arebil has quit [Quit: My keyboard has gone to sleep. ZZZzzz…]
ur5us has quit [Ping timeout: 240 seconds]
arebil has joined #openscad
ferdna has quit [Quit: Leaving]
GNUmoon has quit [Ping timeout: 276 seconds]
GNUmoon has joined #openscad
PovilasCNC has quit [Read error: Connection reset by peer]
PovilasCNC has joined #openscad
teepee has quit [Ping timeout: 276 seconds]
lastrodamo has joined #openscad
teepee has joined #openscad
stonkey has joined #openscad
lastrodamo has quit [Quit: Leaving]
lastrodamo has joined #openscad
ochafik has joined #openscad
ochafik_ has joined #openscad
ochafik has quit [Ping timeout: 256 seconds]
ochafik_ has quit [Remote host closed the connection]
<JakeSays>
there are times when I would kill for the ability to select an object in preview and move it around.
<teepee>
JakeSays: you know that you can move it around via clicking on the number in the translate and using ALT+Cursor or the mouse?
<JakeSays>
teepee: .. what??
<veverak>
z:D
<JakeSays>
teepee: dont tease me like that
<teepee>
just try it :)
<JakeSays>
you mean in the editor?
<teepee>
yes
<teepee>
e.g. translate([10, 0, 0]); click between the 1 and 0 and press ALT + UP
<JakeSays>
LOL that's almost useful
<JakeSays>
it moves in increments of 10
<JakeSays>
ohh interesting. it's based on the cursor position in the number
<JakeSays>
hmmm
<JakeSays>
teepee: well, that's just damn cool!
<JakeSays>
how does the mouse work?
<JakeSays>
ahh alt-scroll
<JakeSays>
teepee: well, that recovers about 87% of my sanity. thanks!
<JakeSays>
it works on any number in the editor!
<teepee>
for mouse there's some preferences to tweak the behavior
<teepee>
yes, it's not perfect, in some cases it's stopping, but it's still very useful
<teepee>
you can also add more decimals with ALT + Cursor Right
<JakeSays>
i dunno. in some ways it's much easier than using the mouse to position.
<JakeSays>
it's also the one feature that makes the editor more useful than an external editor. lol
<teepee>
the jump to source is also quite useful, although it has problems at least on windows :/
<teepee>
especially for inspecting code written by someone else
<JakeSays>
what's jump to source?////
stonkey has quit [Ping timeout: 240 seconds]
<teepee>
right mouse click in preview
<JakeSays>
doesn't do anything
ochafik has joined #openscad
Venular has joined #openscad
<JakeSays>
teepee: is there a magic setting or something?
ochafik has quit [Ping timeout: 268 seconds]
ochafik has joined #openscad
<Venular>
Hello, i ran into something interesting last night and am looking into some insight on why this happens. Here is the issue simplified (i have ran into this issue on something else I'm modelling but that program is a lot bigger and too much to share):
<gbruno>
[github] 3Dietrich edited issue #4059 (open .scad first time --> need to confirm "access to Documents" every time + for every Preset). https://github.com/openscad/openscad/issues/4059
ochafik_ has joined #openscad
ochafik has quit [Read error: Connection reset by peer]
stonkey has quit [Ping timeout: 240 seconds]
ochafik_ has quit [Remote host closed the connection]
TheCoffeMaker has quit [Ping timeout: 240 seconds]
<rapha>
i know the width and height at the midpoints, but how the heck do you figure out how to make the 4 "corners"? is it parts of circles? if so, what diameter? anyone good with such stuff and could give me some tips?
<rapha>
also the 4 sides look to me as if they're also not straight, so might also be parts of a circle, just with a much, much bigger diameter?
<rapha>
well, okay, the left and right sides do seem to have a straight part to them. so i'd need 2 straight lines, 2 very large-diameter circles and 4 smaller-diameter circles (but the the latter possibly in two different variants, 2 for the top corners and 2 for the bottom corners?)
<JakeSays>
teepee: yeah that's fine. go to source isn't something i need right now.
<Jack2247>
rapha: a super ellipse is sometimes used for that. I would use the image - convert it in inkscape into an svg .. import that and use hull to make something fit (or use the Superellipse() )
qeed has joined #openscad
<rapha>
"super ellipse", cool
<rapha>
i already like the name :)
Jack2247 has quit [Quit: Client closed]
<rapha>
will try both of the techniques you describe
<rapha>
it does have the flattened sides ... but the tops are also flat unlike in the picture. perhaps the person who originally designed it combined a superellipse with a circle or a normal ellipse?
<teepee>
although there were 2 where I did not see any difference, so it was 50:50 chance for 1 :)
<Jack22>
wtf is blurple
<Jack22>
(#5539cc)
<JakeSays>
LOL blurple?
<JakeSays>
InPhase: interesting list
<InPhase>
Jack22: It claimed a 2. The details are thin though. Apparently that is something like a top 5% score though from other sites. But maybe I just have a good display I'm using.
<Jack22>
inPhase randall munroe (XKCD) btw has a physics degree and worked for NASA
<InPhase>
It's not very rigorous when the display is not controlled.
<JakeSays>
Jack22: he does??
<InPhase>
Jack22: Yeah, I had forgotten his exact background, but the science training is obvious routinely in his comic content.
<JakeSays>
so i was going to take that color iq test but i failed when i discovered i have no idea what a hue color is
<InPhase>
Jack22: So the two colors that I couldn't tell apart in blue/green... I took a picture of that row with my phone, then took that image and brought it into gimp, and examined the numbers. I got two wrong under blue green because they are actually exactly the same displayed color on my screen.
<InPhase>
My overpriced camera phone cannot numerically distinguish them.
stonkey has quit [Max SendQ exceeded]
<Jack22>
0 points .. shouldn't wear filter glasses for the test - Ü
stonkey has joined #openscad
<JakeSays>
i only wear one contact
stonkey has quit [Max SendQ exceeded]
<Jack22>
InPhase my error also was in blue green (but had blue filter glasses on)
stonkey has joined #openscad
stonkey has quit [Max SendQ exceeded]
<InPhase>
I'm curious what causes the lower scores for others who are not clinically colorblind. It does require a lot of attention I guess, but I wonder if there are just underlying sensitivity differences to color.
<Jack22>
i programmed once a LED to light up in random color and while there are 256³ colors .. you normally can only differentiate 10 without seeing them side by side
<Jack22>
besides some people that are tetrachroma
<Jack22>
it has to do with naming of colors
<InPhase>
I do usually perceive small color changes well. I selected #f7faf5 for the background in that daylight gem colorscheme I posted above, because when I moved any of them up or down by 1, it didn't seem to hue align with the color of the model as well.
<Jack22>
LERA BORODITSKY showed that people with languages that have certain distinction between colors are also better in spotting the difference
<InPhase>
I don't really understand what that hue alignment is about, but, some just match to me and some don't, and moving by one number in a 24-bit value can often change it.
<Jack22>
and as women use colornames like "web_colors" and not like men "rainbow" .. they are better
<InPhase>
Jack22: My language is #000000 through #ffffff. ;)
<Jack22>
my door sticker says not everything is #000000 or #ffffff
<JakeSays>
Jack22: LOL that's cool
<Jack22>
InPhase i would assume that even if you would use that like .. oh this nice #1422AF sky today .. it is processed different to words and didn't form the categories you need
<InPhase>
There are artists who insist not every color fits into #000000 through #ffffff, but for some reason they never manage to take pictures of these other colors.
TheAssassin has quit [Remote host closed the connection]
TheAssassin has joined #openscad
<JakeSays>
i've run in to issues in the past with people who are color'ish'ers
<InPhase>
Jack22: More like, #1422AF sky! Run for cover and close the windows!
<Jack22>
oh there are colors like vantablack or metallic gold .. these can not be printed in CMYK and need to be put extra
<JakeSays>
vantablack is awesome.
<JakeSays>
i'd love to get my hands on some
<JakeSays>
but the guy who has the license is a dick apparently
<Jack22>
koonip capour
<JakeSays>
yeah
<JakeSays>
i think it'd be way cool to paint a boat hull vantablack
<JakeSays>
i think mit has developed a blacker black though
<JakeSays>
yeah him
<InPhase>
Jack22: Off the top of my head, I imagined a beautiful blue sky, and wrote down 6fb3f4. Seems to check out.
<InPhase>
I've poked at hex numbers for colors a lot, so I have a sort of mental guessing algorithm built up that does a decent job of getting close fairly quickly.
<Jack22>
did you know that the color name blue didn't exist too long .. before blue could be produced as a color the people had names like redish purple violett for water and sky
<Jack22>
and always think of the mantis shrimp that has 16 different color receptors
<InPhase>
I fine-tuned it in gimp, and got: 7fc2f4, which is a little better.
<Jack22>
and it looks different on different monitors and on a 10sqft carnvas
<JakeSays>
Jack22: you once told me that it is possible to change settings on only a part of a model in cura. how is that done?
<Jack22>
you need a geometry that overlap with that model and then choose "change overlap"
<JakeSays>
so like a cube of an appropriate size?
<JakeSays>
oh or an eraser
<Jack22>
if you want the change inside the cube yes
<Jack22>
no you need the "per model setting" above the eraser ..
<JakeSays>
if i draw a support blocker it'll allow me to change settings on it
<JakeSays>
right
<JakeSays>
i drew the eraser, selected it, then selected per model settings
<JakeSays>
then "modify settings for overlap"
<Jack22>
yes basically an eraser is an overlaping geometry that changes the support
<JakeSays>
can that geometry be used to change other settings?///////
<Jack22>
e.g. i have a ring as helper geometry so i can remove support circular while the center stays intact
<JakeSays>
these helpers are just stl files?
<Jack22>
right
<Jack22>
3mf - Ü
<JakeSays>
ah ok
<JakeSays>
hmm
<Jack22>
the first symbol make the geometry support .. the 3rd let you choose what to change and if only infill or every setting
<JakeSays>
right
<JakeSays>
and the 4th?
<Jack22>
that is the support eraser
<Jack22>
(first is normal , second is support)
<JakeSays>
ahh ok
<JakeSays>
they really should change the name of the 'support blocker' tool
<Jack22>
when using the 3rd you have to select the settings you want to change inside the overlap
<JakeSays>
right
<JakeSays>
so if i had a multi-extruder printer is that how i would change extruders?
<Jack22>
no for multi material you have seperate geometry and assign an extruder to each geometry
<Jack22>
in multi meterial you wont't have overlap
<JakeSays>
so that means you have to carve up your model?
<Jack22>
the 3mf or STL can contain multiple objects (not when exported via SCAD) you merge them so they get aligned and then select the extruder(material) for each part
<Jack22>
there are also printer that can color surface which works differently
<Jack22>
when you have multiple extruder you set this in cura and cura give you more options to use them
<JakeSays>
ah ok
<Jack22>
in theory you could use the overlap to not print the overlap and then add the overlap again to print with a different extruder ( not sure if that would work)
<JakeSays>
the center. it's too... much or something
<JakeSays>
i originally had just one support but that caused the circle to sag
<rapha>
JakeSays: oh wow, i hadn't noticed your later messages, sorry ... thanks for the effort! that's actually close enough to keep fiddling with it for a bit and then start a print before going to bed!
<rapha>
teepee: about which of all these were your many "no"s?
<JakeSays>
rapha: um, i don't recall having a conversation. perhaps you mean Jack22
<rapha>
InPhase: i think he's an actual physicist btw
<rapha>
oh, Jack already mentioned that
<rapha>
JakeSays: sorry, yes, i meant Jack22 ... i blame the tab key.
<rapha>
well, his name was Jack2247 when he said the things he said.
<JakeSays>
rapha: only two digits of precision matter with him
<rapha>
:)
<rapha>
Jack22: unfortunately i can't find any library that has the functions you used :(
<Jack22>
only in preview the render is fine .. the cube that making the half is scaled with your view ( if i use a big cube the view all option zooms out to far)
<Jack22>
rapha: r defines the size the n variables define the exponent for the rounding corner
<Jack22>
try something like n32=+3.0,n31=1.5,n3=2.5
* rapha
suddenly wishes he had taken math seriously in school
<Jack22>
exactly my thoughts .. teacher should use SCAD to show why
lastrodamo has quit [Quit: Leaving]
<rapha>
man if we'd done something like scad in school my life would have taken a different direction altogether
stonkey has joined #openscad
<rapha>
i see Halb() has many parameters. can it be used to cut the object in half along one of its other axes as well? usually i manually place a large cube and then do difference() for these things...
<Jack22>
yes that is what Halb() does (uses difference or intersection with a cube that is moved)
<Jack22>
Halb(i=0,x=0,y=0,z=0) i =0 is for one side i=1 the other ( and i just see that this modules has no help )
<rapha>
damn!
<rapha>
you just made a nicer one than what i had on my screen :-P
<rapha>
i was using linear_extrude(200) scale([2.5,3]) circle(r=10); and you did that Tz() thing
<Jack22>
Tz is just short for translate([0,0,z]) .. iam lazy as fuck
<rapha>
hmm
<Jack22>
i had T(z=10) but i needed a shortcut for that shortcut .. ( so you understand how lazy i am)
<rapha>
:)
<rapha>
now the only remaining problem is that the thing isn't vertically (with respect to the airplane's fuselage) symmetrical
<rapha>
at least https://bpa.st/ZNIA looks really good with regard to the sizes
<rapha>
and even fits on my printbed
<rapha>
hmm i'll make the walls a little thinner
<rapha>
basically i think i need two of them, an upper and a lower half
<Jack22>
not sure what you mean and also the n=3 is not what the shape from the image was
<rapha>
yeah, but i like your shape better
little_blossom has joined #openscad
<Jack22>
but i has to fit you fuselage or?
<rapha>
https://images.amainhobbies.com/images/large/pkz/pkz5126.jpg was the image, right? see how the thing is flatter on the top half and more curved bottom half and also the hole for the propeller axis is off-center vertically, more to the top?
<Jack22>
no the image from your model front
<rapha>
yes, it's symmetrical on the backside
<rapha>
but not on the front side where the opening for the motor axle is
<rapha>
different cowling, too, even more assymetrical, but makes the point
<Jack22>
n=2.5,n2=3.5
<rapha>
oh, right, your initial n values were Superellipse(+3.1,23.5,r2=27.8)
<rapha>
the 2D one anyways
<rapha>
yeah, got carried away :-/
<rapha>
i'll just print https://bpa.st/ZOYA with some cheap PLA and see how it looks in real life before going any further
<rapha>
also to find the spots for the screw holes
<Jack22>
you can keep the n3 n31 n32 values they dont change the base . only the nose
<Jack22>
and to make is asymetrical (which in theory could be done with different n31 values ) you need to use two halfs
<rapha>
yeah, still experimenting with that (two halfs)
<rapha>
but better first print and get the base right
<rapha>
lol poor little ender2
<rapha>
17 hours even at almost no infill and shit quality, talk to you again tomorrow i guess :P
<Jack22>
try spiral print .. or just print the first cm to check the shape
<rapha>
hmm, less waste, yes.
<rapha>
hmm seems cura doesn't really do that
<rapha>
no matter tho, i do also want to be able to judge the size and strength of it
<rapha>
already learned so much today already tho ... even just looking at the libraries on the openscad homepage, including yours, i'd never done before
<rapha>
i'll leave the print running until the morning and then interrupt it ... and hopefully on the weekend spend more time with your latest sketch, Jack22 ... not really understanding most of what's going on there yet and i think i need to just play with it a bit.
<Jack22>
i have skewed half of it and modified the n31 and n32 values to align for the half
<rapha>
JakeSays: a clock, eexcept it's not behaving like a clock :)
<Jack22>
oh btw for finer grid .. use fn=100 (superellipse has own fn value)
<JakeSays>
rapha: what do you mean?
<rapha>
Jack22: what's M()?
<rapha>
JakeSays: the seconds go back and forth instead of around
<Jack22>
multmatrix what does the skewing
<rapha>
o_O
<JakeSays>
rapha: lol it's a 5 second video
<rapha>
oh.
<rapha>
good point then :)
<JakeSays>
lol
<Jack22>
Superellipse(help=1) show the variables you can use fn =100 and fnz=100 (radial and vertical fragments)
<rapha>
Jack22: ok so then i kind of have an idea what you meant. still want to read the superellipse wikipedia page when i'm less tired and also take a look at those help outputs.
<rapha>
and even the n values for the base might need to be slightly different between the top and bottom halves, but my eyes are almost falling shut now
<Jack22>
n8 then
<rapha>
and then, with a plastic bag over the printer or something, this might actually be a good part to try with ABS
<rapha>
thx for the nice evening and all the new knowledge and g'night!
<Jack22>
gern geschehen
<Jack22>
btw the n and n2 value can be a list = [2.5,2,2.5,2]