Putting it all together - Shapez 2 MAM Build EP13
ฝัง
- เผยแพร่เมื่อ 18 ก.ย. 2024
- In this episode of our Shapez 2 MAM build, we are finally done with our shape staging so we put it all together now and build our 4 layer MAM. There's plenty of challenges ahead with stacking so let's figure out how to stack and crsytal shapes in the right order
FULL PLAYLIST: • Shapez 2 - MaM Build
BLUEPRINTS: drive.google.c...
![วันเกิดเธอ (TO YOU) - PHUWIN [ OFFICIAL MV ]](http://i.ytimg.com/vi/_ROx1Bj3T0Y/mqdefault.jpg)
it would help to decrease travel time by putting each layer in a plus pattern towards a center point that combine all of the shapes and all you have to do is adjust your trains to do a loop instead of going back to the underside and all the way back the same way
omg I thought about that! I actually mention it in episode 18 at the end of the optimisation process
@@DataEngineerPlays "episode 18"? are you hiding another 5 episodes from us? :)
(46:07) When text I need to compare like that isn't obviously similar, or if I need to make sure that there is no tricks like "lI" "0O" , etc, I also use search to make sure its exactly same strings :)
Good idea with the crystal stacking logic on the bin check. I might try to make that work with my MAM. But for now, I've just removed the checks before crystallisation and only check at the end.
As for the AFK / idle game thing... yeah. I find myself just waiting 5-15 minutes for the next milestone award, then I click it to start the next shape, and the next, and the next. It almost feels sad that it's pretty much done now. Maybe if I wait for hours and hours, the generator will throw up something that my MAM can't handle. Maybe. :D
Your analogy about checking for Null is correct, as that part of the MaM isn't producing anything, so nothing is getting backed up. And since your MaM is running at a fixed speed (shapes pr. min) in it's entirety, nothing will ever being backed up. So to sum it up, any trash collection is not doing anything to produce a shape faster. The only part where it would help, is at the final phase, if you are loading shapes onto a train, because of the way loading different shapes to the same wagon works.
I came to this same realization while watching this. If you’re dumping a MaM into the vortex, the trash collection only adds the time it takes for pieces to travel across the trash collection platform.
Yes all the trash collection will be optimised in the next 4 episodes. Although I still retain the null check, I will build the trash collection into the platforms themselves so it doesn't cost any travel time
While watching I was figuring out "what's some other way to do a make anything machine" I was thinking vertical MAM! Though efficiency is halved from normal because it will require a checkerboard destroyer I think it can work!
Small thing I can't figure out without access to the game is how to deal with floating shapes. So I was thinking of a sacrificial quarter to go along the way with the 4 layer quarter I'm not sure if this is the best way or does it work.
'Hard mode, or something fun' 🤣🤣 Hex would be good for next series if your not Shapez-ed out!
The two pipes are also missing at 54:58 and 55:10.
Keep up the good work !
Your final shape check can be smaller. Gate Input: 0, Condition: Not Shape-wire, Output splices back into shape-wire. You're then flushing null and shape changes and not just null. (written at 23:30)
(written at 35:48) The schematic I wrote can be just spliced into the shape-wire feed. There won't be a conflict because the gate only will turn on if the gate's condition, the shape, is null.
hmm interesting I think you described the exact null check I build in a future episode
Plz do the hexagons too 👍🌟
Maybe other factory building games for next series? For example Factorio or Dyson Sphere program?
Also Satisfactory 1.0 released Sep 10
Love the series, don't forget to update the blueprints 😉
I'm also wondering what happens if there is a full layer of crystals? Might need to feed pins into the crystalizer too?
Thanks for reminding me. And you're right the MAM doesn't handle full crystal layers. After episode 18 when the MAM is fully optimised I might take all final suggestions and implement them in 1 more episode
@@DataEngineerPlays That actually sounds like it would be really fun to try different peoples ideas or suggestions and see if they help optimize! And thank you for what you do!
My fully automatic mam is working, 12 x180 belts, 4 layers, pins and crystals. The only problem wyth cristal is that take 15min more to swap beetwin 2 random operator shape because we have to stack layer by layer to presevre crystals. TO make this more efficiant i've créated some complex purge system and i work with quarter parts instead of full shape.
Let me know if you what to see that.
I think rotate + analysis + color better than rotate + cut
what if you analyse a pin? would it come out of the virtual painter with a signal or null?
@@DataEngineerPlays if pin, will get null. It could be add one equal and for check
@@DataEngineerPlays Opps, comment dissapeared, probably youtube don't like link to images :(
You can replace/switch any signal with Gates. I use them to replace Null with Color-u. Connecting to Gate color signal as input and as "Trigger signal", if there is something it wiil pass that gate, if not, another Gate, controlled by same color wire but with Not will pas signal generator with Color-u. (Like you did with button to replace shape with your own before transmitter, but automated, as its own input work as that button). And pins get colored and it works :)
Bypassing Painter if "EQ" to Pin would be more logical, but I already did it with color.
I’m still watching the video so maybe this came up, but what if an entire layer is a crystal? No shape will get created for that layer. I think some logic is needed to use the pins for the crystal segments instead of just leaving the space empty.
yes you're correct, there needs to be that logic
my MAM after adding the first crystal colour it checks if any other crystals on that level are a different colour then removes those corners and goes through another crystalliser (it does this step 3 times for each other corner) but if a layer is a whole crystal it sets the bottom right and top left corners to a circle to allow for a crystal to be built
(it also checks if crystals of different colours are touching and sets the anticlockwise side of the intersection to a circle)
it’s horribly inefficient but it (kinda) works
If i were to do multicolor crystals I'd re-order the operations of the 4 layers and use a 3/4 "mould"
Instead of each MAM section making a horizontal layer, each MAM section would make a vertical quarter. and a blank 3/4 mould would be used to do crystals. Then the 4 quarters could be pieced together. This would work for multicolor crystal layers as long as there's no blank
for crystal only, just send a pin instead of a shape right?
What do you think about new playthrought series on "Turing Complete" game? Its about making your own "computer"
That sounds like an awesome game. I'll look into it!