--- Log opened Wed Feb 21 00:00:05 2024
00:13 -!- Mabel is now known as Anachron
00:51 -!- darsie [~darsie@84-113-55-200.cable.dynamic.surfer.at] has joined #hplusroadmap
02:35 -!- EmmyNoether [~EmmyNoeth@yoke.ch0wn.org] has left #hplusroadmap []
02:45 -!- justanotheruser [~justanoth@gateway/tor-sasl/justanotheruser] has quit [Remote host closed the connection]
02:48 -!- justanotheruser [~justanoth@gateway/tor-sasl/justanotheruser] has joined #hplusroadmap
03:39 -!- millefy4 [~Millefeui@91-160-78-132.subs.proxad.net] has joined #hplusroadmap
03:41 -!- millefy [~Millefeui@91-160-78-132.subs.proxad.net] has quit [Ping timeout: 246 seconds]
03:41 -!- millefy4 is now known as millefy
05:07 < kanzure> greetings
05:08 < hprmbridge> Katylase> Hi! How are you?
05:09 < L29Ah> suffering antihistamine abstinence syndrome
05:10 < nsh> what's that?
05:30 -!- L29Ah [~L29Ah@wikipedia/L29Ah] has quit [Ping timeout: 260 seconds]
05:55 < hprmbridge> kanzure> jacquard loom: https://cdn.discordapp.com/attachments/1064664282450628710/1209860929462861964/1HUQ1GE6aczWT5Ctic7cQBg.png?ex=65e875cb&is=65d600cb&hm=079b3d3800f77c067d9d0b5af5d3d19e3d498d027389036580dccc8bfa98b5ef&
05:55 < hprmbridge> kanzure> von neumann universal constructor: https://cdn.discordapp.com/attachments/1064664282450628710/1209860959397609522/image.png?ex=65e875d2&is=65d600d2&hm=3bbb591487615f109355fec65cf12012a30895362e92ccdf34718a5cbe9e94be&
05:55 < kanzure> perhaps instead of electronics it would be possible to make a fully kinematic self-replicating reprap printer using human-powered mechanisms like hand-crank or bicycle + jacquard loom's jacquard mechanism or punched card tape for specifying the program
05:56 < kanzure> mechanical motion as power input can be abstracted away of course to non-human sources
05:58 < hprmbridge> kanzure> relay reprap diagram: https://cdn.discordapp.com/attachments/1064664282450628710/1209861725403226132/9256771464601539686.png?ex=65e87689&is=65d60189&hm=1d23dfc5a29bf45df2900a7c1a32aeae726a0a04d4293865e1de434a750ac3c7&
06:01 < kanzure> there is already existing literature on using mechanical mechanisms for computation and logic
06:01 < kanzure> such as merkle's rod logic or merkle's mechanical computing using only links and rotary joints https://arxiv.org/abs/1801.03534
06:03 < kanzure> i don't actually know what actually is the state of the art for compiling logic/software circuits to kinematic mechanical mechanism-based computation
06:08 < kanzure> the existence of punch card numeric control NC machines shows that at least xyz translational and rotational motion has been demonstrated mechanically before
06:08 < kanzure> a mechanically powered plastic filament spool feder does not seem unreasonable?
06:08 < kanzure> feeder.
06:40 -!- L29Ah [~L29Ah@wikipedia/L29Ah] has joined #hplusroadmap
06:48 <+gnusha> https://secure.diyhpl.us/cgit/diyhpluswiki/commit/?id=09148b24 Bryan Bishop: transcript: brian mennick >> http://diyhpl.us/diyhpluswiki/transcripts/reprap/brian-mennick-2021/
06:49 < kanzure> this isn't a strict transcript because i put in some more details https://diyhpl.us/wiki/transcripts/reprap/brian-mennick-2021/
06:49 < kanzure> here is minnick's manuscript https://diyhpl.us/~bryan/papers2/self-replication/A%20self-replicating%203d%20printer%20-%20Brian%20Minnick%20-%202021.pdf
06:54 < kanzure> nevermind, i didn't read fenn's comments
06:56 < kanzure> but yeah it is strange that much of reprap self-replication woe has been focused on electric motors
06:59 < fenn> the name is minnick not mennick
06:59 <+gnusha> https://secure.diyhpl.us/cgit/diyhpluswiki/commit/?id=5111fe2b Bryan Bishop: fix name >> http://diyhpl.us/diyhpluswiki/transcripts/reprap/brian-minnick-2021/
07:00 < kanzure> fine...
07:00 < fenn> i wonder why nobody else had tried sintering solder paste
07:01 < fenn> or if that's actually true
07:01 < fenn> you could also mix solder paste with copper powder to prevent it from glorping about with surface tension
07:02 < fenn> solder paste is just solder powder and liquid flux
07:04 < kanzure> updated non-broken link: https://diyhpl.us/wiki/transcripts/reprap/brian-minnick-2021/
07:05 < fenn> flux is some kinda organic acid solution "rosin" which conveniently also acts as a thickener, but i think any kind of acid will work. acetic acid evaporates completely so it might be cleaner than having sticky gunk everywhere
07:05 < L29Ah> > i wonder why nobody else had tried sintering solder paste
07:05 < L29Ah> for what?
07:05 < fenn> for printable conductive traces or coils
07:06 < L29Ah> there are homemade 3d printer heads with driven syringes of solder paste
07:07 < fenn> but do they sinter the printed paste instead of melting it?
07:07 < L29Ah> people don't scream about their great achievements so i guess they have a hard time fighting surface tension
07:08 < L29Ah> you need to make a die and press stuff to sinter it, don't you?
07:09 < kanzure> how do we still not have a compiler for mechanical computation
07:09 < kanzure> this is lame
07:10 < fenn> no that's hot press sintering. the regular kind is just time * temperature integral
07:15 < fenn> the Z1 computer was fully mechanical and much more compact than later relay and tube based computers
07:16 < fenn> iirc it used a combination of gears for multiplication, and packs of thin metal plates for binary memory
07:17 < fenn> i don't think we need a full turing complete computer just to copy a tape and move to coordinates at a constant rate
07:18 < fenn> i'm not even convinced the tape is necessary; you could broadcast on radio to a large number of replicators all operating in synchrony or at fixed offsets so they don't spike the power grid
07:19 < fenn> or data over power line, whether that's electrical or pneumatic or something else
07:21 < fenn> for the 12 bit whiffletree positioner you would probably want something that gradually shifts the input bits from 0 to 1 and vice versa, which implies either hydraulics or like two mechanical bars that physically push all the bits up and down at the same time, like on a loom
07:24 < fenn> it's kinda too bad they didn't figure all this stuff out back in the late 1800s when people knew how to do things
07:30 < L29Ah> fenn: off-the-shelf solder paste is problematic since a lot of it is volatile at your sintering temperature
07:31 < L29Ah> so either you sinter it in a saturated atmosphere or in a quite confined space so that your paste pattern won't deform too much due to liquid loss
07:32  * L29Ah wonders if a paste based on Rose's metal would be approrpiate for typical FDM-printed PCBs
07:34 < fenn> then your motors have a very low temperature limit
07:35 < fenn> anything that sinters will shrink by some amount
07:40 <+gnusha> https://secure.diyhpl.us/cgit/diyhpluswiki/commit/?id=5da2b3bd Bryan Bishop: fix formatting >> 
07:43 < kanzure> backlink.. just me ranting on twitter about these topics: https://twitter.com/kanzure/status/1760327875918483604
07:43 <+gnusha> https://secure.diyhpl.us/cgit/diyhpluswiki/commit/?id=49d7639d Bryan Bishop: backlink >> http://diyhpl.us/diyhpluswiki/transcripts/reprap/brian-minnick-2021/
07:47 < fenn> http://youtu.be/x_ijmjx7Xys the jacquard loom actually has this complication where the punch card pins don't lift the threads directly, they nudge the pins that lift the threads into or out of alignment with a "knife" that lifts the pins that lift the threads, so the punch cards never see any loads from the thread tension
07:49 < fenn> similarly, a whiffletree could use a weak solenoid pin to cause an input to either be driven or not driven by a bar that continuously moves up and down at a slow continuous speed
07:51 < fenn> the continuous speed is because all the inputs must shift at the same rate in order to not make sudden jumps, because the bits will not shift at exactly the same time. consider rolling over a large number of bits at a "round" number like 011111 -> 100000
07:52 < fenn> that first digit has to transition from 0 to 1 at the same rate as all the 1s transition to 0, because the position of the axis is the sum of all their values multiplied by the lever ratios
07:53 < fenn> the first bit is literally half of the axis travel
07:54 < fenn> if you just went from 000000 -> 100000 it would break from the massive acceleration
07:54 < fenn> maybe there is some way to do this with a gray code so it can't ever get bad inputs, but i'm not that smart
08:04 < fenn> also it's nice to smoothly transition between grid positions since you need to make a continuous object and not just a collection of points
08:06 < fenn> huh. i guess if you're broadcasting the motion control instructions as a large number of tiny steps, you don't need any of this. all the complexity goes into the broadcast control unit
08:23 -!- CryptoDavid [uid14990@id-14990.uxbridge.irccloud.com] has joined #hplusroadmap
09:58 < geneh2> there really hasn't been much of a demand for mechanical computers
10:00 < geneh2> and laying out mechanical logic gates is much more difficult
10:03 < geneh2> punch card NC was electronic. The first one was at least. There was a patent for NC control with mechanical analog computers though
10:03 < fenn> it was not a "computer" though
10:04 < fenn> there were "electronic eye" BCD registers and counters for the encoder wheels
10:04 < geneh2> https://patents.google.com/patent/US2852189A/en
10:11 < geneh2> point is it wasn't all mechanical.
10:11 < geneh2> the patent above was basically a CNC as it could trace circular arcs.
10:12 < geneh2> It could be made all mechanical, but you really wouldn't want to do so.
11:17 -!- WiredSoul [~WiredSoul@user/WiredSoul] has joined #hplusroadmap
12:12 -!- CryptoDavid [uid14990@id-14990.uxbridge.irccloud.com] has quit [Quit: Connection closed for inactivity]
12:54 -!- darius___ is now known as abecedarius
13:05 -!- Gooberpatrol66 [~Gooberpat@user/gooberpatrol66] has quit [Ping timeout: 268 seconds]
13:49 -!- Gooberpatrol66 [~Gooberpat@user/gooberpatrol66] has joined #hplusroadmap
14:17 -!- TMM_ [hp@amanda.tmm.cx] has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
14:17 -!- TMM_ [hp@amanda.tmm.cx] has joined #hplusroadmap
14:22 -!- WiredSoul [~WiredSoul@user/WiredSoul] has quit [Quit: Leaving]
15:12 < hprmbridge> nmz787>  https://cdn.discordapp.com/attachments/1064664282450628710/1210001026703298590/PXL_20240221_230959041.MP.jpg?ex=65e8f845&is=65d68345&hm=56b07225db8a83a73fccbb4987290a66d099d86cc180ce70dee9728b3ec59b84&
15:12 -!- jrayhawk [~jrayhawk@user/jrayhawk] has quit [Ping timeout: 256 seconds]
15:24 -!- jrayhawk [~jrayhawk@user/jrayhawk] has joined #hplusroadmap
17:50 -!- darsie [~darsie@84-113-55-200.cable.dynamic.surfer.at] has quit [Ping timeout: 256 seconds]
19:15 -!- TMM_ [hp@amanda.tmm.cx] has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
19:15 -!- TMM_ [hp@amanda.tmm.cx] has joined #hplusroadmap
20:33 -!- mxz [~mxz@user/mxz] has quit [Ping timeout: 256 seconds]
22:39 -!- mxz [~mxz@user/mxz] has joined #hplusroadmap
23:01 -!- Guest19 [~Guest29@141.sub-174-243-249.myvzw.com] has joined #hplusroadmap
23:32 -!- superkuh [~superkuh@user/superkuh] has quit [Ping timeout: 272 seconds]
--- Log closed Thu Feb 22 00:00:06 2024