I'll preface this with man you've helped me so much with my own mam building and when I get stuck I come straight here along with a couple of other you tubers building a mam. One of them in particular has a 1x1 tile painting array that only uses 5x6(7 if you include the bypass belt) squares in the tile that has two paint jumpers to feed the 4 paints. Two of the painters are connected so one jumper feeds the two painters. I'll keep watching as I am sure you'll come up with a solution.
"We can treat it like one whole". and continues to use different platforms) I changed two 1x1 to one 2x1 for trashing corners and halfs to fit "brains" in the middle and belts with bins and filters at sides. (24:33) Slightly smaller version of corner to whole shape -- stack it with rotated, and you already have half of the shape, and now stack it with itself but rotate one of them 180
21:51 Just make full shapes on each channel before transmitting the signal. It's easy to get back to a quarter, colored or not. Alternatively, you can reduce the footprint of this logic part by building it on 3 layers. Make a logic lasagna. :) 26:00 Pass them up, through the entire color-mixer instead. There should be enough space to pass one belt through by jumping all the way. You won't need a space-belt going around. The fact that you need it is a problem in the first place.
you should start using 1 single layer for your logic circuits and only going to 3 layers at the end/start connections if/when necessary, i have done a lot of testing with really big factories and performance is an issue with many logic gates/redundant connections and wires, specially if the 3 layers will use the same signal there is no need to calculate it 3 times; at 40:10 you can route the signal from the top down, you already have it in the color pickers
I do sometimes do that, just not very consistently. Sometimes I get lazy to think about it while I'm recording and just leave the duplicate redundant logic
omg i sat around for like 2 hours trying to figure out the on site trash logic that you did on your shape selector and it just ended up being that, i'm a goober
You could put the logic for the shape painters onto the paint mixing platforms and move it down and across since there seems to be some extra space on those. That should get it where you don't need 4 signal towers for the shape painters.
if you're still interested in "optimising" going forward (making shapes move faster through the mam) there is a super simple tip -> use 'jumpers' where-ever you can. there are 2 reasons, 1st jumpers move shapes faster than belts. you can do a fun race as a 'test', make a long line of jumpers next to a belt, add shapes at the same time and watch the difference. the 2nd reason, is that a 'line of jumpers' at any given time is 'holding' less shapes. when you switch the mam, there will be less 'waste' that needs to be cleared. see the beginning of this video, your shape selector, imagine jumpers going 'up a line', there would be hardly any 'old pins' to clear.
That's a fantastic idea. I already knew jumpers are faster than the equivalent length belt, but always thought of it as a small marginal time saver. But the concept that they "hold" less shapes... that's what makes the big difference!
@@DataEngineerPlays indeed. but also, if you consider the total 'length' of belts used throughout the mam, even that marginal time save can add up to something significant. this would only increase as your mam design moves away from space belts, and towards connected platforms
@@DataEngineerPlays with 4 spaces between a pair, you have room to belt in, jump, belt in, jump, belt in, jump, then it’s ramp down/up until you belt out.
5:05 - so; you see all the shapes jumping over there; right after the 2nd jump there's a column with the perfect space needed to place the belt with the circuit. then the lines for the choosen shape can go in a line. just imagine the logic belt moving 2 belts backwards, and it is replicable on the lower belts too.
Your logic chains to construct and deconstruct shapes could be simplified using a belt reader. They read the shape on the belt, which you can directly feed into a shape analyzer. Compare to your target shape with an equals or not equals and use it to control the belt filters
26:59 maybe if you moved the "paint or bypass" logic to the bottom right or top left of the platform you might be able to fit 4 lanes. (Tho that would be bad for your pc)
I feel you could easily get the signal for the painter from your color mixer. You have a lot of space there and you would even be able to do the quarter to full there.also, there may be an issue with your logic when you get to crystal as it cant be destroyed without breaking it. You should run a test with a missing quarter and a half crystal.
That's probably true... the signal is being made near the top of the painter too... so I could of probably brought the signal down from just above.. duh... still not thinking 'cohesively' enough
8:29 You hop onto the platform, and then... hop 2 more times instead of splicing into the wire immediately? 21:17 A fluid tank before the pipe gate will let the new color annihilate/drain the old color without needing to use it to trash it. 22:21 You can trim one branch off. Shape + 90 Shape Stacked. Stack that output with a 180 rotation of itself. (You don't need to Stack one half, STACK THE OTHER HALF, and then stack them together. Stack One Half, and then split the wire and rotate 180 degrees stacking that) 22:31 Nag nag nag... Ya Know, if you ran your pipes through multiple floors, you wouldn't need belt hoppers and would be able to consolidate belt runs for more contiguous wiring room. 27:56 Awww... now I feel bad about the nagging above. 30:39 Capacity Planning. You're using 3 saturated belts of shapes. That's 12 painters. That's 3 launchers worth of paint. Having 9 useless launchers and 9 unused belts clogging doorways prevents you from hopping a wire signal up a column or over a row. The ONE PLACE I won't yell at you for hopping...
Haha cool ideas as always. Now I've finished recording my MAM series can you please send me a BP of yours? I've been so curious to see your work. Email in my channel description
Beautiful design. One question: the paint mixing area and its delivery are 'in the way'. is there a compelling reason for the mixing area to be specifically there? would piping the mixed colors from outside the production area save transport time? are fluids faster through the space pipes than shapes on the belts? You blow my mind with your ability to do this 'live'! All my best. 🧠🤯
I'm not sure, what happens if you give a belt filter only a color signal? If that works, follow up... what happens when the color doesn't change but the shape changes?
its not something i checked but do you gain space by going to larger platforms? you could change that 4x4 of single platforms to 2x2 of 2x2 platforms which would still be symmetrical and open up space for anything you need.
Yes it definitely does. I did that for the paint mixer episode last ep, It was in the time lapse so you probably didn't notice but I started off with 4 1x1s and switched to 2 2x1s to alleviate space issues. For some reason I forget idea soon after the rest of the series
maybe im wrong, but wouldnt the shape selection be faster if you dragged a belt of each shape to the top row and choose the correct belt there? no rubbish bin needed. there's just enough space for it i think
![Lp. Сердце Вселенной #60 РОЖДЕНИЕ ЛОЛОЛОШКИ [Финал] • Майнкрафт](http://i.ytimg.com/vi/YoR0pAV9FVQ/mqdefault.jpg)
I'll preface this with man you've helped me so much with my own mam building and when I get stuck I come straight here along with a couple of other you tubers building a mam. One of them in particular has a 1x1 tile painting array that only uses 5x6(7 if you include the bypass belt) squares in the tile that has two paint jumpers to feed the 4 paints. Two of the painters are connected so one jumper feeds the two painters. I'll keep watching as I am sure you'll come up with a solution.
"We can treat it like one whole". and continues to use different platforms)
I changed two 1x1 to one 2x1 for trashing corners and halfs to fit "brains" in the middle and belts with bins and filters at sides.
(24:33) Slightly smaller version of corner to whole shape -- stack it with rotated, and you already have half of the shape, and now stack it with itself but rotate one of them 180
Oh man totally never occurred to me to change the platforms to 2x1!
21:51 Just make full shapes on each channel before transmitting the signal. It's easy to get back to a quarter, colored or not. Alternatively, you can reduce the footprint of this logic part by building it on 3 layers. Make a logic lasagna. :)
26:00 Pass them up, through the entire color-mixer instead. There should be enough space to pass one belt through by jumping all the way. You won't need a space-belt going around. The fact that you need it is a problem in the first place.
you should start using 1 single layer for your logic circuits and only going to 3 layers at the end/start connections if/when necessary, i have done a lot of testing with really big factories and performance is an issue with many logic gates/redundant connections and wires, specially if the 3 layers will use the same signal there is no need to calculate it 3 times; at 40:10 you can route the signal from the top down, you already have it in the color pickers
I do sometimes do that, just not very consistently. Sometimes I get lazy to think about it while I'm recording and just leave the duplicate redundant logic
Can't you make a full digital shape from a quarter shape, by just stacking it with itself once, then stacking the half shape with itself once again?
omg i sat around for like 2 hours trying to figure out the on site trash logic that you did on your shape selector and it just ended up being that, i'm a goober
You could put the logic for the shape painters onto the paint mixing platforms and move it down and across since there seems to be some extra space on those. That should get it where you don't need 4 signal towers for the shape painters.
yes that's an excellent idea. I'm still not thinking 'cohesively' enough... still thinking too 'modularly'
if you're still interested in "optimising" going forward (making shapes move faster through the mam) there is a super simple tip -> use 'jumpers' where-ever you can.
there are 2 reasons, 1st jumpers move shapes faster than belts. you can do a fun race as a 'test', make a long line of jumpers next to a belt, add shapes at the same time and watch the difference.
the 2nd reason, is that a 'line of jumpers' at any given time is 'holding' less shapes. when you switch the mam, there will be less 'waste' that needs to be cleared. see the beginning of this video, your shape selector, imagine jumpers going 'up a line', there would be hardly any 'old pins' to clear.
That's a fantastic idea. I already knew jumpers are faster than the equivalent length belt, but always thought of it as a small marginal time saver. But the concept that they "hold" less shapes... that's what makes the big difference!
@@DataEngineerPlays indeed. but also, if you consider the total 'length' of belts used throughout the mam, even that marginal time save can add up to something significant. this would only increase as your mam design moves away from space belts, and towards connected platforms
You can also jump 3 belts over each other in a single lane. Big space saver where you have the runway in the first place.
@@baphnie 3? I've only ever been able to get 2 to share a lane
@@DataEngineerPlays with 4 spaces between a pair, you have room to belt in, jump, belt in, jump, belt in, jump, then it’s ramp down/up until you belt out.
5:05 - so; you see all the shapes jumping over there; right after the 2nd jump there's a column with the perfect space needed to place the belt with the circuit. then the lines for the choosen shape can go in a line. just imagine the logic belt moving 2 belts backwards, and it is replicable on the lower belts too.
Your logic chains to construct and deconstruct shapes could be simplified using a belt reader. They read the shape on the belt, which you can directly feed into a shape analyzer. Compare to your target shape with an equals or not equals and use it to control the belt filters
I hopped to the comments to say this. Belt reader also adds a little wiring but it can be a big space saving for some comparisons.
26:59 maybe if you moved the "paint or bypass" logic to the bottom right or top left of the platform you might be able to fit 4 lanes. (Tho that would be bad for your pc)
I feel you could easily get the signal for the painter from your color mixer. You have a lot of space there and you would even be able to do the quarter to full there.also, there may be an issue with your logic when you get to crystal as it cant be destroyed without breaking it. You should run a test with a missing quarter and a half crystal.
That's probably true... the signal is being made near the top of the painter too... so I could of probably brought the signal down from just above.. duh... still not thinking 'cohesively' enough
8:29 You hop onto the platform, and then... hop 2 more times instead of splicing into the wire immediately?
21:17 A fluid tank before the pipe gate will let the new color annihilate/drain the old color without needing to use it to trash it.
22:21 You can trim one branch off. Shape + 90 Shape Stacked. Stack that output with a 180 rotation of itself. (You don't need to Stack one half, STACK THE OTHER HALF, and then stack them together. Stack One Half, and then split the wire and rotate 180 degrees stacking that)
22:31 Nag nag nag... Ya Know, if you ran your pipes through multiple floors, you wouldn't need belt hoppers and would be able to consolidate belt runs for more contiguous wiring room.
27:56 Awww... now I feel bad about the nagging above.
30:39 Capacity Planning. You're using 3 saturated belts of shapes. That's 12 painters. That's 3 launchers worth of paint. Having 9 useless launchers and 9 unused belts clogging doorways prevents you from hopping a wire signal up a column or over a row. The ONE PLACE I won't yell at you for hopping...
Haha cool ideas as always. Now I've finished recording my MAM series can you please send me a BP of yours? I've been so curious to see your work. Email in my channel description
Beautiful design.
One question: the paint mixing area and its delivery are 'in the way'. is there a compelling reason for the mixing area to be specifically there? would piping the mixed colors from outside the production area save transport time? are fluids faster through the space pipes than shapes on the belts?
You blow my mind with your ability to do this 'live'!
All my best.
🧠🤯
There's no compelling reason no, you could premix outside the MAM and bring in all 7 colors and make your MAM even smaller
I wonder if using fluid tanks to preload and buffer white paint after the second mixer would remove the lag with white painting??
24:16 could you only analyze for color instead of the full shape?
I'm not sure, what happens if you give a belt filter only a color signal? If that works, follow up... what happens when the color doesn't change but the shape changes?
Belt readers should help with checking shapes on the paint belts and reduce the amount of wiring needd
its not something i checked but do you gain space by going to larger platforms?
you could change that 4x4 of single platforms to 2x2 of 2x2 platforms which would still be symmetrical and open up space for anything you need.
Yes it definitely does. I did that for the paint mixer episode last ep, It was in the time lapse so you probably didn't notice but I started off with 4 1x1s and switched to 2 2x1s to alleviate space issues. For some reason I forget idea soon after the rest of the series
maybe im wrong, but wouldnt the shape selection be faster if you dragged a belt of each shape to the top row and choose the correct belt there? no rubbish bin needed. there's just enough space for it i think
Oops Almost skipped some episodes there.
Hello future me.
Hi past me