Doesn't break any records but it's cool looking https://imgur.com/a/0xZ5ACR 140/56/160

Woah, that's cool :D

@ShadowCluster: Amazing disassembly/reassembly you have there :D I did this kind of convoluted solution for some of the story puzzles, but didn't have the patience to do it on this one yet.

Saw several inefficencies with Penta's Area, so here's a 46 Hex, which is +3 from theoretical optimal. +1 from tracks, and 2 from general rotation. Probably unavoidable, but I have been wrong in the past. 45 seems very tight, and I haven't been able to get the juggling right to support it.

70/3328/46: https://imgur.com/hQIVkF1

Who knows, there may be a configuration that allows you to not require disassembling it.

Also those solutions makes me realise which puzzles are fairly easy to optimize and which puzzles are not: This was an easy one as it can be solved by only adding and removing connections in a smart way. At the opposite, my previous puzzle is probably a nighmare to optimize X)

@PentaPig: Awesome work you did there :) It's also cool to see all the solutions together in the same album.

Thanks. If you can script one arm to move precisely every 4 cycles adding the other two becomes trivial. I noticed some inefficiencies in the area solution. There is now a 55/802/49 in the album.

@PentaPig, these are some impressive solutions. The fact that the 3 arms on the same track do not smash into each other is nice

I made a mistake while estimating cost. It's 40/748/94 now. I added the gif to the album.

The 6. input can be grabbed on cycle 11. It needs to move across 3 sets of glyphs to the ouput, for a theoretical minimum of 16 cycles. This probably isn't possible, but 17 cycles certainly is. My scores:
Cost: 50/177/89
Cycles: 810/17/316
Area: 75/693/51
https://imgur.com/a/F0MPYVm
I didn't attempt sum optimization, because Lag Grenade's solution is probably optimal.

Ok, so here are my entries for Area (72) and Cost (55):

https://imgur.com/a/1Ml2KQX

Yep, 18 cycles beats mine quite soundly. I overlooked the possibility to set up 2 parallel tracks to increase throughput in this one.

Wonderful method for both of you, and I guess you are right about the maximum speed Stratege. Let's see if anyone is up for the size or the cost challenge now :)

Got 18 Cycles and here I'm pretty sure it doesn't get faster as every 2 cycles a reagent is taken from input and product put into output.
Total score is 660/18/296

https://imgur.com/a/NjDSWiN

220/31/133. I think that's the absolute minimum in cycles.

https://steamhost.cn/steamcommunity_com/sharedfiles/filedetails/?id=1575649371

@Lag Grenade: Cool solution, looks less like a factory than mine :D

70/63/105

https://imgur.com/q9YUvvm