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
<peepsalot> i wonder how much of recent issue spam can be blamed on nabble stopping services
<InPhase> teepee: Maybe we need to track down a webdev volunteer.
<InPhase> Some young padawan looking for a resume item.
<InPhase> We're not a nothing. "Created new website for the OpenSCAD project" has to be worth something in a first webdev job hunt. :)
<linext> i'm a web developer
<linext> check this out: http://jbcse.com/3dpartpricelib2/
<InPhase> linext: :) I make many interfaces that look like that for scientific software. People not accustomed to scientific software seem to get really upset when they see them. ;)
<linext> it's a work in progress
<InPhase> Sometimes though, you just got to get to the point. And sometimes numbers are the point.
<linext> i'm thinking of putting the web customizer somewhere on my domain
<linext> i wonder if the .STL file that's generated from OpenSCAD web assembly can be uploaded instead of downloaded
LordOfBikes has quit [Ping timeout: 268 seconds]
LordOfBikes has joined #openscad
J1A849 has joined #openscad
J1A84 has quit [Ping timeout: 252 seconds]
<joseph_> teepee InPhase: I'm sure you're wondering why I haven't checked in yet this week. Unfortunately my time has been consumed with responding to a serious unexpected external issue (if you need to know more, we can discuss my situation individually, not in the public IRC. It is fully unrelated to GSoC). Still, I'm trying to squeeze in hours of work time when I can. I did want to apprise you of my proposed solution for the failing CI
<joseph_> tests. First to disambiguate, I'm trying to set a different color for the "cut faces", which I think in previous discussions might have sometimes called "back faces", although I tend to think of back faces as usually referring to the invisible side opposite the face normal. Anyway, this is in reference to the problem that peepsalot warned about last month, which is that this information about the cut faces is lost when converting
<joseph_> from Nef3 polyhedrons to polysets. So I am charting out creation of a new function with Nef3 polyhedron as input which outputs a VBO storing a boolean marker of vertices for whose face is marked. This data will then be available as an attribute in the shader, and the cut face color will be provided as a new uniform. I might run into initial difficulty trying to ensure that the order of vertices is correct, and I expect a possible
<joseph_> headache trying to modify the existing VBO interleaving to fit it in. When you can, please let me know your reactions to this approach. And as always, let me know if you think I should shift tracks and spend time on something different first, before returning to this big main issue.
<peepsalot> joseph_: back faces and cut faces aren't really the same. back face is the rendering of a polygon on the "inside" of the geometry. these are normally culled from display (eg, search project for `glCullFace(GL_BACK);`), except in ThrownTogether View where they are highlighted magenta (useful for debugging winding order issues of polyhedron calls etc.)
<peepsalot> cut faces are just the new faces caused from a differeence operation. coincidentally, because of the way OpenCSG works these are technically backfaces, but in F6 Render view that shouldn't be the case.
<peepsalot> i *think* the NefPolyhedron stuff uses some custom attributes to indicate which faces are cut, rather than the winding order. but I never 100% understood how that aspect of the code works, so I might be wrong
<joseph_> peepsalot: Thanks for clarifying. I didn't know about that feature of ThrownTogether before, but now I know it is a separate issue that I need to handle in addition to the cut faces color. So that's why I misunderstood InPhase's previous comment. Fortunately I believe that magenta (or any other color) back faces can be produced without needing any new VBO because I think I can just disable back face culling and evaluate gl_
<joseph_> FrontFacing in the shader
zauberfisch_ has quit []
zauberfisch has joined #openscad
teepee has quit [Ping timeout: 268 seconds]
teepee has joined #openscad
<peepsalot> joseph_: yes that would be the shader based way of doing it. (also related is the fragment shader discard keyword, https://www.khronos.org/opengl/wiki/Fragment#Fragment_discard )
aiyion has quit [Remote host closed the connection]
aiyion has joined #openscad
<peepsalot> joseph_: transparent objects is one other case where back faces are drawn iirc, but we don't handle color/transparency in render view either (unless adding that was also part of your proposed changes)
califax has quit [Remote host closed the connection]
califax has joined #openscad
hrberg has joined #openscad
qeed has quit [Ping timeout: 252 seconds]
qeed has joined #openscad
jonasbits has joined #openscad
califax has quit [Read error: Connection reset by peer]
califax has joined #openscad
califax has quit [Remote host closed the connection]
califax has joined #openscad
ur5us has joined #openscad
califax has quit [Remote host closed the connection]
TheAssassin has quit [Remote host closed the connection]
califax has joined #openscad
TheAssassin has joined #openscad
TheAssassin has quit [Remote host closed the connection]
TheAssassin has joined #openscad
ur5us has quit [Ping timeout: 255 seconds]
aiyion has quit [Remote host closed the connection]
aiyion has joined #openscad
ur5us has joined #openscad
ur5us has quit [Ping timeout: 268 seconds]
buZz has quit [Ping timeout: 268 seconds]
buZz has joined #openscad
buZz is now known as Guest193
qeed_ has joined #openscad
TheAssassin has quit [Remote host closed the connection]
qeed has quit [Ping timeout: 268 seconds]
TheAssassin has joined #openscad
teepee_ has joined #openscad
teepee has quit [Ping timeout: 268 seconds]
teepee_ is now known as teepee
ali1234 has quit [Quit: Leaving.]
guerd871 has joined #openscad
guerd87 has quit [Read error: Connection reset by peer]
Guest193 is now known as buZz
rawgreaze_ has quit [Quit: ZNC 1.8.2 - https://znc.in]
rawgreaze has joined #openscad
rawgreaze has joined #openscad
rawgreaze has quit [Changing host]
GNUmoon2 has quit [Ping timeout: 268 seconds]
GNUmoon2 has joined #openscad
pie_ has quit [Ping timeout: 240 seconds]
J1A849 has quit [Quit: Client closed]
J1A849 has joined #openscad
GNUmoon2 has quit [Remote host closed the connection]
GNUmoon2 has joined #openscad
pie_ has joined #openscad
pie_ has quit []
pie_ 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
pie_ has quit []
pie_ has joined #openscad
J1A849 has quit [Quit: Client closed]
J1A849 has joined #openscad
aiyion has quit [Remote host closed the connection]
teepee has quit [Remote host closed the connection]
TheAssassin has quit [Remote host closed the connection]
GNUmoon2 has quit [Remote host closed the connection]
califax has quit [Write error: Broken pipe]
fling has quit [Remote host closed the connection]
teepee has joined #openscad
TheAssassin has joined #openscad
califax has joined #openscad
GNUmoon2 has joined #openscad
aiyion has joined #openscad
fling has joined #openscad
califax has quit [Remote host closed the connection]
califax has joined #openscad
<Friithian> does openscad specify units in the output file?
linext has quit [Quit: Leaving]
<J1A849> Friithian depends  SVG and 3mf is defined in mm.  STL is unit less
<Friithian> ok cool
<Friithian> maybe I should switch to 3mf
<J1A849> yeah and all these " i didn't check and now i have a benchy for ants" is  history
<dalias> 3mf is awful for lots of reasons
<dalias> most boil down to "[potentially] embedding all sorts of software-specific junk metadata into the file that you didn't intend to have there"
<dalias> the fact that you *can decide to* interpret the unit in STL as anything you want is not any kind of deficiency. file formats are all a matter of human convention and agreement; when the format does tag a unit, as e.g. mm, it's human convention/agreement that determines what "mm" means
<dalias> in the case of STL, the convention is that the unit is mm. if you're doing something else wacky, that's certainly possible, just like making a 3mf file that says it's mm but that's actually inches is possible. but you're breaking the convention and folks who get the file from you will be justifiably annoyed
<dalias> TL;DR STL is perfectly fine and STL's unit is mm
<Friithian> heck, the problem with having dual extruders is that nothing is optimized for it
<Friithian> I guess I have to design a door chock myself
<dalias> what are you trying to do with them?
<Friithian> I have some TPC (stretchy rubbery stuff) that I want as the bottom grip surface but the rest of it can just be PLA
<dalias> do they bond?
aiyion has quit [Remote host closed the connection]
<dalias> i'd think you'd need to either design a mechanically interlocking interface in the multi-material print to ensure they stay together, or just do them as separate prints fastened together somehow. but maybe they bond? iirc one youtuber did something like that successfully a couple years back, maybe cnc kitchen?
aiyion has joined #openscad
califax has quit [Remote host closed the connection]
<Friithian> yeah the PLA and TPC bind pretty well actually
califax has joined #openscad
<Friithian> I did a 10mm square ontop of another 10mm square of PLA and they were connected. With enough force they prob could've been seperated but eeh
<dalias> cool. i should try that sometime
<Friithian> this was TPC-91A, not the cheapest material ($52/750g) but it is fun
<dalias> i wonder if it's a mechanical bond between fine structures resulting from how the nozzle laid material, or something that'd work even if the surface was very smooth to begin with
<dalias> expensive stuff
<Friithian> natural PVA is $100 for 1kg
<dalias> really? at that price why not pvb instead? :-p
<dalias> (soluble in ipa and can be had for more like $35/kg :)
<Friithian> water is easier to use than ipa, and these printer cores are built for natural PVA
<dalias> oooh yousu's is only $22/kg now on amazon. i should buy some of this
<dalias> 23*
<buZz> 22 usd is cheap?
<buZz> last 1kg PLA i bought was 16 eu/kg
<buZz> oh, you're buying PVA for 22/kg? dang, link?
<buZz> PVB?
<buZz> PVB dissolves in IPA? weird
<buZz> Butanol, isopropanol and MEK are good solvents for PVB. Ethyl acetate can dissolve PVB.
<buZz> gee
<buZz> yeah MEK :D
<buZz> that'll dissolve your nose too :D worked SOMUCH with MEK
<buZz> made me shithigh
<Friithian> MEK is a good solvent for most plastic
<Friithian> I love MEK
<buZz> Friithian: my job then was building LEGO statues :D 3 meters high dinosaurs
<Friithian> you MEKed them together? whew
<buZz> had to be strong enough to let children climb on them
<Friithian> aaaaah
<buZz> yeah MEK is the go-to solvent for glueing LEGO statues
<buZz> if you want them to survive transport etc ;)
<Friithian> I hope you had the proper PPE for that daamn
<buZz> rofl
<buZz> nope ;P just took -very- often breaks
snaked has quit [Ping timeout: 268 seconds]
<dalias> i've never bought their pvb before (been meaning to try it tho), but yousu silk PLAs are nice
fling has quit [Remote host closed the connection]
teepee has quit [Remote host closed the connection]
fling has joined #openscad
<dalias> and they offer it as a 4-color minispool pack which is :) :) :)
teepee has joined #openscad
<dalias> slightly more ($29 for 1kg total) but still way less than polymaker's 1kg spools
<buZz> i never ever had a good experience with a silk PLA
<dalias> it needs lower temps and drying
<dalias> hardly matters if you're printing easy geometry, but overhangs are awful without drying. they turn into a foamed-up die-swelled mess
<Friithian> if ya wanted to see that
<dalias> i've done successful 85% dragons in silk tho, and moderately successful (but some of the smallest tail segments fused and broke and needed to be just glued immobile) down to 75% or maybe even 66%
TheAssassin has quit [Remote host closed the connection]
TheAssassin has joined #openscad
<Scopeuk> Friithian so that looks like a sodimm pc module (probably a cm3) and a fairly large motherboard. interesting
snaked has joined #openscad
<Friithian> oh it is very interesting
<Friithian> USB camera, usb extender for the front panel, the screen is just an HDMI touch screen
<Friithian> running debian buster for arm
TheAssassin has quit [Quit: No Ping reply in 180 seconds.]
TheAssassin has joined #openscad
<Scopeuk> at least that bit is all standard, my understanding is the ultimakers are designed to be a bit "black box", I've not worked with them
<Friithian> I can ssh into this one so
<Friithian> 5~currently troubleshooting a camera issue, I just ssh'ed in and opened dmesg, reinserted the usb, nothing happened. Quick deduction it is a hardware issue
<Scopeuk> that seams likely
<Friithian> I wonder what would happen if I put something in the mSD slot that the board has
<Friithian> well, one of the mSD slots the board has
<Scopeuk> I would guess it will just mount it, but I suppose its possible there is no reader support compiled in or they are reserved for overloading boot loaders or similar
ali1234 has joined #openscad
<Friithian> I kinda want an ultimaker for myself now, so I could actually hack it and do fun stuff
Guest53 has joined #openscad
GNUmoon2 has quit [Ping timeout: 268 seconds]
Guest53 has quit [Client Quit]
GNUmoon2 has joined #openscad
califax has quit [Remote host closed the connection]
califax has joined #openscad
snaked has quit [Ping timeout: 255 seconds]
teepee_ has joined #openscad
teepee has quit [Ping timeout: 268 seconds]
teepee_ is now known as teepee
<buZz> Friithian: is the usb port's power even enabled? might be in OTG mode
<Friithian> it is, I did a test with a usb drive, I also switched it to the front panel USB port, no dice
<buZz> usb drives often also work on 3.3V
<buZz> webcams often do not
<Friithian> I also tried plugging it into my laptop
<buZz> ah :P
<Friithian> not even a peep about USB stuff from dmesg, so, I assume it is dead
<buZz> too bad you're in the US
<buZz> we have a idle ultimaker 1 with dual extruder needing love
<buZz> ultimakers suck imho, especially the first ones, they are 100% wood, SUPERnoisy while printing, easy to bind, no bedheating, etc
<Friithian> I've only dealt with S5s, and I love em
<Friithian> especially given I didn't pay for it
<buZz> oh sure, 3000?
<buZz> ultimaker donates em around often to local schools/fablabs/etc
<Friithian> the S5 on its own for $6-7k
<Friithian> pro bundle for about $10k
<buZz> so some local schools ended up with piles of unmaintainable halfproprietary shitty printers
<buZz> of which we got one
<buZz> new ultimakers are 100% proprietary i think?
<Friithian> you think I can convince them to send us a pro bundle or two :D (This is a university makerspace)
<buZz> good reason to completely ignore imho :D
<buZz> perhaps, or just get a budget from uni, any $ US isnt spending on weapons means global peace potential
<buZz> ;)
<Friithian> honestly the fact they have an API on the printer themselves for interfacing with it over the network and I can ssh into them is pretty nice
<Friithian> hahahaha a budget, you think we have a budget
<buZz> no clue, i've had that on printers since early repraps
<Friithian> 95% of the machines we have are from donor's money and not the university's
<buZz> Friithian: so its not a university's makerspace, just a hobbyroom renting its own space?
<buZz> or free space from uni?
<Friithian> no it's the universty's makerspace, I am on university payroll for this
<buZz> lol but no budget? :D
<buZz> how much do they spend on that rugbyclone US is so nuts about, with the padded shoulder outfits
<Friithian> the university is trying to become D1 so…
<Friithian> we ask to buy things, the university will most likely say yes
<buZz> 1 dimensional?
<Friithian> division one
<buZz> is that a makerspace thing?
<Friithian> no, it's a sports thing
<buZz> oh, English Football League One thing?
<Friithian> sorta, it's a classification for university sports stuff
<buZz> is what google tells me 'division one' is
<buZz> aha
<buZz> only sport i took in highschool is math
<buZz> \o/
<Friithian> lol
<buZz> Friithian: cool @ https://github.com/gunqqer/pxltrm :D
<Friithian> oh yeah I was going to do something with that fork…
<buZz> lag commits, it seems ;)
<Friithian> I honestly forgot about that…
<buZz> i did 'terminal pixel art editing' a lot back when that was a normal thing for communication platforms to use ;) landlines, modems, BBS :)
<buZz> TheDraw and aciddraw are two i used back in the day
<Friithian> ok no I dont remember doing this at all… what the fuck
<buZz> its one of your 'popular repos'
<buZz> :P
<Friithian> huh
<Friithian> oh this was 4 years ago no wonder I dont remember it
<Friithian> musicScripts and StreamDye are stuff I've really actually done stuff with
<Friithian> oh that reminds me I should put my makerbot scripts up there
snaked has joined #openscad
<buZz> hmhm
<buZz> Friithian: did you know, github has a great .stl previewer aswell
<buZz> might be a nice place for your 3d printing projects ;)
<Friithian> most of the STLs I make aree cobbled together messes :D
qeed has joined #openscad
qeed_ has quit [Ping timeout: 268 seconds]
<InPhase> dalias: I've received a few, but not many, stl's that were made in inches. The usual symptom is "Why didn't it load in Cura?" *pause* "Oh wait, there it is in the middle..."
<InPhase> Most stl producers conform to mm after a little time in the 3D printing communities though.
<buZz> lol yeah, -very- rare
qeed_ has joined #openscad
qeed has quit [Ping timeout: 255 seconds]
ccox_ has joined #openscad
ccox has quit [Ping timeout: 268 seconds]
ccox has joined #openscad
ccox_ has quit [Ping timeout: 252 seconds]
teepee has quit [Remote host closed the connection]
teepee has joined #openscad
<teepee> real work in progress
ccox_ has joined #openscad
<teepee> so I guess yes and no
<teepee> obviously a full yes would be so much better
<ali1234> it doesn't seem to exist in nightly
<ali1234> so i thought maybe it was abandoned and removed or something
<teepee> correct, it did not make the jump to that state yet
<teepee> ideally it would be combined with the roof() stuff
<ali1234> what is a roof?
ccox has quit [Ping timeout: 268 seconds]
ccox_ has quit [Ping timeout: 252 seconds]
<teepee> yes
ali1234 has quit [Remote host closed the connection]
ccox has joined #openscad
ali1234 has joined #openscad
califax has quit [Remote host closed the connection]
GNUmoon2 has quit [Remote host closed the connection]
califax has joined #openscad
GNUmoon2 has joined #openscad
<ali1234> i'm trying to make a stencil btw
<ali1234> so i want to turn a curve path into a slot with round ends and sloped sides
<teepee> roof should work for that
<ali1234> yeah. ideally i would also want to minkowski a circle with a bezier path or something too
<ali1234> if offset_extrude is implemented then can't roof be implemented by just saying height = offset
<teepee> I suppose roof is pretty much a special case of offset extrude
<teepee> although not with the simple offset parameter but fixed angle
<ali1234> angle is just atan2(height, offset) no?
<ali1234> it could be optional arguments, like d/r for circles
<teepee> yes, but for a random input what offset would you give to produce the same result as roof()
<ali1234> specify any two
<ali1234> im not exactly sure what roof is supposed to do (if you explained it i missed it when getting dc'd)
<ali1234> is it supposed to go to a point?
<teepee> as the name implies it's similar to a roof of a building
<teepee> with 45° angle
<ali1234> well there's lot of different types of roofs :)
<teepee> pff
<ali1234> 45 degree ones are not very common cos they look evil :)
<teepee> yes, computers are both ugly and evil
<ali1234> but yeah, is it supposed to take an angle, and then compute the height that results in a point?
<ali1234> if so that's clearly more involved
<teepee> no, angle is fixed, it will just go to the height it will need
<ali1234> "need" for what?
<teepee> to end up in a line
<ali1234> okay that's what i mean by point
<ali1234> as in there's no flat bit on top
<teepee> yep
<ali1234> i see. yeah that does require something more than o_e
<teepee> for flat part a separate difference is needed
<teepee> and for different angles some scaling
<teepee> yes, it's like do offset extrude at angle till all geometry is collapsed into lines or points
<ali1234> that seems like it is going to be significantly harder
<teepee> harder in what respect?
<ali1234> well o_e doesn't do anything that openscad can't already do
<ali1234> extrude and offset already exist
<ali1234> this is just combining them in a way that isn't currently allowed by the language
<ali1234> roof requires actually calculating a bunch of constraints
<teepee> and that's the problem, the "simple" offset extrude gets complicated for more complex shapes
<teepee> the underlying algorithm for roof is supposed to help with that
<ali1234> why does it get complicated? if the offset makes parts of the geometry disappear?
GNUmoon2 has quit [Remote host closed the connection]
<teepee> I don't remember all the details, but there's quite some discussion in the github PR
GNUmoon2 has joined #openscad
<ali1234> i guess you'd want to use roof as like a "limit"
<ali1234> blender's bevel also has this problem: it just refuses to make the bevel any larger than the smallest thing you try to bevel
<ali1234> that's also a todo item
teepee_ has joined #openscad
teepee has quit [Ping timeout: 268 seconds]
teepee_ is now known as teepee
califax has quit [Ping timeout: 268 seconds]
califax has joined #openscad
<J1A849> i would assume the problem starts when your offset splits your polygon
snaked has quit [Ping timeout: 252 seconds]
snaked has joined #openscad
snaked has quit [Ping timeout: 255 seconds]
snaked has joined #openscad
aiyion has quit [Remote host closed the connection]
aiyion has joined #openscad
snaked has quit [Ping timeout: 244 seconds]
TheAssassin has quit [Remote host closed the connection]
fling has quit [Write error: Connection reset by peer]
teepee has quit [Remote host closed the connection]
TheAssassin has joined #openscad
fling has joined #openscad
teepee has joined #openscad
snakedGT has joined #openscad
hrberg has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]