<kintel>
J24K I think the "stripes" can come from two places: 1) our vertex quantization or 2) floating point inaccuracy in Manifold. I'd explore 1) first. Do you see any warnings of the kind we discussed earlier? That warning will trigger a vertex quantization
pca006132 has quit [Remote host closed the connection]
pca006132 has joined #openscad
pca006132 has quit [Remote host closed the connection]
pca006132 has joined #openscad
<J24k>
pca006132 for the vertical artifacts/edges - as iam not sure why they occur it is hard to replicate it is an difference with an oval shaped ring.
ccox has quit [Ping timeout: 246 seconds]
kintel has joined #openscad
<kintel>
J24K Could you share the design, as .scad or .csg ?
<J24k>
I am just trying to replicate this on a minimal case. I think there are vertices very close together
<J24k>
I could get something by using roof but the original design didn't use roof at that location
<teepee>
kintel: statement on the qscintilla list: "QScintilla is effectively in maintenance mode - I will keep it working with new versions of Qt but there is little incentive to do a major upgrade."
pca006132 has quit [Remote host closed the connection]
pca006132 has joined #openscad
<J24k>
kintel if i render only the part where the issue occur - it doesn't
<kintel>
Meh, of course the build fails. This is why I wanted some overlap time.
<kintel>
J24k I can reproduce the lines. Will look into it
noonien808310429 has joined #openscad
<teepee>
yeah, not sure if I missing some notificaiton mail, I only saw it yesterday as banner on the website when looking at the build-fail notification
<kintel>
Yeah, there is also a missing notification..
<kintel>
I think it used to go to the old IRC channel, and CircleCI somehow messed up the IRC config
<kintel>
J24k I think the root cause of the artifacts is 32-bit float inaccuracy
<kintel>
You have two linear_extrude(height=1000) in your design, which expand the range of the object so far that 32-bit precision isn't enough to represent details. If you rewrite those, the artifacts go away
<kintel>
pca006132 Good case for double support :)
kintel has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<teepee>
hmm, latest Scintilla has some sort of Qt support
jan64 has joined #openscad
<jan64>
hm, bit of a random question - is there a way to access the current time / date in an openscad script?
<jan64>
trying to do some prototypes and i'd want to stamp the 'version number' on each one
<jan64>
found a github ticket from like 2013 that has some discussion, but that seems to still be open :P
<teepee>
without external help, no, not possible yet
<jan64>
ah well
pca006132 has quit [Remote host closed the connection]
<jan64>
i guess i'll keep incrementing a version string before sending it to the printer
pca006132 has joined #openscad
<teepee>
if you always update to the latest dev snapshot, that version string would be an approximation ;-)
teepee has quit [Quit: bye...]
teepee has joined #openscad
marlasinger has joined #openscad
ccox has joined #openscad
marlasinger has quit [Client Quit]
gunnbr has joined #openscad
othx has quit [Remote host closed the connection]
othx has joined #openscad
pca006132 has quit [Remote host closed the connection]
pca006132 has joined #openscad
neothechosen has joined #openscad
<J24k>
kintel 1000 is not much and the problem is that i don't know how big an object is so a steep angle could easy lead to objects that are bigger 1000
<J24k>
But yes reducing that solved the problem
<Scopeuk>
With floating point it's the difference in scale that causes issues rather than the absolute size
<Scopeuk>
At 1000 scale you minimum step should be .00006103515625 but that's a single bit and could easily be lost else where, if you had. Arad -> degrees conversion etc
<J24k>
it is the object that is cutting
<J24k>
You can't determine the height of a roof operation
<J24k>
https://imgur.com/a/026g4mT this happens if that object is not big enough .. but i may switch back to intersections
kintel has joined #openscad
<kintel>
J24k It's possible something else is going on too, but I slimmed down the model to only basic operations and it still failed
kintel has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
Guest17 has joined #openscad
Guest17 has quit [Client Quit]
teepee_ has joined #openscad
teepee has quit [Ping timeout: 260 seconds]
teepee_ is now known as teepee
teepee_ has joined #openscad
teepee has quit [Quit: bye...]
teepee_ is now known as teepee
mmu_man has quit [Ping timeout: 264 seconds]
pca006132 has quit [Remote host closed the connection]