<tcurdt>
This is giving a super weird result and I don't see the problem yet.
<tcurdt>
function circle_polygon(r, vertices) = [for (i = [0:vertices-1]) [r*cos(2*PI*i/vertices), r*sin(2*PI*i/vertices)]]; !polygon(circle_polygon(20, 32));
<tcurdt>
nevermind 🙄
LordOfBikes has quit [Ping timeout: 255 seconds]
TylerTork has joined #openscad
LordOfBikes has joined #openscad
<TylerTork>
The "Customizer" on Thingiverse seems to be total $%#! and the developer(s) don't answer support requests. Lots of things see not to work, and there are no useful error messages to give hints about what it doesn't like. Am I alone in thinking this tool is abandoned and useless?
<TylerTork>
No comment?
<InPhase>
You are not alone. We've been whining about it since 2019.
<TylerTork>
Is anyone supposedly working on it? Can I do something to encourage that? Is it managed thru github or all private code? Can I offer a bounty?
<InPhase>
We even talked for a while to an employee there, but there aren't internal resources interested in investing in the site, so it's not really going to advance.
<InPhase>
It's not our end. It's inside the company that they'd need to make changes.
<InPhase>
OTHER sites have taken our webgl release of OpenSCAD and put up better customizers.
<InPhase>
Thingiverse is just antiquated.
<TylerTork>
OOH that's what I like to hear! Who's doing it right? I'll move all my stuff from Thingiverse to there
<TylerTork>
If they have a way to actually pay license fees to developers that would also be a plus
<InPhase>
Unlike thingiverse's version, which ran on a server which got overloaded regularly and cost them too much money, this runs in browser.
<InPhase>
So it might not be the wisest choice on a phone or something, but on a regular computer it scales a lot better, because you're not bogged down by other people using it.
<InPhase>
J24k43: You are still using Printables a lot, right? Did they do anything yet in the direction of a customizer?
<TylerTork>
the makerworld customizer is MUCH nicer than Thingiverse one, but does it have an equivalent to the $preview test? Sometimes I like to model things in one orientation (e.g. exploded assembly diagram) and then when they export have them oriented for printing
<InPhase>
I don't think so.
<InPhase>
But you can fake the equivalent with a bool for the customizer to switch between two configurations.
<InPhase>
You might lose that functionality in regular OpenSCAD eventually anyway. Advancements with Manifold are so impressive that the entire Preview subsection might be obsolete soon.
<InPhase>
Preview still has some unique capabilities, like color, but we will need to figure out color in the render path anyway so that we can do multi-material support.
<TylerTork>
I keep hearing about new things I never knew.
<InPhase>
If you haven't been using the latest nightly releases, render runs about as fast, and sometimes faster, than preview does for a lot of models.
<InPhase>
Some models are still slower, but in summary it's really on the same scale. So the idea of preview for big speed gains is already obsolete.
<InPhase>
To call it game changing is a bit of an understatement. :)
<TylerTork>
Okay, that's excellent news, but as I indicated, for me preview isn't just about speed. It's about usability. I don't want the user to have an extra step to choose the "printing orientation" before exporting, and especially I want to prevent people from exporting an STL file with the parts in the exploded arrangement, which would be impossible to
<TylerTork>
print. I want to make designs that are as foolproof as possible
<TylerTork>
Or for that matter, the "assembled with cross section to verify clearance" arrngement.
<TylerTork>
So right now I use $preview to intercept the render operation and make sure the output is reasonable for printing
<TylerTork>
who should I talk with about finding a way to satisfy these use cases? I don't insist it be done the same way, but I'd like to have SOME way
TylerTork has quit [Quit: Client closed]
<InPhase>
But I just gave the way. :)
qeed__ has joined #openscad
qeed_ has quit [Ping timeout: 255 seconds]
gimzmoe_ has joined #openscad
gimzmoe has quit [Ping timeout: 256 seconds]
qeed__ has quit [Quit: qeed__]
teepee_ has joined #openscad
teepee has quit [Ping timeout: 260 seconds]
teepee_ is now known as teepee
<guso78>
Teepee, creating a Pyramid means oversampling and calculating Points which are Not in Input Data.
<teepee>
there is probably a lot of room for a more flexible heightmap() with a much clearer outcome
<J24k43>
tcurdt guess you found that 2PI need to be 360
<J24k43>
inPhase no they are on the STEP track (but that was also forked from bambu) and afaik you can't change parameter there
<J24k43>
InPhase - $preview can use the same render but we still should keep that F5 $preview variable to switch lower resolutions.. i have models that still need render times of several seconds, you don't want that while working and designing. More important i use assembly so in $preview the parts are assembled while F6 will place them on the build plate
<J24k43>
to print
snakedGT has joined #openscad
<teepee>
there was some talk about a real-time-CSG algorithm, unfortunately there's not much public code but that would be cool to have
<J24k43>
if it is faster it would be handy for animations
<teepee>
it was meant for games :-)
snakedLX has quit [Ping timeout: 264 seconds]
<J24k43>
.oO( the new openSCAD GE game Engine)
cart_ has joined #openscad
<t4nk_fn>
written exclusively in python for optimal performance
<teepee>
"1. Upload your original 3D models to MakerWorld in .3MF or .STL format. Including a Print Profile with your model can help you gain more Likes and Downloads."
<teepee>
there seems to be something missing from that list of formats ;-)
<J24k43>
with 3mf they meaning normally the bambustudio project files
<J24k43>
ah it is in the details page ; Note: For this contest, models uploaded before the contest period will be allowed as contest entries. All entries must include .scad files with your model uploads to obtain the customize button.
<teepee>
ah, ok. I did not see that yet
<teepee>
I suppose the rules page is always the same
<J24k43>
guess the rules never change .. but you are right that this is confusing .. however bambu is confusing - makerworld is and bambu studio ..
<J24k43>
in BS changing the nozzle make all your user processes disappear .. but even better - all your custom filament settings are gone too
<J24k43>
so now i have a PETg for 0.4 and another for 0.2 nozzle .. and according a print process for 0.4 and one for 0.2 ..
<J24k43>
but i do like that bambu shop is also offering all these little things like screws and dovels, pins, washer ,bearings, motors - and reasonable priced .. they really making that into a one stop shop
<Scopeuk>
I guess if you want to do vertical integration you either go whole hog or you fail
<J24k43>
interestingly bambu offers a USB connector right on the inside so you can power additional LED
<Scopeuk>
yeh I think big tree tech use it for there multi printer control box thing
<InPhase>
J24k43: But how does "lowpoly = true;" differ from having $preview with a separate gui button?
<InPhase>
J24k43: If we completely purged the preview subsystem, people could even put "$preview = false;" in designs to make it picked up by customizer.
<InPhase>
Or $preview = is_undef($preview) ? ...
<InPhase>
If we didn't have a preview system, I'm confident we wouldn't add all those separate buttons and menu options just to work with what is a dedicated boolean change, when we have lots of support for doing this in the general case. :)
<J24k43>
InPhase - it is about that you update your design with F5 or F6 .. also how external editor will update with F5 getting a "preview" not the full resolution render
<J24k43>
but sure adding $preview as parameter will be used if there is no F5/F6
Tom85 has joined #openscad
<Tom85>
Hi everyone, I have a question. I'm compiling OpenSCAD for Windows, and everything works fine, the program compiles and runs. I'm using the scripts source ./scripts/setenv-mingw-xbuild.sh 64, ./scripts/mingw-x-build-dependencies.sh 64, and ./scripts/release-common.sh mingw64. I would like the program to compile using the USE_GLAD definition, but
<Tom85>
I'm not sure how to set this up when compiling for Windows.
<teepee>
I'm not sure that's a supported combination
<Tom85>
in code i can see some attempts: #ifdef USE_GLAD
<teepee>
from the comment it sound like help would be good
Tom85 has quit [Quit: Client closed]
teepee_ has joined #openscad
teepee_ has quit [Remote host closed the connection]
teepee_ has joined #openscad
teepee has quit [Ping timeout: 260 seconds]
teepee_ is now known as teepee
<t4nk_fn>
brrr, I'm in over my head with that stuff.. still messing around trying to understand opengl altogether ;)
J24k43 has quit [Quit: Client closed]
J24k43 has joined #openscad
mmu_man has quit [Ping timeout: 268 seconds]
mmu_man has joined #openscad
otsakir has joined #openscad
<otsakir>
hey! i've addressed the changes requested in this pull request https://github.com/openscad/openscad/pull/5031 (quite) some time ago. It's still blocked. Can someone have a look ? Is there anything more to be done ?
SamantazFox has quit [Quit: Bye]
paddymahoney has quit [Ping timeout: 264 seconds]
<teepee>
otsakir: nice, I'll try, but hope kintel beats me to it ;-)
paddymahoney has joined #openscad
<otsakir>
teepee: great, thanks!
<otsakir>
Btw, i think i'm a little bit confused about the workflow for contributing to openscad. I assume it's not enough to poke people (reviewers) from github issues/PRs. Instead, it's this channel that keeps things going. Is this correct ?
<teepee>
it's not enough people for an actual workflow unfortunately
<teepee>
both github and channel here are fine in general, I'd say
<teepee>
of course in some cases it might be simpler to have a quick discussion here when needed
<otsakir>
I see. Thanks.
<teepee>
basically an occational nudge is appreciated ;-)
<otsakir>
ok clear. I just wanted to know if some mentions just slip through or it's because you people have an overview and plan for everything but my PR was further down the queue. Obviously it's the first.
<teepee>
yes, notices can get lost and the list of unfinished open topics is huge, I'm struggling to finish my own PRs lately
<J24k43>
i only find PWFA.. and i am sure you didn't meant that
krushia has joined #openscad
<InPhase>
J24k43: Oh, that's interesting and weird. It's multicolor, but using a long bowden tube which it retracts all the way out of, then auto-inserts all the way down the long bowden tube for the next color.
<InPhase>
I guess that's one way to get a lot of colors.
<InPhase>
But at the expense of a long color-change lag.
<J24k43>
InPhase that is how all systems work with a single direct exruder .. the AMS (bambu) can add 4×4spool units and so you have some meter ptfe tubing
<InPhase>
I'd really like to see something like a 5 color mixing chamber hot end with CMYKW.
<InPhase>
Although I haven't figured out how you keep the pressure one-way...
<J24k43>
they exist but you would need static mixer within to get better results and they still generate lot of waste when purging