teepee changed the topic of #openscad to: OpenSCAD - The Programmers Solid 3D CAD Modeller | This channel is logged! | 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
mtm has quit [Ping timeout: 252 seconds]
mtm has joined #openscad
mmu_man has quit [Ping timeout: 248 seconds]
oldlaptop has joined #openscad
paddymahoney has quit [Ping timeout: 246 seconds]
mmu_man has joined #openscad
paddymahoney has joined #openscad
mmu_man has quit [Ping timeout: 252 seconds]
mmu_man has joined #openscad
peeps[zen] has quit [Read error: Connection reset by peer]
peepsalot has joined #openscad
<UltimateCodeWarr> Has anyone ever seen any type of good design for slide in mounts.   Thinking something like X shaped Extruded Aluminum or the where you can have a hollow side of it integrated with a hull, and the corresponding solid could slide on in for a tool less connection?
<UltimateCodeWarr> Picatinny isn't bad, just  lookin for something that might be a bit easier.
J24k49 has joined #openscad
J24k60 has quit [Ping timeout: 256 seconds]
califax has quit [Remote host closed the connection]
califax has joined #openscad
<UltimateCodeWarr> Are there any Reverse Engineering Tools to Convert STL->OpenSCAD?
LordOfBikes has quit [Ping timeout: 265 seconds]
<buZz> not really, no
LordOfBikes has joined #openscad
<UltimateCodeWarr> I figured it was a long shot, but I know there are some guys in this forum who love a mathematical challenge.
<UltimateCodeWarr> I found a Picatinny Rail, but can't find the mount for it in OpenSCAD.   It's like a blue M&M / beanie baby / white buffalo.
stefanct__ has joined #openscad
<buZz> why not design what you need? :)
<buZz> or get chatgpt to do it for you
stefanct has quit [Ping timeout: 252 seconds]
stefanct__ is now known as stefanct
snaked has joined #openscad
<UltimateCodeWarr> Well, I don't like reinventing the wheel.
<UltimateCodeWarr> ChatGPT can't figure out even the female polygon to mount the rail.
<UltimateCodeWarr> Neither can Claude, at least ....my ability to describe it.
<UltimateCodeWarr> The nice thing if you can find something paired is that someone else when through all the trouble to design, build, test it with a 3D printer most likely so the tolerances should be spot-on.
mmu_man has quit [Ping timeout: 260 seconds]
<UltimateCodeWarr> I did create my own DIN-style sliding/locking camera mount:    https://imgur.com/a/qLxpb3k
germ has quit [Remote host closed the connection]
<gbruno> [github] hzeller synchronize pull request #5312 (Make headers include relative to project-root.) https://github.com/openscad/openscad/pull/5312
germ has joined #openscad
<gbruno> [github] hzeller synchronize pull request #5312 (Make headers include relative to project-root.) https://github.com/openscad/openscad/pull/5312
<pca006132> guso78: it probably means that the glm version you have is a bit too old
<InPhase> UltimateCodeWarr: LLMs have tended to be notoriously bad at OpenSCAD.
<pca006132> UltimateCodeWarr: I think there were tools for compiling stl to openscad primitives, and run some solver to simplify the expression
<pca006132> but those are probably far from the 'real' openscad primitives you expect
<pca006132> they don't really tell you the design intents
<UltimateCodeWarr> Well, I just felt A great Relief in the FORCE, CAD employees and aficionados everywhere just drew a sigh of relief.
<pca006132> and I think it is probably much harder to deal with curves
<pca006132> I feel like converting design drawings into openscad is probably simpler than converting stl to openscad with design intents...
<UltimateCodeWarr> OCR is pretty cool tech.
<UltimateCodeWarr> Haven't seen a 3D scanner yet that can transform everyday objects back into Geometrical Primitives.
<pca006132> computer vision is hard
<pca006132> I would expect you need something like semantic segmentation in 3D to split an assembly of object into components
<pca006132> and then start to look into how to get back the primitives
<pca006132> plus errors in 3D scanned point cloud is making things harder, because you now have to account for errors in the "disassembly" of your object
<pca006132> the paper on simplifying openscad primitives is a precise method, theoretically sound, but this also makes it nearly impossible to handle errors
<UltimateCodeWarr> So, If I have a funky polygon, is there anyway to scale it up/down so that there is enough tolerance for things to slide over it like a sheath?   Word on the street is 0.3 mm would be a loose slide, 0.15mm larger would be a tight slide.
<UltimateCodeWarr> Or would it just be cleaner to adjust the sides 1 by 1.
myosotis has quit [Quit: myosotis]
<UltimateCodeWarr> So when you export to STL to get ready for a print, should you manually comment out parts of the drawing and export each piece to a separate STL, or is there a way when you import the entire model into the slicer to break up the pieces and lay them out on the build tray?
<J24k49> UltimateCodeWarr as STL have only triangles not objects like 3mf only if the algorithm can tell the objects apart (e.g. if they are not touching) you can separate them into parts or objects in most slicers
<UltimateCodeWarr> I'll probably need a tutorial on the separation,  wasn't obvious in Bambu Slicer, at least with 1 minute of fooling aorund with it.
<UltimateCodeWarr> Thanks for that Snap Lock Device, it's got a SCAD too, that's great!
<UltimateCodeWarr> I thought about just grafting in a lego block somewhere and then the rest would be childs play
<UltimateCodeWarr> I was playing around with scaling the Picatinny Rail up by  1.1 and then subtracting that from a cube in order to get the Female Sleve for it: https://pastebin.com/nBAg0pCX
<UltimateCodeWarr> Was about to 3D print it to see if my Vodomath skills were dead on.
<J24k49> just right click ↦ split into parts .. or use the  "P" or "O" symbols on top
<UltimateCodeWarr> So after the import do this?
<J24k49> before no object is there - Ü
<J24k49> you can also cut objects in parts
<J24k49> (hotkey C)
<J24k49> depending on the rail you can difference it (if the rail is smooth) but you need to lock it somehow with a set screw
<J24k49> or just use the holder i posted that has a printed screw to lock
<UltimateCodeWarr> Yes, it would need some mods to get it to lock into place for sure.   For my application I would print it large so that it would be walled in by sidewalls that are unscrewable.
<UltimateCodeWarr> So it's not going to move once the sides are shut.
<J24k49> you can also difference a smaller rail and add a vertical slot so it works as a clamp
<UltimateCodeWarr> This guy had a good design, but it wwasn't openscad.    https://www.myminifactory.com/object/3d-print-picatinny-female-to-female-mater-adapter-116901
<UltimateCodeWarr> He cut a slit so it had some give to it so it could be cinched up.    Not good for NATO, but good for Airsoft
<J24k49> my design comes with the 2D shape you just extrude for that
<UltimateCodeWarr> You have a variation of the Picatinny rail?
<J24k49> RailBlock(s=[30,50]);
lockywolf has joined #openscad
<UltimateCodeWarr> That's really cool now that I understand what it is.    It's a different kind of female mount
<UltimateCodeWarr> Different kind of rail too
<J24k49> https://en.wikipedia.org/wiki/NATO_Accessory_Rail#/media/File:STANAG_4694_%E2%80%9CNATO_Accessory_Rail%E2%80%9D_1.jpg
<UltimateCodeWarr> Your mount is like a Trek Mountain Bike Compared to my Chukudu
<J24k49> chukudu is like an E-scooter without E
<UltimateCodeWarr> Bone-Antler  Flintstone Tech.
<J24k49> you also can just use the matching profiles in a slicer https://www.printables.com/model/278478-nato-rail/files#preview.file.ocV7I to modify your objects
<UltimateCodeWarr> Is there a way to combine multiple make files?
<UltimateCodeWarr> So I can print one cornicopia of parts.
<J24k49> select parts ↦ merge  so they become one object
<J24k49> else you can just put multiple objects on your build plate
<UltimateCodeWarr> So I downloaded that file, there wasn't STL files, just .3mf files.  I'm used to dragging and dropping STL's on the Bambu Labs Build plate.    I can't do that with .3mf.      I don't understand quite how to merge .3mf into 1 .3mf and import that into Bambu Studio
<J24k49> no you just drag them all into BS - if you want to print them together as one unified part you can merge them
<J24k49> When draging 3mf files in you will be asked if you want to import geometry only - YES
<UltimateCodeWarr> Dragging would only load one .3mf at a time for me in windows.  I could do an import, multi-select the files and import them.  But it asked me about the geometries and converting to mm's  so...lets try a print.
<J24k49> after the first 3mf - every 3mf will ask you to import geometry only and adding to the existing
guso78k has joined #openscad
<guso78k> pca006132, when compiling with mxe , which glm installation is the right one ?
<guso78k> a) the one in /usr/include /usr/lib  b) any  glm in ~/openscad_dependencies c) other option ?
<UltimateCodeWarr> Ok JK, sent it to the X1C For Printing.    1hr 13 mins.     There was some sort of 8-segment numbers under the rails that I had to split into objects and delete.  Was that intentional?
<J24k49> i included a version number which cura slicer just omitted cuz of the small size - but other slicer have troubles .. so i am not doing this anymore (and waiting for an option to include this as meta data in the 3mf)
<UltimateCodeWarr> When I imported the model, it looked like it was resting on it.  Not being familiar with it, I was wondering how it was going to print it being on stilts, but when I hit the Auto Table Button it put the mounts up at a 45 deg angle.  So They are printing now at a 45 deg angle, wonder how that will impact the performace.
<UltimateCodeWarr> "Auto Orient Button"
<UltimateCodeWarr> Maybe easier to just emboss a QR Code into the model itself?
paddymahoney has quit [Ping timeout: 252 seconds]
paddymahoney has joined #openscad
<J24k49> it is a separate object you can delete with a click in the object browser
<UltimateCodeWarr> I'm printing one of those.   I would have to modify it with some sort of stilts to hold up a rasbperry pi board.
<UltimateCodeWarr> I might also use this system instead of my DIN camera mount.   .
<UltimateCodeWarr> It's just nice to have multiple mounting options that are standardized so that I can mount more hardware inside the camera hull.
<UltimateCodeWarr> Looks like Bambu's default Infill % is 15%.    Is this a good number for mounts like this?
misterfish has joined #openscad
teepee has quit [Remote host closed the connection]
teepee has joined #openscad
Joel has quit [Ping timeout: 245 seconds]
Joel has joined #openscad
<UltimateCodeWarr> JK, the Nato rails were supurb.    Is there a way to lengthen the bottom (Z) of the rail so that it can be grafted into another object without affecting the screw slide ability?
<UltimateCodeWarr> Worse case I could always raise a cube out of the material to join the floating rail.
cart_ has joined #openscad
Guest14 has joined #openscad
Guest14 has quit [Client Quit]
<J24k49> UltimateCodeWarr there is a parameter within the module that can be changed
<J24k49> hFloor  will lift the rail up
<J24k49> or use hMax is probably better
<J24k49> UltimateCodeWarr have updated and uploaded v3  where you have a parameter "ext" to extend the rail to join
foul_owl has quit [Ping timeout: 246 seconds]
aiyion1 has quit [Remote host closed the connection]
aiyion1 has joined #openscad
foul_owl has joined #openscad
foul_owl has quit [Read error: Connection reset by peer]
foul_owl has joined #openscad
mmu_man has joined #openscad
Guest74 has joined #openscad
<Guest74> / Define the size of the cube
<Guest74> size = 20;
<Guest74> / Generate the cube
<Guest74> cube([size, size, size]);
Guest74 has quit [Client Quit]
<pca006132> guso78k: probably ~/openscad_dependencies
teepee_ has joined #openscad
teepee has quit [Ping timeout: 260 seconds]
teepee_ is now known as teepee
snaked has quit [Quit: Leaving]
cart_ has quit [Ping timeout: 260 seconds]
Ischwitch has joined #openscad
Ingvix has quit [Ping timeout: 248 seconds]
Ischwitch is now known as Ingvix
<teepee> guso78k: i don't think ~/openscad_dependencies should exist in MXE context at all
<guso78k> teepee, sorry, for inaccurate memory.
<guso78k> openscad scripts have definitely created ~/openscad_deps/mxe in my home directory
<guso78k>  my basic question is, where I can check/upgrade my glm version which is used for mxe
mtm has quit [Ping timeout: 260 seconds]
mtm has joined #openscad
deathonater has joined #openscad
Smeef has quit [Ping timeout: 248 seconds]
<teepee> I'm not aware of any valid combination of scripts that should do that
cart_ has joined #openscad
hyperair has quit [Ping timeout: 252 seconds]
<guso78k> its in the very master branch  , and documentation refers to that file
<guso78k> after all, its not my intent to discuss about the right place of mxe build, but rather glm ...
<teepee> mingw/msys build is totally different from mxe
<teepee> they don't even share the same platform to run on. msys is on windows mxe on linux
<guso78k> at least in the past they appeared to have something in common.
<guso78k> probably I still use very old fashioned scripts to compile in my computer
<teepee> no
<teepee> the builds are different, the msys build did not exist at the time
<teepee> but you are right with that script, I mixed up that one
<teepee> we certainly have confusing script names
<teepee> I checked the docker image for MXE and that does not use glm
<teepee> so my guess is, it's (wrongly) using the system version and that happens to work because it's just header files
<guso78k> Up to now, I still successfully create working windows installer with my notes in https://bpa.st/GM5TM
<teepee> so basically the build scripts are probably wrong and work only due to lucky combination of other files it's picking up
<guso78k> So I am wondering, why recent MXE CI builds with manifold where manifold requires glm
<guso78k> lucky me :')
<teepee> the nightly builds work too, but it sounds like they should not
<guso78k> ... with manifold works , as manifold requires glm
guso78k has quit [Quit: Client closed]
ubuntourist has joined #openscad
<J24k49> probably nothing can be done here if that is the driver https://www.reddit.com/r/openscad/comments/1flchg1/comment/loi41g8/   as it seems to be also in the official built - it is probably a very special client hardware issue
<J24k49> strange preview error
<teepee> yep. maybe trying "force goldfeather" algorithm helps, it generates different preview geometry
myosotis has joined #openscad
<pca006132> I think we can force it to use llvmpipe
<pca006132> something like this
<pca006132> ...but this will be slower due to software rendering
kintel has joined #openscad
ubuntourist has quit [Quit: Leaving]
<kintel> teepee Any comments on the libsvg include paths in https://github.com/openscad/openscad/pull/5312 ? Trying to keep up with hzeller while he's motivated to clean up stuff :)
mmu_man has quit [Read error: Connection reset by peer]
TheCoffeMaker has joined #openscad
misterfish has quit [Ping timeout: 252 seconds]
hyperair has joined #openscad
<teepee> kintel: strong nod to 'I think we should let libsvg includes be referenced as #include "libsvg/*.h"'
<teepee> I don't mind the location much, I don't really have an opinion between the 2 locations
<teepee> heh, "Files changed 435"
<kintel> I hope that once the cleanup is done, formatting and code style can be largely automated by common editor addons.
<teepee> cool, till the next re-format in 2 years ;-)
<kintel> heh, yeah, but getting off uncrustify is nice in itself
<kintel> Reformat-on-merge would be nice. Scary, but nice
<teepee> yes, switching to clang tools seems like a good step
<teepee> hmm, I would not like automatic reformat too much, block merging would be a bit safer but potentially annoying at times
<kintel> fair, it's mostly useful as an editor tool. Not having to think about formatting, or single-click reformat is nice sometimes
<pca006132> automatic reformat is just too scary
<pca006132> from our experience in manifold, it is fine to ask people to format their code
<pca006132> I have so many rants for cmake lol
<pca006132> we originally thought that FetchContent is a good thing, but apparently it is driving people mad when they try to integrate with their existing project, redownloading stuff over and over etc.
<pca006132> I don't really notice this as I have a 1 Gbps internet connection and downloading small libraries are typically pretty fast for me
<teepee> yes. also there are build systems that have no network access during build at all
<pca006132> that is actually fine
<pca006132> there is a FetchContent path or something that user can override
<pca006132> and because it just timeout when it tries to download stuff, it is easy to debug and find the setting
<pca006132> and I'm thinking about renaming MANIFOLD_PAR option to just ON and OFF
<kintel> Whatever you do, don't invent a new build system to "fix" CMake's issues ;)
<kintel> If you want to see a complex build setup, just look at gettext :)
<kintel> they also have multiple included libraries, with support for building them inline or using external ones. I think they bundle the sources though, which is important for reproducible builds
<kintel> ..and it's autotools, and I don't think you want to go there :/
<teepee> standards!
<othx> standards is https://xkcd.com/927/
cart_ has quit [Ping timeout: 244 seconds]
<pca006132> oh I did use autotools
<pca006132> that thing is painful
<pca006132> (by using I mean run it with other projects and set some parameters)
<gbruno> [github] hzeller opened issue #5319 (Render with Manifold: transparent solids can look 'streaky') https://github.com/openscad/openscad/issues/5319
<gbruno> [github] hzeller edited issue #5319 (Render with Manifold: transparent solids can look 'streaky') https://github.com/openscad/openscad/issues/5319
<gbruno> [github] hzeller edited issue #5319 (Render with Manifold: transparent solids can look 'streaky') https://github.com/openscad/openscad/issues/5319
cart_ has joined #openscad
L29Ah has left #openscad [#openscad]
aiyion1 has quit [Ping timeout: 260 seconds]
aiyion1 has joined #openscad
L29Ah has joined #openscad
guerd87 has quit [Read error: Connection reset by peer]
teepee_ has joined #openscad
teepee has quit [Ping timeout: 260 seconds]
teepee_ is now known as teepee
cart_ has quit [Quit: Konversation terminated!]
cart_ has joined #openscad
guso78k has joined #openscad
teepee has quit [Remote host closed the connection]
teepee has joined #openscad
guerd87 has joined #openscad
UltimateCodeWarr has quit [Quit: Client closed]
<guso78k> My Animation in openscad is not very smooth and is stalling/jumping quite a lot.
<guso78k> Are there altenative methods from dumping pictures and creating a video ?
<othx> guso78k linked to YouTube video "Morphing a Cube into an Octahedreon" => 1 IRC mentions
guso78k has quit [Quit: Client closed]
kintel has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
mmu_man has joined #openscad
paddymahoney has quit [Ping timeout: 248 seconds]
<teepee> if you export the images, there should be no jumping
<buZz> i think they did it manually, the 'dump pictures' checkbox under animate wasnt checked
paddymahoney has joined #openscad
guso78k has joined #openscad
sinned69158 has quit [Ping timeout: 276 seconds]
guso78k has quit [Ping timeout: 256 seconds]
sinned69158 has joined #openscad
mmu_man has quit [Ping timeout: 244 seconds]
mmu_man has joined #openscad
L29Ah has left #openscad [#openscad]
<gbruno> [github] sesse opened issue #5320 (SIGSEGV with Minkowski operation using `manifold`) https://github.com/openscad/openscad/issues/5320
L29Ah has joined #openscad
L29Ah has left #openscad [#openscad]
L29Ah has joined #openscad
mmu_man has quit [Ping timeout: 246 seconds]
mmu_man has joined #openscad
<gbruno> [github] hzeller synchronize pull request #5312 (Make headers include relative to project-root.) https://github.com/openscad/openscad/pull/5312