One week ago I couldn't even spell nuclear physicist and now I are one!

[ksleet]

One of the neat things that SpaceChem does is track your efficiency -- how many instructions you used to solve a puzzle and how many cycles it took to complete -- and then show you a bar graph that compares it against how many symbols and cycles other players took, letting you know the average time taken as well as the outliers in both directions.

Comments

[kpreid]

• If the task can reasonably be accomplished by a single waldo, then use both of them to do the same thing out of phase and double the output rate.
• If you can use one rotation instead of another square's movement to move a molecule into the output zone, do so; rotation takes 1 cycle but going out and back takes 2.
• You can place bonders on the input or output areas and make/break bonds before picking up or after dropping them. For example, if you have methane and need to split off a hydrogen as in “Settling into the Routine”, drop it so it spans the two output areas, break the bonds, and then perform output on each side.
• Bond instructions can be executed by either waldo and trigger all bonders. You can have one waldo be carrying a molecule and doing something else while the other waldo triggers a bonder the first is passing over.
• A waldo program that looks like a double loop is usually not efficient.
• When chaining reactors, consider whether the overall efficiency can be improved by changing where or in what orientation one reactor places its

[kpreid]

As to 7 symbols on Double Bonds, I think it must be something like this, because that has 7 symbols and if it were any fewer it wouldn't have the necessary functions (2 direction symbols, 1 input, 1 output, 1 grab at input, 1 drop at output, 1 add bond). This pattern creates a double bond, but I can't figure out a place to put the output symbol such that it doesn't output monatomic oxygen.

[ksleet]

That's clever, I like how it uses the Bond+ to avoid having to actually move one of the atoms into position. I do recall experimenting with something like that, but I ran into the same problem you did, of how to make the output only happen after the final product is assembled. You could have the other waldo do it, but then you're back to nine symbols again.

I've been doing the "drop the molecule so it spans both output volumes then debond it" thing -- that's very handy for dividing up atoms and sending them to different reactors, especially on the more complex stages. Having both waldos do the same thing, just out of phase, is one trick I hadn't tried. I'll bet that's how the most extreme times on those simple reactors were accomplished.

I wish you could save multiple solutions for a reactor in this game. It is a bit annoying that I can't have both my "fewest symbols" and "best time" around, since they're rarely the same solution. It would also be nice if the stat screen tracked your record bests instead of just the current results.

[kpreid]

Indeed. For that matter, it would be nice to see a report of how your solution compares to your previous solution.

I suspect that they may have omitted any import/export of solutions to discourage people copying other people's solutions into their own game and distorting the histograms. But saving multiple solutions wouldn't have that problem.