Which speed solution is better?

[ghost]

the one that is a few ms faster (but the timer shows the same amount of seconds), but a few commands longer or the one that is a few ms slower, but a few commands shorter?

[hingston]

I'd probably lean towards a few commands shorter just because it's a noticeable difference in the game. Anyone else have an opinion on this?

[ghost]

Sure, however continuously improving solution by a few ms can eventually shave off a whole second (that's basically what happened with defrag disordered solution). Also it's literally a SPEED solution, so even a small speed improvement matters in my opinion.

[NeunEinser]

The main problem I see with this is, how can you easily, and preferably automatically, tell the exact time in ms?
Also, I think all solutions would need to be reevaluated and show the exact time in the table.

[ghost]

there are a few commands that take 0.016s (for this game it's an indivisible time unit). the more of these you can insert at the start of the solution before the timer starts showing that it became slower - the faster your solution is.

[NeunEinser]

Ah, that's actually a clever way of measuring this.
I imagine, setting a mem value to some number would create that kind of delay?

[ghost]

yes, if myitem == myitem and telling to memory slot when it's empty also works.