peepsalot has quit [Read error: Connection reset by peer]
peepsalot has joined #openscad
default_ has quit [Ping timeout: 268 seconds]
The_Jag has joined #openscad
The_Jag_ has quit [Ping timeout: 268 seconds]
ferdna has quit [Quit: Leaving]
ur5us has quit [Ping timeout: 264 seconds]
gunnbr has joined #openscad
snaked has joined #openscad
snaked is now known as MustangJ
MustangJ is now known as snaked
GNUmoon has quit [Ping timeout: 276 seconds]
The_Jag has quit [Quit: The_Jag]
incal has left #openscad [ERC (IRC client for Emacs 29.0.50)]
GNUmoon has joined #openscad
arebil has quit [Quit: My keyboard has gone to sleep. ZZZzzz…]
GNUmoon has quit [Remote host closed the connection]
GNUmoon has joined #openscad
TheAssassin has quit [Remote host closed the connection]
TheAssassin has joined #openscad
TheAssassin has quit [Remote host closed the connection]
TheAssassin has joined #openscad
JoyaDK has joined #openscad
<JoyaDK>
Hi all, I'm new here, and need help to customize a text in a nameplate. I want the font "Questrial" -which is a Google font, and I can see it in the program. But it won't use it. i can see, that it isn't on the list of currently registered fonts with openscad. Is there a way, that i can get it registered? Thx a lot
<teepee>
which OS and in which program can you see it?
<JoyaDK>
I'm using windows 64bit
qeed_ has joined #openscad
<JoyaDK>
I found the program on Thingiverse... (Sweeping_Name_Plate_VZX.scad
<JoyaDK>
Hmm... I deleted the font cahce -and on reopening the program it updated -but did not fix the issue *sigh*
<JoyaDK>
i don't have a fontconfig folder in the windows system32
<JoyaDK>
hackwin commented on 10 Apr 2017 •
<JoyaDK>
Deleting the C:\Users[User]\AppData\Local\fontconfig folder fixed it for me for the GUI. Deleting C:\Windows\System32\config\systemprofile\AppData\Local\fontconfig fixed it for me for the CLI.
The_Jag has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
lastrodamo has joined #openscad
<JoyaDK>
I think I'm not smart enough to fix this... Will have to accept my failure *sigh*
<teepee>
:(
<teepee>
it's an annoying issue and difficult to solve for multiple reasons :(
<JoyaDK>
tried this: Maybe not the proper way, but I've had success with "use <path/to/font.ttf>" (did not work9
<teepee>
yes, it's only in that file mentioned in the comment
<JoyaDK>
Yeah maybe I'm not installing it as admin...
mhroncok has joined #openscad
ur5us has joined #openscad
<JoyaDK>
YAYAYAYAYAYAY I fixed it!!!
<JoyaDK>
Thank you sooooo much teepee!
<JoyaDK>
No more drag and drop fonts in my fonts folder... -just right click and install for all users...
ur5us has quit [Ping timeout: 260 seconds]
JoyaDK has quit [Quit: Client closed]
ochafik has joined #openscad
ochafik has quit [Ping timeout: 260 seconds]
ochafik has joined #openscad
ccox_ has joined #openscad
ccox has quit [Ping timeout: 268 seconds]
ferdna has joined #openscad
ferdna has quit [Remote host closed the connection]
ferdna has joined #openscad
ferdna has quit [Quit: Leaving]
Jack21 has joined #openscad
<Jack21>
is it known that "rands" not working proper in 2021.10.04?
<InPhase>
Not to my knowledge. What is not working properly?
<Jack21>
for (i=[0:10])rotate(rands(-30,30,2))cylinder(80,d=rands(5,20,1)[0]);
<Jack21>
echo(rands(-1,1,1,1));
<Jack21>
when you define a seed - all rands without seed will use that
<InPhase>
Seeds are sticky, yes.
<InPhase>
This is intentional, and many years old.
<InPhase>
I made that change after some discussion, as it makes it possible to do animation with rands in a sensible manner.
<Jack21>
so as soon you define one rands with seed in the hole program - none without is working anymore?
<InPhase>
They work, they just follow the seed, which is normal for rand behavior elsewhere.
<Jack21>
why should the rand for rotation follow the seed from a rand for a different object ?
<InPhase>
If you want unseeded rands, grab them first, rather than the old behavior which is if you wanted consistent random seeds you had to grab them first.
<InPhase>
It only matters if you want to mix true random with seeded random, which has more limited applications (although I've used it), in which case you pregrab the true random parts.
hisacro_ has quit [Ping timeout: 245 seconds]
<InPhase>
A better solution would be RNG objects, but we don't have that sort of structure yet.
<Jack21>
in 2019 this was different
<InPhase>
Correct. Late 2019 was the change.
<InPhase>
I realized the issue while trying to do animated procedural generation with rands and it was nearly impossible because we had no way to sticky a seed and still get rands in a functionally nested manner.
<InPhase>
The old behavior was weakening the ability to scale up in complexity.
<Jack21>
how is that an issue - if you want seeds you define seeds?
<Jack21>
also that the "sticky" rand is active after a second preview? or at least something changes between the first and second refresh
<peepsalot>
InPhase: hmm, trying the example, the behavior does seem odd. first F5 gives a random result for non-seeded values, then subsequent F5 keep the seed. should it not reset on each F5?
<InPhase>
peepsalot: It does reset on each F5.
<Jack21>
when i make these "random treey "things i just put a random for seed generation so each random get a random seed but without change.
<InPhase>
peepsalot: When in seed = -1 mode, the default.
<InPhase>
Jack21: So the problem is that every needle needs to be randomized differently, in a recursive manner, but in a way that can be made consistent across animation frames.
<peepsalot>
"<InPhase> If you want unseeded rands, grab them first" Jack21's example is doing that though? it behaves diffrently between 1st F5 and subsequent F5
<Jack21>
you use seeds=rands(0,1,amount_of_needles,42);
<InPhase>
peepsalot: Oh, Jack's example...
<InPhase>
peepsalot: So the sticky is sticking between runs there, yes...
<Jack21>
it feels strange that i can have an object with randomness and when i define an echo for a seed number all randomness stops in all scopes
<InPhase>
peepsalot: Oddly, the seed is not sticky between runs on the tree. Something is different...
<InPhase>
That aspect was tested, and still works on the original test. :)
<peepsalot>
which original test?
<InPhase>
Wait, my mistake. The tree is no longer working right.
<InPhase>
Somebody broke it.
<InPhase>
It only randomizes between the first and second preview, then it's locked in.
<InPhase>
They were unseeded values generated at the top.
<InPhase>
Now these do not come out unseeded.
<peepsalot>
doh, do we need to run a git bisect?
<InPhase>
Something is definitely different. The tree doesn't even rotate anymore...
<InPhase>
But I need to get to work right now. But yes, this requires investigating.
<Jack21>
for (i=[0:10])rotate(rands(-30,30,2,i))cylinder(80,d=rands(5,20,1)[0]); iam using these to find a nice positoning and then i fix the first seed but still want different diameters on each refresh (without changing a seed) guess this is not possible anymore
The_Jag has joined #openscad
The_Jag has quit [Client Quit]
The_Jag has joined #openscad
hisacro has joined #openscad
The_Jag has quit [Client Quit]
The_Jag has joined #openscad
The_Jag has quit [Client Quit]
The_Jag has joined #openscad
EkpyroticFrood has joined #openscad
<muesli>
let's say I got a cylinder with a variable amount of "sides" (cylinder(height, d=width, $fn=sides)), is there a way to determine the distance between the center of the cylinder and the center-point of each side?
<Jack21>
using sin gives you half the side length.. in a hexagon the radius and sides are equal
mhroncok has quit [Ping timeout: 268 seconds]
mhroncok has joined #openscad
The_Jag has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
ochafik has quit [Remote host closed the connection]
ochafik has joined #openscad
ochafik has quit [Ping timeout: 268 seconds]
ochafik has joined #openscad
<muesli>
Jack21: one thing that still strikes me as odd: if we assume 180 sides here, the distance would be exactly r, which it shouldn't be unless it's a perfect cylinder/circle
<peepsalot>
cos(1) != 1
<peepsalot>
muesli: you would need infinity sides for the distance to be r
<peepsalot>
the math holds up
<muesli>
oh, of course, i'm an idiot
<muesli>
thanks again. I feel like I'm back in 8th grade 😄
<Jack21>
radius is the hypothenuse and you have half of the center angle to calculate the adjacent side ( Ankathete )
<Jack21>
But oSCAD is perfect to show pupils why they need the math in 8th grade - i wish they would use this in school
mhroncok has quit [Quit: Leaving.]
ochafik has quit [Remote host closed the connection]
ochafik has joined #openscad
ochafik has quit [Ping timeout: 260 seconds]
Guest91 has joined #openscad
<Guest91>
Hello there!
<Guest91>
Can i change names on a stl file?
<InPhase>
The filename? Sure.
<Guest91>
Names in the project i mean!
<Guest91>
Sorry!
<InPhase>
There are no other names in a standard stl file other than the header information that is typically ignored.
<Guest91>
i got an stl file that i want change the names of the months, anyone could help me?
Guest91 is now known as Dini
<peepsalot>
you want to replace 3d text that is already rendered into an STL?
<Dini>
Yes!
<InPhase>
Dini: Ah, that's just a standard modeling problem. You use some intersections or differences to erase the old text, and then some more operations to add the new text.
<InPhase>
You're missing ub.scad in the upload. It does not work as-is.
<Jack21>
if i add the lib people end up with different version as i nearly change things every day - i thought it would be better to reference a central dl location
<Jack21>
also started some documentation on github
<InPhase>
Ok. The reference is missing though.
<Jack21>
in scad file ? there should be the link
<InPhase>
Where?
<Jack21>
v.gd/ubaer
<InPhase>
That's on a variable, not at the include.
<Jack21>
very top it also show in the console
<InPhase>
Oh, that assert? I didn't read that far down. I don't debug errors when includes are missing. :)
<Jack21>
i thought this would be a good idea (as it also checks if the version is right) .. so should i add a link on top of everything?
<InPhase>
Standard practice seems to be an inline comment right after the include.
<InPhase>
Although the assert is fine as well, but I had no mental reason to read it.
<InPhase>
Also the include should probably match the case-sensitivity of your download. :) You have ub.scad for include, and UB.scad for the download, so it only works on Windows.
<InPhase>
Got it displaying something now though with that fixed.
<Jack21>
it was inline .. seems it went down when i added the [Hidden] tab ..
<InPhase>
Ah.
<Jack21>
that case sensitive thing is strange as it is the file i upload .. seems there was a file with upper case in the very beginning
<Jack21>
will make a new download .. (also i like to switch to github but filedownload is not easy only as release with all other files there
<InPhase>
My German game is weak, but I set lamellen and schirm to true to try to make the whole thing display, but the top displayed at z=0 for some reason when I did that.
<InPhase>
Maybe it's only supposed to be one at a time?
<Jack21>
so the dl should be fixed now
<Jack21>
the lamellen is bit slow so this should be seperated .. the "stiel" only will show the "schirm" in position .. schirm only is for printing
<Jack21>
you can use schirm and lamellen so both are leveled for preview
<InPhase>
Oh, there it is. I was struggling to find your for loops in Lamellen...
<InPhase>
I was trying to figure out how you actually made the shapes, but you have aggressively hidden the code with a lack of whitespace. :)
<Jack21>
it is a bit tricky to get the angle right at the end of the "stiel" for display, and you wouldn't use that for printing
<InPhase>
I knew there had to be some loops somewhere.
<Jack21>
line 126 and 134
<Jack21>
also the "Polar" module is a radial loop with $idx counting up
Guest90 has joined #openscad
<InPhase>
Yeah, I see it now. hull over stepped cylinder pairs in a sequence.
<Jack21>
and the wave is by tilting them with a sin() and some randomness
<InPhase>
Imagine just replacing every cylinder h and r with a loop of 4 points, and then a list of these lists of 4 points, and it traces out the whole shape. ClosePoints then polyhedrons it from there.
<Jack21>
already seen your library a while ago - but a loop with hull is simpler - and i render this only once .. if i need this more often i would take a different approach
<InPhase>
To preserve the width, you apply RotZ to the lists of 4 points so that they are oriented along the direction of travel, which you can determine with the same step logic.
* InPhase
nods.
<InPhase>
Either way, it's very pretty as a result. :)
<InPhase>
Heading out to get my kid.
<Jack21>
.. do we start that advent calender soon?
<teepee>
there was some idea using cadhub this time, but it does not seem to get much attention, so I guess we just go ahead with the github repo
<teepee>
ok, you should have an invite in your mailbox
<teepee>
I suppose we need a folder per day this time, so we could store also extra stuff like STL maybe
<Jack21>
(3mf are so much smaller)
<teepee>
right, good idea :)
<teepee>
I spend so much time fighting with it, so we probably should push it a bit more
<Jack21>
i just had an 3mf file which the microsoft 3Dbuilder made weird by only opening and closing (no colors no other)
<Jack21>
the original loads in scad in seconds .. the file from 3Dbuilder nearly same size took 20 sec or so
<teepee>
maybe it split into multiple parts?
<Jack21>
not sure who to blame .. yeah but there was also some transformation so CURA (slicer) made 2700% scale (for the org. size)
<Jack21>
however 3mf allows browser to display in color (and with mapping) which is why i think it would be great if the colors in oSCAD could saved with the vertices in 3mf
<Jack21>
not sure how difficult a 3mf viewer in browser is - but would be a great way of display the advent calender results
<teepee>
oops, I don't mind the boring full name either :)
<Guest73>
I just need to make shit for 3d printers for my iot devices
<Guest73>
like cases and stuff
<Guest73>
The thing is, every time i send my blender file i exported out to cura it looks like it went thru a paper shredder
<teepee>
that seems a good use case for OpenSCAD indeed, as there's a number of designs you can customize for size
<Guest73>
blender took some time to learn so if i gotta learn something else I want to make sure i learn the right thing
<teepee>
blender is great for organic designs and such stuff. I don't think it can handle parametric designs as it's not really meant for mechanical cad
Jack21 has quit [Quit: Client closed]
<teepee>
the tricky part is that blender is a basic mesh editor so it allows for any mesh where slicers tend to require solid models (no missing faces or holes)
Jack21 has joined #openscad
<Jack21>
you can do this in blender but not with the gui you would start programming this in python
<Jack21>
FreeCAD can also handle parameters - but openSCAD is easier to learn - but it is a different approach to other CAD software
<dalias>
blender is not cad at all, and horrible for cad
<Jack21>
solid works or Fusion360 also not bad but this is not free software so i don't like it
<dalias>
for mechanical designs i would generally prefer openscad over pretty much anything else
russnelson has joined #openscad
<russnelson>
#join esp8266
<Jack21>
only when you start with freeform surfaces or rounding somthing on a spherical object - it gets difficult with openscad
<teepee>
na, #esp32 :P
<russnelson>
;)
<Jack21>
guest73: big advantage of scad are the amount of libraries you can use
<russnelson>
That gives a hint to my openscad development practice: start by defining x, y, and z sizes and thickness. Then make your boxes accordingly.
<Jack21>
i have a small script that takes any existing 3Dmodel and makes space for a PCB with a lid in it .. so maybe using different software synesthetic
<russnelson>
I've gotten into the habit of creating a box for any open PCB that I buy. Eliminates shorting one against another.
ochafik has quit [Remote host closed the connection]
The_Jag has joined #openscad
ochafik has joined #openscad
ochafik has quit [Remote host closed the connection]
<InPhase>
teepee: I like scadvent. scad, advent, and event sort of all mashed into one word.
<teepee>
there seems to be no good version than all lower case though
<teepee>
OpenSCADvent has some benefits in that regard :)
ochafik has joined #openscad
<InPhase>
I think just Scadvent or SCADvent, and then separately label it the OpenSCAD Advent calendar.
<InPhase>
It's a good idea to brand the event. :) Gives it a hashtag too.
<Jack21>
oSCADvent
<InPhase>
SCAD is a thing all by itself, since that's the file extension. :)
<Jack21>
stands for scripted computer aided design?
<Jack21>
oh another "bug" when you safe a file and while the dialog is open you rename the existing file - you get an overwrite warning but nothing is there to overwrite - seems it only checks for filenames when opening the dialog not when actually saving the file
lastrodamo has quit [Quit: Leaving]
<JakeSays>
Jack21: that's not a openscad bug.
<JakeSays>
oh well i suppose it depends on how the open dialog is being used
<InPhase>
Jack21: The troubling question then is what happens when you do the opposite, open the dialog, rename a file in place of what you're about to save it to, and then press save? :)
<InPhase>
Does it warn, clobber, or fail catastrophically?
<Jack21>
you get the warning
<Jack21>
two times !
<Jack21>
so the second time is the correct dialog .. but the first one is either not needed or wrong
<Jack21>
interesting that the system sound (windows) for the dialogs are different in volume
<InPhase>
Windows dialog sounds have caused me so much mental anguish over the years.
<JakeSays>
i disable all system sounds.
<JakeSays>
can't stand 'em
<InPhase>
Also, "bloop-bloop" from unplugging and reinserting usb devices.
Guest73 has quit [Ping timeout: 256 seconds]
<Jack21>
also windows is not able to make the mixer working -- they play in full volume even if the mixer is at 2% .. but when you have the mixer open it is working as it should
<Jack21>
.oO luckly i have a second monitor i seldom use so i move the mixer there