Conveyor Tech

[microfarad]

The purpose of this thread, and my return to the forum, is to discover the mechanics of conveyor and assess its possible implications on PG tech.

Here are some discoveries:

Crossover
0q3109G13N*viafJZVgOK8y0JS.swhi49be871HV64RrQnAeyUz6BzeTMZ0ZLqu5EeVjyLEnV6NzpLFPvGzKqq8a9UMsWTTSZMPggKus97XwrGuU*LAHQw7*lIqtuQ01hT7tIVchnC*dTVhmlGyiPrFT4JVrJRh30Glr9IEfgApbQarTlX7GY5rXldb9ulH*hWKBQz4p*kT*Jz5RPutMKS6dfoMF.kc0GaYsPc3gwrQc6Y32CdnF139yFgwqOYE2BrDrotifHarbztj0tp3ngNm0tnWeWvKUpoGewmnqP3fUhZ2wXcmti226XWIksecObdZl1m6BYZV7Em7Z4rrk9H8SqFItWjOxDPiD4RDHd8SV6RLxr1o1W1z5HwIe3iMOjyyTsngGYaZVX73izY1Xm6oGjWYuweMEfpIllrppCcg.GfrQQi7NN8rcQgjEK9s476poU4IkzNBYH4J6R1XBMw5hbKL3oSXrXNb3eL4VB6TJtH6Dbq40eGsnbocPowIlXDPQOFu5t8XTORvvIUBqVwu52SejLKSR6rVHMsUDrb71fwSbZ*kekRSPZcRPkQVx.LQDdNxH44YTeNwvYI0qs2DJTSjxAKPDPVTxGTydpEbhF9yQcPeQH63yJTfK6f.cnR8DPa
Can we formally express these complex behaviors?
13dA2P*o1KK2grsPThzgvJZnaO7**4i4X4yo*RM7bkAvS33PQNjkNYXYGpkBCXaoWkvumiuaJ*Xe*ivCrfB1uD.n5CnX4HqZ8QmL1OH9YB85USHXKPhiGeWe7h00GZqs3ap0wcISPrBGrMQ1b2JJgcWYt0mAW84yp30Aw1XKvJ4AEeSY4jkl99SU4hCsJZu00irsG4ocpNhgV4pEL113ObmobBWQoy9*ZtD9griL9crPrniz*qUmoGkhdJE3mRwrZCv2QEOUvT1ILEPtJ*gE1RP*Sz9NBrzXO.vNHyOBo5FBdQqljV6YWxqJjPaxwHGDfUOuoZfqHmmuzZeStcQHw2dZ7SrW4h.k1WQO8Yx6hu3vpb8CES244aMnYscPr.IlcvLPVxq5yej9GJnpjEeN3gRcUCBE94*uQFxEJDQHW6eqFJ6G5MqT8o6i5GM4IOYo1NjahFxOcIYKA1wm048zDWvCmEfadLLHyR.F3PVJWA6VSTVxMOjsRyV46kWOLcKk6HVR*J.7S0dBmJ*lw.5.TKsF0ocdNwIH4nMdDptxkWYtlh8JbkwUeA8lbeO7oWinXlf5ulImuyByKlJOUgXLJEKbkm4Dvu*C.3RwtXnsOf*IHFbXbih1pOCKGRfSEA
Dots always go around the outside
uoIaxBXLubWTLdo.WbijMw*Z4Mw1onSTFoqvo78FliG4*wXCIBAaFxgrVju4b*mZk1quO2LaxHCeN2KCT7c1GZJnjIe8N0xc6B*GF9UCiQ70GDUaQ8ieusv20MPkNUjO4PgkzRBoU8I.oxD5DS5Y5usJOUcxFMIDSTMxLVtcsEMJ1vJdAyXg7OJRAw3pTot5EzoougVcRv6gtgUEzli3mBToDbDQQYVECfRu5hwacuZeQlwCGg81TEt1WkNjtcZRanWiTp1m.WOyIp0Q0kbBFAMwIi6I5aoaMlsIVjNEyg2BFqPl5nhY8LPJ5t9xI*zD1m.vTz*5.kc9Iz.jOuCWLSnIe2i9N4sjy7A02X3BgeLfI5IdzJvlR4GSm43WnTkFTHnohN2gkQhA29pk8C1Z5eoIOEYwF6RziFlFbDEJ6UNUQH8wGKwtTm848AFmT2x6ED.mNFViqS6WOFU4IUW0mxHnDL1PQTrl9MnP0KG5Z7FBQVsFGQ42T.spGScR7KIjKZPPjZ5n.mSzIZ4POcsKL25b8aLF7WykybOurdvaMqWeAhxC0k91DAs3ecDwIlvC1WywOYTcxn4gNYSETe5IzMNw2e7ILWpa6htIFfMEiwF3*tny0mprjfzwlHodfmYrGi4oBRPHG7
Test patterns for study
.1ML*RKd*IJkf1sCzZr3ZLtgOSAmlODkc.Ob0DoZ1BXXuUuBuTu5NnMmQhJZSYyPa.Erfz3gbPmzCcKD1bNSG1Yij*o4wt6ofy24W.ZOb2x3pg2lvlmYqB2JvfIxkXcDzcWakRcC68l0gRtaWU5fjwsOmScodhbo*A9ms8eiLRnZ9qc73nKAELgx8fHDdq6pqx35jvYPEgxLYvZneySzdTSyQpCKDxuYUmyKNoT8qX47eLFxgGzsdqF7gMVlzUhPk9ekNx9VXaGJDt8t7owwAM9BNmPZ0ubMif7abhcu6u*qQdaGGYMTT6aiGX8r5fy3McvrFaOfit1b0aP1AXfNlc4bIuKD5mWxMZb1O1fIjIF3l9UKL1nqynSwXFPR87WcVIPnhx5hBJOLURy1S1xvgkB9I5BecNCnqI4bEABDL4Lz2Cgv89YLZZ9BuIX2x7Lavc3T*0FTn0LJuU5mKfXgt6EAv6d4hUx4iQXWep1yawsoe3smwA28ilpAyLQWG*qN8E*XqyjuFHT*raL7gXLBnq5B0NmJCUkziqLwQ6cirBX7YfqsflIvl564eTD.cKcdtjyYJkVy5ZTFk*7VLPCA46grwhAWzYGD3S4WNlyQYvpU5rEqbHotgpP2*9QuAEULOu8IDc7Kyrk2ILvN6a6IQbBQCh3oyqWHpAXA7iEb6U0G6KY584385Wnv82XHVALdC5GHL7tj73t1HIlbRNTeMpiPhTynyL87lB5HScbPZryLVYbnLdNyO3bDLXrba604TkZxai2d7*Rhhi3FXfn2iD3jPo1CLabw6lXCFj0Gi.PSmX2uwKmOYy2flUI1uMcthPZ1oR2TH8JI21jo84V*5WYDu.obls6Jyz2brwV8Kf64e.VWYxjjlVVSYzFqrrvPEHYZzJAerjv4mnFs*dfQRs*SDNMBxm1lzPE0bp0SHWKQE2wN3v.hvDlxekts*nsNzizG*SAtpdDttDfUCB5QKMfMIAo7FpojiMnE9Z0LbH
Properties of horizontal junctions
sUjxsqz0rnQGQGyeHb7x6LygTXawKVCOZ7GdLh1hN*h2*aynQP4iBnd5dknHRsNbRs0FNtXJqauVOtWxIoIsVMd6Y.JUYV7GpK3cwMYwP5xsbIYIlNGVtQ2NLv2Z2nsLHsnZeqG*Bd9pdKHo2muLmKBajoRCwwlwfnhCWpAJ3XBDoclH*ZTSo7lj*b*5ejBpxYE5tVhHKCoTuVgvoQOq*.f52GvjLEYvv5cFm77JJKUVl37xb6rseYCKr1CWY9uMnUxX8SQzhF3n7SRKrnPzwLgCdpwamxEIZmAagohu9Tsy01jOA4VL7WjaA2zCTA9wE52EeH2oB2R.dH3QjKnNWmCbTkQa9ryhQs720kHp4JWduBFKQSjIpf*Z.BlBf3Rzw4SB36zNWjZaJMTewJlbttSKgHCyzPuAkIy*NuxAd1J7xECZnB.kylCVnDStc5lWcWGtfanhAtedcam3iX2EX5p*S6lLpDzB.6vzt4OXKN1je0O9Y5g7VnvYUJfA9RTyQiM9ZktL0zkPiMz0yLFDnnyyiNiKxVOYeKSKXMn87fP4Rs2WHpmjSN2SHrIq6zc0LmqUloL2CXCEWoKggtcddJJNQj2enrMUY.Iehypq8Flrk.veEdst3vhzQuQgfqgdO0thU*
In-congruent behaviors between vertical junctions and a potentially useful behavior
rl1mq51AJWG9Km2ZHVkpIsuh5NUfm5xrGH92hZMqrE94mK3BBQe8prAfewIKBC*YSapTgrxEleTwtcQMAuQ16vFLr7VzYFhBn4lzu6EXRLNjoKkYBkCZ6A.IBekwSWQCFdTws5y1c9.t9QcJ3VKUExdlJTt74L3mN1SzTL1v.6OsYfFIeizVb8FkegP6yrSYla9asceuLrnqvcfGpZRT.7XzpMXPaULltLHP.Nm5T4f9XJmk3*o70R0sDvGUQnae9cdVma63JhJNgxBpW.v.jQAFmyQddqkHqfXnYS4ucBsFk46vLoEb4JS*UZQQ3jOEj8jzSLFSajyE9wxZz8VYMmAq45tmOrS9OZyWFIkjIYpETAxJpxgwLlRkceYVk57Xz7.5PCTS4Af2wgmh9evqk.1GHs8Vc6cLl0OS.hCqPFbnrR7cA5*.T6sjhMcTcLx1vat2Af*bDBPFXUFQ7SfjoEntNHMhAVGtrYHVQG7tp5gmXGaruw5uqhNwxGHZadX*sGUApkGPMi0DLW4xWi2ykia5lAqju29AhhQjz86t1vdoauxeGawTW.sk8ye0qDCslbH.L4utN5qiDNx8EV653xpr.5ZTfDNhw6JTp4o1GjXHbvOrlyguYOO9lw8XuoyNhmk9OEmXuTfjKEn9UB34JlmrE9eTGzhTzMVhqK.tN7duBemS1j2HCPtM*BIHk3cdzCZHaE2D7TR1hE3a1B4LOFo0fXYe.fMUjYIeaWpq7pgwFuyg6zEdBPyM6xi.HpO82aR*xSm*HC9pzVHNtQZQw.GhdO03mGpsbUdmiS6iFFVWJG44PLs9Kn4vvpc9sxI*bqN9yssLVblPpstzvZA63Btq.TjSWAn4U5PdWmMvVxynw6QZSALLQ44My2
I thought conveyor was deterministic but one of the test patterns exhibits stochastic behavior, so this really changes my incipient theories on the matter.

Stochastic behavior:
HOS7JoOtGBf*z9gS40gK3PVzRumdQyZjv40dEww3YS4UGULzdTrvww1r5QY4mFDIuGW6QkBjr1IXPIA5JNu8UpDuZDTGqliXo2k*Z0FZ0JMly4FBs1z6xbFqwXPrPkzJAfwbphRvGu2ryhQHskkbFD8yahuKpZCYWe8Kfgf8Avm7MHOWYrm5hH2qcpISnxcgYqZSxs23KfOZvw07p*oA.R3x3rJJKzdb7SbNvUaBQDz7cQaZiVMkXxaViPtsvMpGg1mdJ3HxmG8tS3GJM6YSx37L.bNzfjZBwcdzpa6XGtVjCO7cBXrh657QBdNoSlZ4FWcn4cFDnnHRENvyeEcCT6wLYtSMXfbdGpmmcfcroEMmqbS.6mv6JmD.*VyZ2YDs0TkKXykeZo2HsZEZkvseRC550gLjNYXR4fbjDh4nkuT.QZVC8yCX5O.G8wkuVoQECzVuL*oMksCAVaKkLXcZO5NIbh7woZpCXpGgrdIkKqBYeZI6YcWBl2IwYW2IgKvUy9Ix5BpdcOghABoF0E02Dg*3W2VltAhPa1llj3EpEGN*IFCPmIj8*sVvoKFHbSxdsJ.HlHVDCE29*SQjrPeWuzTG53DuIBvE10twDPt6kAkAITtkOO5ZLytIOUdvFOluiVeELT9IsEQ4rbUBI6rypeD8qyCYwl4vVO.XXz

[Qwerty]

I haven't done much experimentation with these things. So the click choice determines which direction it moves? And it moves dots of solid blocks individually?

[droctagonapus]

I found an odd thing.

If you have a tube, 3 thick including the walls, and the walls are made of opposite types of converyor

1) The dot inside the tube moves 2 spaces per frame

2) Upon reaching the end of the tube, the dot will either continue to stick to one side of the tube or the other, and generally will continue to stick to that one each time it goes around. (the majority of dots will stay on the same half)

However, which half it is isn't determined by weather it is the clockwise or counterclockwise one, or which one is top or bottom.

Instead, it appears to depend on the length of the tube!

shaving or adding one pixel to each end of the tube appears to switch whether the dots stick to one side of the tube or the other.

I don't see an obvious way to alter the length of a tube, but its probably useful to know when creating tubes, so as to make sure the dots go the right way when it reaches the mouth.

Also, it might be possible that the reason that this happens is due to some internal state of the dot, in which case it might be possible to make it so that dots that took one path to get there would go one direction, and those that took another path might take a different direction, at the mouth? I don't know if this is the case or not, but it seems plausible.

/endlist

also, thought to try to make a laser splitter using it, but it doesn't seem to be working.

I found that if you put the two types of conveyor side to side, with one piece of metal on it, the metal will alternate between the two positions. I was able to use this to make a thing that should always reflect a laser in one of two directions, and never absorb it. However, when I had one of the outputs feed back into the thing, it doesn't produce a continuous stream of output. I would post a code but the get set isn't working for me (pressing get doesn't do anything, pressing set yields "Error" if the box is empty, doesn't do anything if there is space removed code in the box), probably because my Java is out of date.

Maybe it is because the thing is too small?

here is a diagram. c and C stand for different types of conveyor. If it doesn't work, try flipping them.
m stands for metal.
0 is empty space

00000000
0000m000
in->0m00
00000m00
00000cC0
00000000

This seems to work somewhat as a thing to reflect the laser randomly in one of two directions, (and it is very compact imo) but I haven't been able to use it to split a single beam in a way that is exploitable to create infinite laser.

It could probably be influenced by another dot of metal being carried by another conveyor, so it could probably be used as a conveyor to laser converter.

another good splitter (slightly bigger, but different input output angles, and possibly more elegant from certain points of view) is

cC000
m0000
m0000<---in
0m000
Cc000

I haven't got this one to actually produce more laser though.

/end laser based digression

with careful spacing of a single conveyor, it is possible to make a dot occasionally skip a square.

Creating a tube of both types of conveyors allows the dot to skip a square every time.

Using both types of conveyor, I have found a way to make a single dot jump up to 14 blocks at a time.
(!!!!!!!!!!!!!!)

the pattern is fairly simple. Just use a tube, and remove every other line from it.

so like

c0c0c0c0c0c0c0c0
0000000000000000
C0C0C0C0C0C0C0C0

The dot just blinks from one spot to 14 squares away! (and then continues at a somewhat slower rate after that point, but still faster than 2 squares per frame)

This is very fast!

I wonder if it could be used to accelerate S-Ball to nitro ignition speed.

(I previously attempted to accelerate the S-Ball with a simple tube, but it stuck to the ends, except in a certain situation I did not understand.)

edit:
wait I guess it doesn't matter that it doesn't work as a laser source because there is still laser-ball+glass.

[endy123]

Jul 22, 2014 19:10:20 GMT droctagonapus said:

I got a thing working where the first dot of metal that travels on the thing will get stuck, but the second one won't. This could be used to count, and along with an or gate could create an and gate.

However these gates aren't resettable.

(at least I haven't figured out how to make them resettable yet)

generally the thing works like

(my get and set are broken)
(where 0 is empty, c is conveyor)

000000000in0000
000ccccccccc000
000000000000000
00c000000000000

So thats a thing.

I was able to make it resettable. New dots at certain positions will "release" the stuck one.

Resetting this with a laser is interesting.