Opus Magnum Cycle Optimization Tutorial
Vložit
- čas přidán 5. 01. 2019
- I've seen a few requests for this sort of video to exist. This video features a simple puzzle I created, and the progression of solutions from a 67 cycle initial solve down to the optimal of 15. In so doing, we cover reset cost, throughput, latency, regrabs, and most importantly track loops.
- Hry
Hey thank you for a really terrific rundown of this game's expected thought processes! I had just started and was feeling the burn of inefficiency, curious about how I should be thinking, and this put me on the right track. I loved that you graduated the thought processes linearly to show acceptable, valid methods, because by the time you arrive at your optimal state I really got it and also felt like I could start following these processes myself.
Excellent video! I never knew track loops were so powerful in cycle optimization.
Void.
My problem is I never have trouble understanding the solution or any explanations/tips but when I am left to my own devices, I struggle to generate concepts or strategies. This video helped me because it gives me a list of things to check
ik its been years but this video is GREAT. Really well spoken too
This is the guidance I needed to get track loops to work for me. I’ve been struggling with this for a while now and this made it click. Thanks for that!
My brain just exploded. Thanks for proving I am not smart enough to play this game in such an amazing way :D. I'll stick to my clever underoptimized for everything solution, and maybe will try to apply some of the things you taught here in the future!
Nah my ego wont accept that, i will learn and defeat by force like sekiro
@@peepfacts6546 ikr..I shall force the rusty gears in my brain to turn..and I shall prevail! hehe
Great video! Great flow, concise, and informative. Approaching the puzzles from a latency + throughput makes so much sense and helps a ton. Now, back to bothering all you hot shots in the discord with more questions
Outstanding video. Just got the game and was trying to establish a terminology for optimizing, then searched and found this perfect video. Thank you!
Amazing. Unbelievably cool.
Thanks a lot for uploading this very useful video man. I finally understood how to create loops with Tracks. You are sooo awesome.
it seems like there could be one more cycle imrovement - take hexagonal manipulators and save turn back action
Great video!
Thanks for this! I had zero clue how cycles were counted so this helped a lot. I understood the obvious basics (more individual input = more cycles), and I understood that more single input arms tended to have less cycles, but I wasn't entirely sure how it counted it all.
I'm still on chapter IV right now so still learning, but my early machines were really cheap and space efficient, but hella cycle heavy. That was sort of my thing for a while; cheap and small, but as some of the puzzles got more complex I started to realise I was gonna need more than one or two arms doing EVERYTHING unless I wanted to spend ages walking the machine through each individual action lol. I especially struggled with one last night when I tried to make everything compact and everything kept crashing into each other, but then I went back to the drawing board, used multiple arms on and experimented with track loops and suddenly it all fell into place. literally. So now I'm trying more of that, and most machines now are low to average cycles and small area, but expensive. :P Now I'm watching this video and understanding WHY my track loop worked so damn well, and I'm encouraged to try multiple arms on the loop.
I don't think I'll ever be able to fully grasp how to do the crazy small cycle builds for bigger puzzles, there's just too much going on in those for me to ever understand how you even begin to build them, but you've helped me a lot with the breaking down and doing one arm. Stair thing makes a lot of sense too. Gonna go try it out! Edit: Oh, just saw you have another vid! gonna check it
Really great explanation - one thing that could be clarified further is why the arm/task with longest cycle length determines the period ("throughput"). And then why you can repeat other arms/tasks as long as the total cycle length is less than the period. For example, @ 9:43, we can still repeat arm 2 once in the task pane and it would be an 8 length cycle (which is less than 9) so the throughput would be 2 products for every 9 cycles. That also clarifies why moving to a 12 cycle loop is better because it will produce 3 products for every 12 cycles. I find that general understanding was helpful for me as a new player and glossed over/assumed implicitly. But again, fantastic video.
I just discovered this game and found the ideas here quite useful. Thanks for sharing!
Really impressive, thank you.
You, sir, are a genius. I didn't even know you could copy arms, not to mention the holy staircase trick! :P
Thank you very much. The game does not explain much. This is awesome.
this was very helpful my dude! thanks!
excellent video :D thank you
This video is wonderful but discussing the variables in the equations you mention are very difficult to follow. Great video this definitely helped me a ton in this game!
Super helpful thanks!
I just started playing this game. I'm looking forward to getting my self-confidence dashed once I can view the percentile at the end.
Very good explanations like a DSA class
You just made me feel really dumb, bro. Reckon I needed that. thanks.
Thanks Obama, very cool
You're a godsend
A whole new dimension to Opus Magnum just opened :D
Thank you so much for doing this step by step process breakdown!
So the idea with cycle optimization is using 'forward momentum' in a way. Rather than taking time to 'reset' you can have multiple arms on a looping track continue the output.
.... But this is JUST for Cycle Optimization huh 🤔 I was attempting to design for all 3 qualities (heat, cycle, space) but i assume it's better to optimize just for one huh?.... Which I assume that Cycles is the one most focus on? 🤔
I kinda like the challenge of keeping all 3 aspects in mind for the design though. 🤔
I'm still confused a little, but I'm starting to get a hang of it, I guess :D
The staircase technique is genius
I think I figured out all these on my own except for the staircase trick. That one would have been a real time-saver when figuring out periodics.
I haven't even played this game (I player Codex of Alchemical Engineering tho) and I'm already fascinated
To me, I always focus on trying to keep the size of the actual device down, the more compact the better.
The way you went form solution to the next better solution was insane here.
Like no thinking how to rearrange things or commands. I guess you have fiddled a lot with this particular problem already.
I was like: "OK, that lloks optiomal" [dude improves it 4 more times]
Copying the arm and the coding is going to save SO much time
Been having issues coordinating multiple arms ... like konrad509, still a little confused, but got plenty of food for thought.
This is cycle optimized, but what about cost effective? or Space Effective.., do these things just not matter? - serious question
The leader boards have a category for each of those - this tutorial is focused specifically on cycle optimization.
The method I use myself (and I'm sure many others use it) is to develop multiple solutions to each puzzle, with each solution being optimized for cost, space, or cycles. After all, you can't maximize all three at the same time; they contradict each other.
Everyone else: yes yes, latency and throughput...
Me: You can freakin' copy lever arms and their INSTRUCTION?!
Same lol why didn't the game teach me that
Everything made sense, until seeing 8 arms moving in unison turned my brain to mush.
damn some people are built different
The puzzle solution is too clean. Most tasks involve single element input and turning it into a 4 or more element structure.
WITCHCRAFT!