Scopeuk has quit [Read error: Connection reset by peer]
Scopeuk5 is now known as Scopeuk
califax has quit [Remote host closed the connection]
califax has joined #openscad
Andy-pants has joined #openscad
castaway has quit [Ping timeout: 264 seconds]
J23 has quit [Quit: Client closed]
J23 has joined #openscad
<Andy-pants>
Is it possible to create a single hull of all of the actions performed in a for loop? (e.g. a 10 iteration loop, that creates a cube in a different location in each iteration)
<Andy-pants>
and creating a hull of those 10 cubes
<InPhase>
Sequential paired hull, or total hull?
<InPhase>
Yes to both, but they are very different things.
<Andy-pants>
Either or, actually. Since in this case, it would result in the same shape in the end. But, I assume that single hull would be better optimized for rendering.
<Andy-pants>
Code still needs a bit more debugging, clean-up, and refactoring. But the model generates correctly. Although, rendering the "bridge" takes quite some time. Which is what I'm looking to optimize
<InPhase>
hull of rounded cube is a much messier proposition than hull of a cube. That's going to be slow.
<InPhase>
You'll see it's not actually even more code. And if you're already running bezier_curves in there, I have confidence you can handle that math. :)
J23 has quit [Quit: Client closed]
J23 has joined #openscad
<Andy-pants>
hah, the bezier curve was easy. It's 3 points :D
<Andy-pants>
I'll dig into this though. Thank you for the example
<InPhase>
Probably that "-fillet_radius-0.1" should be reduced back to "-fillet_radius", and instead that should be an x coordinate -0.1 on point 1, and a +0.1 on point 2, to get a nice clean overlap.
<InPhase>
That was a leftover hack as I lined up the left side first.
<InPhase>
There's a fractional offset difference between the top of the curve and the piece connecting to that end, but I think that's a residual defect because of the asymmetric shape of that top piece's rounding.
<InPhase>
The rest of the ClosePoints curve could be smoothed better with t_step = 0.4/norm(point2-point1); Even with all those points it renders in 5 seconds in the master branch version of OpenSCAD.
<InPhase>
1 minute in the last release.
snaked has quit [Remote host closed the connection]
teepee has quit [Remote host closed the connection]
teepee has joined #openscad
L29Ah has quit [Ping timeout: 255 seconds]
Andy-pants has quit [Ping timeout: 255 seconds]
Andy-pants has joined #openscad
snakedGT has quit [Remote host closed the connection]
snakedGT has joined #openscad
LordOfBikes has quit [Ping timeout: 255 seconds]
LordOfBikes has joined #openscad
<InPhase>
Andy-Pants: I don't even know what this thing is supposed to do... But you nerd-sniped me, and I made a very smooth version. https://i.imgur.com/aILHIQd.png
<InPhase>
Andy-Pants: sliiiightly overboard. $fs = 0.4; at the top, and "for (a=[0:16:360])" at line 142, will give the equivalent print on most standard printers.
<InPhase>
Dropping that down to those reasonable values gives a 22 second render on the master branch, while keeping all those nice little rounded hole edges.
<Andy-pants>
But, close points is definitely what I'm looking for. I was looking for a path extrude which would keep something normal to the x plane. And, although not quite as straightforward, this gets the job done.
<InPhase>
Ah, one of those things.
ur5us has quit [Ping timeout: 248 seconds]
J2326 has joined #openscad
use-value has quit [Remote host closed the connection]
use-value has joined #openscad
J23 has quit [Ping timeout: 260 seconds]
TheAssassin has quit [Quit: No Ping reply in 180 seconds.]
<Vasiliy475>
how can I groupe expression in propper way?
<guso78>
my first useful manifold program.
<Vasiliy475>
how can I groupe expressions in propper way?
<guso78>
most of the code is for output
<teepee>
guso78: aha, nice, STL export
<guso78>
teepee, its really minimalistic and cannot do tesselation
<guso78>
does not calculate norm vectors either
<teepee>
use *.obj then, no tesselation needed then :)
<buZz>
teepee: lol
<teepee>
or was that *.off
<buZz>
tiktok videos :D
<teepee>
that's the currently used time measurements, no?
use-value has joined #openscad
<teepee>
Vasiliy475: what do you mean? do you have a short example you can share via pastebin?
Andy-pants has quit [Remote host closed the connection]
Andy-pants has joined #openscad
use-value has quit [Client Quit]
<teepee>
guso78: did you try one of those SDF examples, exporting the mesh for those?
guso7876 has joined #openscad
<Vasiliy475>
cuteepee normally in c { curved brackets uses for expressions grouping, of course there aspect with variables name
<guso7876>
no, did not look into the SDF examples yet
<teepee>
I suppose libfive is much more featureful, but if that stuff works same, one library is better than two :)
<teepee>
Vasiliy475: you can still group as you want you just can't assign variables that live in an outer scope
use-value has joined #openscad
<teepee>
or more generally you can't re-assign varaibles at all
Andy-pants has quit [Ping timeout: 255 seconds]
<Vasiliy475>
teepee please , one more question. line if(cb)difference() {stairsPlatets1(); platesCuttingAuxBody1();} If use thisdifference function it slows down my pc, what the problem?
<Vasiliy475>
teepee please , one more question. line 520 if(cb)difference() {stairsPlatets1(); platesCuttingAuxBody1();} If use thisdifference function it slows down my pc, what the problem?
<guso7876>
teepee, just looked at the sdf stuff included in manifold: there is one sdf directory and one sdf_test.cc file , which does not even have a main as it part of a whole test suite.
<guso7876>
the testsuite does not compile in my place without minor interaction.
<guso7876>
still try to provide missed header files
<teepee>
difference means more stuff to preview and/or calculate, so it's not too surprising. what are the details of that slowdown?
<teepee>
like when rotating the view after F5 (preview)
<teepee>
guso7876: I got only an error in gtest with that fixed, it did compile
<guso7876>
i am just wondering if i do generally wrong
<teepee>
do you have libglm-dev (debian name) maybe glm-devel on fedora?
<guso7876>
can i "make install" of the manifold ?
<J2326>
Vasiliy475 your linear_extrude uses 20 slices but there is no need for
<Vasiliy475>
teepee: yes in preview mode, after stl generation model rotating smooth but has no collors :( teepee thank you. Got to go and construct :)
<teepee>
Vasiliy475: that's the preview display logic then, the dev snapshots have some improved display code that can be enabled in preferences via the vbo-* options
<teepee>
or maybe vertex-buffer-*
<Vasiliy475>
J2326: used extrude from examle did not think about that. Which value you advice?
<teepee>
I can never remember which name those are using
<teepee>
guso7876: I don't see anything install-related in CMakeLists.txt but cmake might provide something by default
<teepee>
nope, on my system it did not allow make install
<guso7876>
i got such basic errors like: #include "manifold.h" , and file not found
<guso7876>
my own manifold test works way better :)
<teepee>
looks like that's not yet automated
<teepee>
but I did not look through the documentation
<J2326>
Vasiliy475 as these are just cubes slices=1 is enough .. also you don't need all these spheres(r=1) .. put them into a union and switch them off
guso7876 has quit [Quit: Client closed]
<J2326>
nice stairs btw
<Vasiliy475>
thank you, that a lot of time spends
<J2326>
maybe a tutorial would saved some of it
<J2326>
you could made a module Step() that is then called in a loop for the positions
<J2326>
there are a lot of options to make this much shorter
<J2326>
!tutorial
<Vasiliy475>
yes I think that will be a refactoring.
<J2326>
have you worked with for loops?
L29Ah has left #openscad [#openscad]
<Vasiliy475>
but there a lot individual parameters for 11 of 18 steps,
<Vasiliy475>
for loop used in my project
<Vasiliy475>
where it makes sense
<J2326>
you could use much more as you can put the parameters in an array (vector) and then loop will execute each of the parameter sets
<Vasiliy475>
this will harden understanding.
<J2326>
just a question of how familiar you are with the code
<Vasiliy475>
of course it will pleasure for programmers eyes
<Vasiliy475>
you will be surprise , 7 years in c. But this project my first one. In openscad language there is a lot features, there is also my learning progect, that why pure code stile.
<Vasiliy475>
Ok , leaving now. will read later.
Vasiliy475 has quit [Ping timeout: 260 seconds]
Andy-pants has quit [Ping timeout: 252 seconds]
Andy-pants has joined #openscad
Andy-pants has quit [Ping timeout: 255 seconds]
Vasiliy745 has joined #openscad
Vasiliy745 has quit [Client Quit]
Vasiliy745 has joined #openscad
Vasiliy745 has quit [Quit: Ping timeout (120 seconds)]
Andy-pants has joined #openscad
califax has quit [Ping timeout: 255 seconds]
califax has joined #openscad
snakedGT has quit [Quit: Leaving]
califax has quit [Ping timeout: 255 seconds]
Andy-pants has quit [Remote host closed the connection]
Andy-pants has joined #openscad
califax has joined #openscad
Andy-pants has quit [Ping timeout: 248 seconds]
Vasiliy475 has joined #openscad
KimK has quit [Ping timeout: 252 seconds]
Andy-pants has joined #openscad
use-value has quit [Quit: use-value]
califax has quit [Ping timeout: 255 seconds]
use-value has joined #openscad
L29Ah has joined #openscad
Andy-pants has quit [Ping timeout: 255 seconds]
califax has joined #openscad
Andy-pants has joined #openscad
tachoknight has joined #openscad
Andy-pants has quit [Ping timeout: 256 seconds]
Andy-pants has joined #openscad
Andy-pants has quit [Remote host closed the connection]
Andy-pants has joined #openscad
<lf94>
InPhase: I'm taking a mental day today from work. That means looking at OpenSCAD + libfive :)
<lf94>
Not sure if you saw my little nodejs binding example last night
<Av8r>
I'm working on adding some Qsettings for an export dialog box. These will not be visible (at least for now) in the "Preferences" dialog. I was going to give them a long name (eg, "ExportPdfOpts_ShowDsgnFN") to ensure uniqueness, although it looks like I could put them in a Group with a unique name (eg, "Export PDF Options"). Questions:
<Av8r>
(a) are contracted names okay (as first example)? (b) is it okay to use a dedicated group name (as second example)? (c) is it safe to use the Qsettings methods beginGroup() and endGroup() within a single method (the function using the modal dialog box)? and (d) it appears from Preferences a slash in a target name is interpreted as splitting the
<Av8r>
name into a Group and Target. Is this correct? (eg, is that a valid means to not use the beginGroup() and endGroup() methods?)
snaked has quit [Remote host closed the connection]
snaked has joined #openscad
Vasiliy475 has joined #openscad
Guest83 has joined #openscad
Guest83 has quit [Client Quit]
<teepee>
"egui is not a framework. egui is a library you call into, not an environment you program for."
<teepee>
very good
<teepee>
not sure about the immediate mode though
ur5us has joined #openscad
guso78 has joined #openscad
ur5us has quit [Ping timeout: 248 seconds]
Guest105 has joined #openscad
fling has quit [Remote host closed the connection]
fling has joined #openscad
Myster-D has joined #openscad
<Myster-D>
Hello. I'm trying to use openscad from the command line, and I have to pass a string parameter. I'm on Windows, and using a Powershell script. I've read the Wiki and tried all suggested solutions (doubling the double quotes, escaping double quotes using backslash) but I always get a " WARNING: Ignoring unknown variable 'mytext' in file ..."
<Myster-D>
any idea ?
<Myster-D>
my command looks like this : openscad.exe -o "$output_dir\key$index.stl" $source -D "keyIndex=""$index"";keyStyle=""$key_style"""
<Myster-D>
keyIndex is an int, and I can use it without any problem