teepee changed the topic of #openscad to: OpenSCAD - The Programmers Solid 3D CAD Modeller | This channel is logged! | https://openscad.org/advent-calendar-2021/ | Website: http://www.openscad.org/ | FAQ: https://goo.gl/pcT7y3 | Request features / report bugs: https://goo.gl/lj0JRI | Tutorial: https://bit.ly/37P6z0B | Books: https://bit.ly/3xlLcQq | FOSDEM 2020: https://bit.ly/35xZGy6 | Logs: https://bit.ly/32MfbH5
lastrodamo has quit [Quit: Leaving]
GNUmoon has joined #openscad
<peeps[zen]> InPhase: well, i tried reaching out to the spiral_extrude developer in github comments. hopefully something comes of it.
<peeps[zen]> i haven't transcribed all the preconditions/corner cases we discussed yet, slightly worried about overwhelming them
<peeps[zen]> i was thinking some more about the child height vs pitch requirements. we could potentially make it even more flexible if we allow pitch < height, as long as the 2D geometry doesn't intersect itself when translated by Y=pitch
<peeps[zen]> eg: translate([0,2]) rotate(45) square([1,10]); rotate(45) square([1,10]);
<peeps[zen]> just ignore that it intersects the Y axis though ;)
<peeps[zen]> or... maybe that doesn't matter either?
ferdna has joined #openscad
ur5us_ has quit [Ping timeout: 240 seconds]
LordOfBikes has quit [Ping timeout: 256 seconds]
LordOfBikes has joined #openscad
Jack229 has joined #openscad
Jack22 has quit [Ping timeout: 256 seconds]
Jack229 has quit [Quit: Client closed]
Jack229 has joined #openscad
linext has quit [Read error: Connection reset by peer]
Zauberfisch has quit [Quit: Zauberfisch]
Zauberfisch has joined #openscad
submariner has quit [Remote host closed the connection]
peepsalot has quit [Quit: Leaving]
peepsalot has joined #openscad
arebil has joined #openscad
<InPhase> peeps[zen]: Hmm. I guess crossing the y-axis is okay for certain conditions.
<InPhase> But it's mostly a mess I guess.
<gbruno> [github] spuder opened issue #4028 (Crash when `--animate=4` but works when `--animate=3`). https://github.com/openscad/openscad/issues/4028
<gbruno> [github] spuder edited issue #4028 (Crash when `--animate=4` but works when `--animate=3`). https://github.com/openscad/openscad/issues/4028
miique has joined #openscad
PovilasCNC has joined #openscad
ccox has joined #openscad
ccox_ has quit [Ping timeout: 268 seconds]
ccox_ has joined #openscad
ccox has quit [Ping timeout: 268 seconds]
Guest3 has joined #openscad
Guest3 has quit [Client Quit]
ccox_ has quit []
ccox has joined #openscad
snaked has quit [Remote host closed the connection]
snaked has joined #openscad
<Jack229> inPhase .. in these "staircase"  extrusions or horn, some auger  you will cross the y-axis .. not so much in threads  https://pasteboard.co/h7j56ncSd3PI.png
splud has quit [Ping timeout: 240 seconds]
ferdna has quit [Quit: Leaving]
lastrodamo has joined #openscad
little_blossom has quit [Remote host closed the connection]
little_blossom has joined #openscad
aiyion has quit [Remote host closed the connection]
aiyion has joined #openscad
<pa> InPhase: thanks , yes i'm using ClosePoints since the inception of this model :-)
<pa> i think remodeling was the right approach, i'm just a bit slow because i have to update my 2d modeling app
<pa> plus, i discovered the curve i have is not the curve i initially used, that must have gotten lost in those initial pre-git days
<pa> so i have to recover that too
<pa> next feature: load raster as bg lol so i can model the curve faster
splud has joined #openscad
lastrodamo has quit [Quit: Leaving]
<JakeSays> Jack229: you wouldn't happen to have an English wrapper for your library, would ya?
<Jack229> haha
<Jack229> sure  https://www.dict.cc/
<Jack229> but every thing new is more english
<JakeSays> so that would be a no then
<Jack229> if i translate all modules and variables  - you still need to learn what they mean and do or?
<Jack229> i already put english explanations where possible
<JakeSays> i'm betting that i could figure out 90% of it just with naming
<JakeSays> i was just more curious than anything
<Jack229> you mean  x, y, z,  d h
<Jack229> JakeSays  is it the module names or variables causing confusion?
<Jack229> e.g.  i have a module Kreis which means circle but circle already exist so i can't use it
<JakeSays> you could use Circle
<Jack229> i also have a function kreis()
<JakeSays> so far it's just module/function names
<Jack229> Box()  or Ring()  wouldn't change
<Jack229> or Pille would be Pill
linext has joined #openscad
rvt_ has joined #openscad
<Jack229> in the readme.me  i have them all translated
<JakeSays> can modules and functions have the same name?
<Jack229> yes they can but i try to difference by capital letter for modules
<JakeSays> makes sense
<Jack229> because  else the color scheme can't diffentiate
<Jack229> https://github.com/UBaer21/UB.scad#functions  the list below the image isn't helping?
rvt_ has quit [Quit: rvt_]
<JakeSays> i'll look at it again tomorrow. i'm done for today
PovilasCNC has quit [Read error: Connection reset by peer]
TheAssassin has quit [Remote host closed the connection]
TheAssassin has joined #openscad
la1yv has quit [Read error: Connection reset by peer]
la1yv has joined #openscad
<Jack229> just tried to translate a module with www.deepl.com  which worked for most parts ..  but there are not many german words
linext has quit [Read error: Connection reset by peer]
la1yv has quit [Read error: Connection reset by peer]
la1yv has joined #openscad
linext has joined #openscad
linext has quit [Read error: Connection reset by peer]
rvt_ has joined #openscad
rvt_ has quit [Quit: rvt_]
rvt_ has joined #openscad
rvt_ has quit [Quit: rvt_]
rvt_ has joined #openscad
kaph has joined #openscad
rvt_ has quit [Quit: rvt_]
rvt_ has joined #openscad
rvt_ has quit [Quit: rvt_]
qeed has quit [Read error: Connection reset by peer]
qeed has joined #openscad
<lagash> Jack229: have you tried LibreTranslate? I've found it sometimes performs better than DeepL, AND it's libre software :)
<InPhase> Jack229: You can dual-language your library with secondary modules that just call the primary module. I haven't seen anyone do it, but, you sure could. :)
<Jack229> lagash just tested and interestingly  it  replaced some functions with a single letter ..  however  these tranlation tools are not made for code .. but  i ll keep an eye on libretranslate  thanks
<Jack229> InPhase yeah i am already doing it for compatibility when i change  names.   But that didn't change that i have no idea in what i should change Zylinder
<Jack229> some words are so similar  like Prisma and Prism   and  some  like  `Umkreis()` would be  `circumscribed_circle()`
arebil has quit [Quit: My keyboard has gone to sleep. ZZZzzz…]
<Jack229> lagash may i ask if you used libretranslate to translate into german  cuz this is were most fail and deepl is the only one i know to get this right (most of the time)
<Jack229> not sure how well they perform the other way
arebil has joined #openscad
<lagash> Jack229: translate INTO German? not as much, I'm learning German, but I know better than to rely on machine translation :)
<Jack229> that is the point of deepl  it is deeplearning  AI which is not translating words but search for similar occurrences  in text corpora
<Jack229> that way the translated result feels very natural
<Jack229> seems libre translate  is also using a neural system ( argos translate)
<peeps[zen]> teepee: do you know how codeql and lgtm work? it looks like its not possible to omit source files or directories from analysis if its a compiled language. like anything involved in the build is forced to be analyzed afaict
<teepee> I don't know why they behave strangely, CodeQL did show a list of issues allowing manual selection of "won't fix"
<teepee> at this point, only the windows build seems to fail, both CodeQL and lgtm are at least geen
<teepee> *green
<peeps[zen]> and then while reading lgtm docs, i saw a bunch of info about codeql, so is it using the same underlying tech?
<teepee> that's possible I suppose
<peeps[zen]> configuring these i so opaque im at my wits end
<peeps[zen]> i guess i'll just ignore them. once the changes are merged do you think they will stop reporting those same problems?
<peeps[zen]> i think i can get the windows build to work again, i'll just have to use two different approaches to integrate mimalloc depending on target OS
<peeps[zen]> maybe i'll send the reports upstream at least
<teepee> yeah, it's a bit strange, so macOS needs special integration and linux and windows full or so?
<teepee> a bit surprising for a project that looks quite mature with regard to platform supoort
<peeps[zen]> linux works pretty much any way you do it. mac didn't like the static override build option, windows didn't like the more direct function overrides
<peeps[zen]> they basically said *shrug* try clang-cl?
<peeps[zen]> which i didn't realize was a thing, but seems like a big change to convert to that
<peeps[zen]> the docs on mimalloc re: which build/link options are valid per OS platform are still confusing
<peeps[zen]> i've also been trying to get that other PR on VS builds to work (completely separate from mimalloc efforts for now), but getting kinda stuck. i'll write up my results in that PR soon
miique has quit [Ping timeout: 240 seconds]
rvt_ has joined #openscad
<teepee> cool, the VS PR would be nice to have
<teepee> sorry, just back home I'm a bit distracted soldering my 8266 boards :)
<peepsalot> ooh, watcha makin?
<teepee> the board was designed for my heng lamp, brightness control, 2 touch buttons and home assistant support
<teepee> but as aisler always makes batches of 3, I have 2 spare boards, and one will go into that signal lamp tower I printed lately
<peepsalot> what's a heng lamp? hanging?
rvt_ has quit [Quit: rvt_]
<teepee> not sure if it's a company or person name
<peepsalot> ah, looks like one of those dyson fans
<teepee> the old one I have unfortunately was just 5V led strip and *very* bright
<teepee> I think the new gen has some buttons, but i like my homeassistant one even better .9
<teepee> b
<teepee> oops, throwing around keyboards, not so ideal :)
<teepee> next board shall have esp32-c3 though, if I can make that work
linext has joined #openscad
Colt has quit [Remote host closed the connection]
Colt has joined #openscad
Colt has quit [Read error: Connection reset by peer]
miique has joined #openscad
<t4nk_freenode> when I have, for instance.. two modules, two cubes...
<t4nk_freenode> and I want to join them together exactly
<t4nk_freenode> is just popping them together by eye the correct way to make sure they will 3d print as one?
<t4nk_freenode> or so I need a mm overlap for instance?
<t4nk_freenode> what is the correct way?
<t4nk_freenode> normally I would do just the latter and pray.. and it seems to work
<t4nk_freenode> but now I hae a part, and I need to extend it on one side by 12.5mm
rvt_ has joined #openscad
<t4nk_freenode> to space it, and to use the extension to bolt it down
<teepee> if you don't do funny rotations, you can mostly get away with direct alignment
<teepee> the recommended way is having some overlap though
<teepee> it does not need to be mm, I usually use 0.01 or so
<t4nk_freenode> I aligned them flush and sliced in cura.. looks ok
rvt_ has quit [Quit: rvt_]
rvt_ has joined #openscad
othx has quit [Ping timeout: 256 seconds]
othx has joined #openscad
rvt_ has quit [Quit: rvt_]
linext has quit [Read error: Connection reset by peer]
miique has quit [Ping timeout: 240 seconds]
<teepee> peeps[zen]: it lives https://imgur.com/a/swHqHT8
<teepee> raspi for scale :)
<veverak> nice :)
<teepee> I guess I need to print 2 more levels. I bought 5 of the LED rings
<peepsalot> very cool
ferdna has joined #openscad
<InPhase> teepee: Cool. Now print 921599 more and you have a 720p display.
Guest63 has joined #openscad
Guest63 has quit [Client Quit]
<teepee> uh, that one all white is 2.2A @ 5V :)
<veverak> step downs are a thing :)))
<teepee> but those don't change power consumption other than increasing a bit due to losses
<teepee> just helps with thinner wires :)
<veverak> and it's usually easier to find 1kW power source for 48V rather than 5V
<teepee> a better candidate would be the table tennis wall, scaled up to maybe 480p?
<veverak> teepee: ad. wires: the I^2 term in "how much heat the wire does" is a nasty one
<teepee> problem would be also that out of that 1kW, probably 800W would end up as heat
<veverak> yup
<veverak> so you have to throw in another xxxW on cooling
<veverak> tbh I always wanted to build water cooled LEDs :)))
<teepee> it's a *signal* tower, not a tool to burn peoples eyes out
<veverak> damn
<veverak> :/
<veverak> reminds me that I have to check if the laser on my robot is strong enough to do that
<teepee> if you want that, there was a video using a 200W led IIRC?
<teepee> ouch, 1.5kW
<teepee> but it claims CRI 95+
<othx> teepee linked to YouTube video "Testing the world's biggest LED" => 1 IRC mentions
<veverak> :D :D
<teepee> actually that might be a newer video than the one I meant
<teepee> yeah, I think the one I remembered is https://www.youtube.com/watch?v=Q-hQXaxO-cc
<othx> teepee linked to YouTube video "Unboxing a $500 LED..." => 1 IRC mentions