Shapez 2 - MAM (Make Anything Machine) - Phase 1 (4 Layers with Colours)
ฝัง
- เผยแพร่เมื่อ 19 ก.ย. 2024
- Phase 3 (4 Layers and Crystals) out now... • Shapez 2 - MAM (Make A...
Make Anything Machine - 4 Layers with Colours (No Pins or Crystals)
If you like this content please subscribe.
I just uploaded the next phase, which is a 4 Layer MAM with Crystals... th-cam.com/video/2pNIZ8Jg33o/w-d-xo.html
Blueprint for Phase 3 is in the Shapez 2 Discord.
Great work. I couldn't really wrap my head around what the game meant by make anything machine in the tutorial. Having seen your solution, I'm not quite sure I am cut out to build something similar. I might have to go back to something simpler like Crusader Kings :D Thank you for sharing this!
Ok so I might not empathise with someone who doesnt have CS background but let me tell you one thing. You have gone very far in order to make very clear and easily understandable design to which I applaud you.
Very cool build! I could never comprehend building something so advanced lol
Thank you very much. :D
I'm still watching the video. And here you're talking about the issues with paint waste.
Just a thought, not having looked into it myself yet. Can't you paint the whole incoming shape before splitting them in 4? That way you have very little input waste and it should also speed up the general painting process as you paint 4 corners in 1 operation instead of 1 corner per operation.
This was done in Phase 2 :D
You've got way way more space pipes than you need. I first use a blueprint to gather 3 fully extended fluid miners onto all 12 belts (which, come to think of it, might not be necessary), then gather FOUR of those into one space pipe. In other words, one space pipe can handle the output from 48 fluid world squares. Yes, they have that much throughput.
Thanks for posting this. I've seen reference to a make anything factory in game and wasnt sure how much of a pain that would be mechanically. This was nice and modular to be able to understand it.
These are a fantastic way to make a MAM, well done!
Thanks for this, a worthy MAM, and helpful for learners. Was gonna suggest putting fluid in tanks but if you've pivoted to pre-painting whole shapes maybe you don't need it.
you've heard of a MAM, now prepare for a MEM
meeem? (TF2 Meme)
This is amazing. I played this game for 8 hours this is just beautiful to see. I'm completely impressed. I'm curious, what is your background? Have you just spent a lot of time playing factory games? Do you just enjoy the shapez franchise? Are you an engineer by trade? Thanks for sharing this - as someone that enjoys logic and things of that nature this game is super cool to me and I just don't know that I want to take the time to build something like this.. however, I'm super happy to see it made... cheers!
I'm a professional Software Engineer. I've played pretty much every factory game, and most of the main mods in Factorio, including the dreaded Space Exploration. (Don't even ask me how much time I spent on that mod...)
Why not paint the bits before splitting them?
You are a genius. :D I will definitely do this... should fix all my colour throughput issues completely. :D
@@HairymuppetGaming Glad I could help! Edit: Incredible work, by the way, I'd love to poke around this as a blueprint once you get it to a point you feel good sharing it.
I was coming here to say the same thing! It means selecting from 32 input shapes instead of 4, but it completely eliminates the complexity of the coloring that led to the inefficiencies. I bet that makes up for the extra distance needed on the shape selector part of the machine.
Yes, that was also my idea when watching the last video. My idea was to deliver them one train per shape but with one wagon for every color. That should solve the bottleneck.
@@andrewholaway4113 Actually what I'm planning is still having the same 4 shape inputs, and just colouring them (in the MAM) before cutting them. Theoretically it means you only need one belt throughput for each colouring block, to obtain 4 belts total throughput for that quarter.
How did i never notice i could send 3 layers over space belts? I have 45 hours playtime...
I like how organized your MAM is. Im just trying to cramp everything together as much as possible to make a very compact MAM. Maybe i should reconsider.
Couldn't you color the shapes before cutting? You are already using one belt per quadrant layer, so if you paint just before cutting...
Yup, made this improvement today, but the time I get to lvl 50 and start doing pins the MAM will have this. :D
The only thing I can think of that might cause an issue is if a layer has more than two different color crystals, and you could theoretically make crystal quadrants and add it afterwards. But the stacker is the only issue I can think of for that matter, as having 3 crystal colors on layer 2+ would mean a swap would be needed.
You could do this by assembling each quarter separately, all the way up to level 4, and then using swappers to combine them at the end. This would allow for any combination of crystals in any quadrant, but would fail for shapes that have empty gaps below the top level. (Though according to shapez2.wiki.gg/wiki/Shapes#Pins, the Vortex will not differentiate between empty quadrants and pins when accepting shapes, so maybe you just put pins everywhere)
The biggest problem I could see is if you have a layer with a pin/empty being shared with a crystal. Hopefully this will never happen, but if it does, then some more extreme swapping will probably be required...
@@HairymuppetGaming I'm just starting my own MAM and have been trying to take that into account. I was thinking a setup where it would read any non shape part, and if it doesn't read as a crystal it could throw a pin into that slot. Swapping could do to put the crystals on the upper layers, but to do that you'd need a factory dedicated to just making crystals to swap in
@@Funnyvid16 Sounds like a good approach. My current idea for crystals is to just do the crystal right at the end... so build up the first layer without crystals, once it's finished add the crystal bits for that layer (if required) then stack the 2nd layer (without crystals) on top of it. If the 2nd layer needs crystals, then you apply the 2nd layer crystals to the entire 1st and 2nd layer stack. As long as there are no crystals on top of empty spaces/pins? I think it will work, unless of course there are different crystal colours on the same level, but I'll wait and see if that is an actual level requirement before worrying about it too much... *fingers crossed*
24:25 Is there a particular reason to not just use one belt with the 3 layers for full throughput? Or just for visual? btw amazing design! so tidy
Cool man... i made a mam 4 layers seeing you video, it woks! than i just made a second mam paintig the shaps befor cut it... what a pain in the ass to make it work, but is working!!.. i think that demand the same time to make the shaps but take much less paint... thanks for the tips
I'm a little confused about something. Your system looks at the desired shape and says "Oh I need a circle bit in this spot, send some circles to get cut into 4ths, rotated, and then painted." All the ways you are sending four belts from one area to the next with all your little quarter circles. Why not paint, then cut, and finally rotate? Seems like it would save paint as you mentioned that can be a hold up if you get unlucky with how many things need to be painted.
Interesting setup. Was thinking about ways to build a MAM and thought best would be having all full shapes pre painted in every color. It would increase throughput a lot but would need a lot of platforms to sort them. Could maybe send trains that have one shape in every color on each wagon. Then pick which are needed and process.
But guess that approach gets very complicated and would need a ton of platform points to set up. Guess to get there you would need a setup like yours to generate those platform points first.
Also wonder why you don't let the train jump into the void directly?
thank you for posting this! im about to struggle through my first MAM. im going to try for a full space belt, so 12 full belts. I did notice one thing, youve got one belt going into one cutter array, giving you 4 times the output. this actually isnt helping you, your gaining 4 times the output by cutting the shape twice and losing 4 times the output feeding it through one array. you could for example, only take in a single belt of input and split it into into all 4 of your arrays and produce 4 belts of output.
This is pretty cool, would you consider sharing a blueprint for us to tinker around with?
I found the Shapez2 MAM much easier to make than the first game! I didn't need a tutorial or the like this time ;)
Some good ideas from your vid though.
I would recommend merging all 12 pipes at the colour offloader station so they unload from the train evenly
How come you dont just throw the trains straight into the vortex?
Nice. I'm excited to build my own in the future.
that was crazy try to get 36 per tick 12 belts on 3 levels!!!!
That's a _very_ cool setup!
Is there a reason that you unload the train instead of just jumping it into the center of the vortex to deliver the parts? I thought that was basically an instant "unload" but wondering if I missed something. Thanks!
It is a great diagnostic tool. I can look directly at the hub and, if any of the belts going into the hub suddenly empty out/have gaps, I know there is an issue with either the factory producing that shape, or the train transporting it. Currently there's some weird behaviour with Early Access, so it happens even if my factory is perfect, but I have discovered a bunch of train throughput issues with this diagnosis, which is very nice. :D
Im playing this game and loving every second. But you have blown my mind :)
How come you don't just launch train into void?
you can input all three floors into the vortex
The throughput isn't high enough to need all 3 layers so this way you have the aesthetics of seeing all the shapes flying in.
@@dopeslope7 This right here! :D
Beyond impressive, I'm still strugling with 2 layers 😆
I uploaded a blueprint onto the Shapez 2 Discord server, in the blueprints channel search for Phase 1 MAM by hairymuppet
That's awesome, thanks for crafting this.
what's the building that let's you read shape layers layer by layer? i can't find how to seperate the signals
The unstacker, under the simulated buildings.
Great job. Impressive that you're already at this point with the game being so new. These are the first MAMs and there is so many things to improve to speed up building and delivery. I'm a little disappointed to learn that you have to claim each set of items. If I remember correctly, in the first game it would change as soon as a piece was completed allowing you to turn it on an walk away.
One small tip (I'm not trying to be a dick). You need to work on getting your mic set up correctly. The coughing is fine and I understand you wanted to get this out because you are one of the first to make a MAM. I'm talking about the mouth sounds. It drives me nuts. Pull the mic back a bit or adjust the mic threshold , or get pop filter or windscreen.
This is the downside of doing a 50 minute commentary while wearing braces, I suspect. I'll move the mic a bit further away for the next video, see how it goes...
Hey! This is cool!
But I have questions. In SHPAEZ 1 I made only one input for colour and with signal chose which colour to enter. But in your case there a separate platform for each colour.
Isn’t more efficient to chose which color to bring?
There is no fluid filter in Shapez 2... yet...
Please share the world save or blueprint for us to tinker, and thanks for the pins theory
I'm trying to build it myself without using your blueprints. Having a little trouble with the stack logic. I got the shape and color section working, once I hit the stackers I'm having trouble dumping and routeing. Could you go into some detail on that section in the next video please.
I just finished casting the next video before I saw your comment. There is a short explanation in the next video, around the 20 minute mark.
Basically, say you are merging A and B. If B == null, send A directly to the output of the platform. If B != null, send A to the stacker. To check if something is null, use the Equals Logic Gate, and pass in the shape on one side, and a Signal Producer with "null" as it's configuration in the other side.
Awesome
That looks very cool!
Could you share the save so I can look at it too please?
One of the problems that I can see (basic shapez player) is that you only have one set of four coming out and splitting it into three. That means you are only running at one third efficiency. I dont understand where the rest of the throughput is? Shouldnt there be three outputs? Would this make the MAM a lot bigger, you would have to triple what its doing? Or look and see where the inefficency in your system is? Ive never made one of these myself yet.
Not 100% sure what you are referring to... but if you're talking about right at the end, I split it into 3 lanes to put it on all three layers of the train carriage for throughput reasons. All 3 carriages are then fed into the hub at the other end of the line.
Im working on a MAM soon and my thoughts were to have a stock pile of every shape of every color and use the code to filter through it. Or would that be more difficult?
More difficult, the logic you use to open/close flow of your pre coloured shape will be almost same to guide specific shape to a color station, plus pre coloring all shapes will take up more area and may complicate abit
@@avada_kedavra4484 makes sense thanks for the feedback. Iv never made a MAM before, never tried in shapez1. I'll try and start small and expand as I go
Cool
Love your name.
discord chanel?
this is like watching a chinese movie without speaking chinese...... you will undertand nothing.
Hi, sorry I'm not watching 50 minutes. How do you controll the flow of paint? Or basically how do you pick which paint is being selected for the final step?
Go to 14:38
@@hampst0rхехе 1488 хехе
I have a separate block for each colour, and I only route the blank shape through the required block.
@@HairymuppetGaming ah I get it, all painters are fully loaded, you don't control the flow of paint, you redirect the shapes to the correct painter platform loaded with the correct platform