-2017, vBulletin Solutions, Inc.