yes , my solutions tends to use many cycles, but are decent on cost and area. i have tried to optimisme for cycle efficiency, but I end up being bored and not liking the machines in the end

It's something I really like about this game and the other Zachtronics ones that there are virtually infinite correct solutions and it's completely up to the player what to strive for.

I find something immensely satisfying about optimising for speed, trying to get it so that things arrive "just in time", and parallelising as much as possible.
Optimising for space doesn't appeal to me anywhere near as much (I need an artificial constraint like in the bonus levels to force me to build small and even then I was still targeting speed with the silver paint machine). It find it really interesting to watch your solutions because the approach is so different.

I occasionally go back and tinker with some of these puzzles. Faster, smaller, cheaper; you start seeing optimizations that eluded you.

Unstable Compound Cheap literally the cheapest possible. There is nothing that can possibly be removed.
Curious Lipstick Fast it's actually kinda crazy how slow my old 153 cycle solution was by comparison.

That Curious Lipstick is crazy fast!

It's pretty close to the theoretical optimum. You need 18 salts to create each output, and for 6 outputs you need to draw 108 salt atoms. Two salt atoms are available on cycle 1, and presuming you grab them and remove them two more will be available on cycle 3, and so and so forth every odd cycle. This means you'll have 108 salt atoms drawn by the 107th cycle if you never miss a beat. From there it's a matter of getting them to the output in the right form in the fewest possible cycles. At very minimum it would take four more cycles (one to pick it up, one to move it and bond it with the existing structure, another to move it over the output area, another to drop it) so the theoretical optimum is 111 cycles.

When I finished this design I immediately saw my mistake; I use my last two salts to create a death atom, which requires an additional three cycles to drop the salts, pick up the death atom, and move the death atom. In order to do it faster I'd need to use both of the last two salts as the two salts in the final output. That is impossible with the layout I have here. I'm still pretty happy with 114 cycles, though.

Haha, that's pretty much how I started every level - by calculating the maximum theoretical speed. Then I'd get so hung up on trying to achieve that that I'd get stuck. Eventually I'd give up and just try and make a working solution often to find that I'd still be on the fastest actually populated column on the histogram which is good enough for me. I don't think I've ever come as close to the theoretical maximum as you have.