<InPhase>
lf94: What if you did dual color the easiest way possible, printing a plate full of card backs in black, maybe 2 layers, then one layer of white on the face, with the text left as holes?
<kintel>
InPhase I think I got clang with sanitizers now, but got carried away with GitHub Actions manual build GUI :)
<teepee>
just saying, the bird is dead, I just grumpily continue to use it for openscad stuff
<teepee>
well "for me" - just my totally personal perspective
<guso78>
ahh ok, did not make the twitter link in order to continue conversation there. i feel that libera is the better option.
<teepee>
that's fine, just saying, there's nothing that will be posted there, except after hell freezes over ;-)
<teepee>
so far I like Mastodon, no algorithm shoving stuff into timelines and such
<Scopeuk>
Mastodon is federated/decentralised?
<teepee>
yes
<guso78>
Don't know Mastodon. maybe i google it and see what pops up in my screen
<Scopeuk>
haven't seen a lot of it since the initial buzz with twitters first melt down. A bit like all of the communication tools it is only as useful as the other people/communicators on it. the local flooding/road issues etc here are all available on twitter but not mastodon to my knowledge
<ente>
I used to have an account but my problem remains the same as on twitter
<ente>
infinite scroll is bad for my ability to escape a website
<ente>
so I spend too much time reading random people's random thoughts
<ente>
maybe if I follow fewer people
<guso78>
teepee, are there any further traces, why the CI builds crash randomly, recently ?
<teepee>
Scopeuk: slow move still going on, remains to be seen what happens, but I like that a number of places decided to now run their own mastodon instances including heise (big IT publisher), ARD (our public funded tv/radio) and even government itself
<teepee>
plus the federated is not just mastodon, it's the "fediverse" with mastodon the more twitter like part, but there's others like peertube (like youtube) and pixelfed with focus on sharing pictures
<Scopeuk>
fair enough. decentralised with discoverability is probably the holy grail for this sort of stuff
<teepee>
I don't know if it's the best solution or if it will scale with 10 times the user base, but it's way better than giving all that power to a dickhead of a billionaire
kintel has joined #openscad
<kintel>
guso78 In terms of CI crashes, I'm working on custom CI build, including instrumentation for sanitizers. My hope is that some of those will give us some ideas: https://github.com/openscad/openscad/pull/4622
obriencj is now known as siege
kintel has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<guso78>
teepee, i like this channel. it makes me lookup words which are not yet in my vocabulary ... (its the referring one ...)
Virindi has quit [Quit: ZNC 1.8.2+deb2+b1 - https://znc.in]
Virindi has joined #openscad
peeps[work] has joined #openscad
<gbruno>
[github] Loki44 opened issue #4623 ( [manifold] Minkowski failed with error, falling back to Nef operation: CGAL ERROR: assertion violation!) https://github.com/openscad/openscad/issues/4623
<gbruno>
[github] Loki44 edited issue #4623 ( [manifold] Minkowski failed with error, falling back to Nef operation: CGAL ERROR: assertion violation!) https://github.com/openscad/openscad/issues/4623
teepee_ has joined #openscad
teepee has quit [Ping timeout: 240 seconds]
teepee_ is now known as teepee
Guest54 has joined #openscad
<Guest54>
hi?
<Guest54>
are there designers here? ones for hire that will sign an nda?
<teepee>
possible, you may need to hang around a bit so people can see the offer
<Guest54>
thank you teepee, that is good news. i will be back when i can spend more time here.
<teepee>
Guest54: alternative would be asking on the mailing list
<Guest54>
it is not an uncommon request?
<teepee>
not unheard of, but not happening very often
<teepee>
I suppose OpenSCAD is more used in hobby/private/free use
<Guest54>
understood. thank you.
Guest54 has quit [Quit: thank you]
<J2345>
before i sign an NDA i like to know what this is about Ü
<InPhase>
Same. A little background helps in assessing if it is an appropriate project to take on.
<InPhase>
Order of operations like: Brief description of roughly what the design is like without needing to disclose trade secrets, discussion of pricing models and price ranges, and then an nda could make sense for more details.
<InPhase>
I would never sign an nda for a $50 payout. :)
<L29Ah>
how about $55?
<InPhase>
I'm a physicist, so assume 1-3 orders of magnitude for the ranges of all numbers. :)
<L29Ah>
i hope you don't operate like this when writing openscad code :]
<InPhase>
I try to make them parametric across a broad range!
<InPhase>
People have taken my wall anchors and printed them like 6 centimeters in diameter to fill pipe holes... Still worked apparently!
<L29Ah>
oh so did you patch out the ridiculous resolution limit of openscad?
<InPhase>
I have for personal use.
<InPhase>
I made a version with 3 orders of magnitude better resolution. It's an easy change.
<InPhase>
Allegedly this is supposed to make things slower, but I didn't notice a performance difference.
<InPhase>
I'm not sure if anyone still knows what models this difference appeared with.
Ckat has quit [Ping timeout: 276 seconds]
<InPhase>
I did at the time notice the rate of random CGAL assertions goes down with the better resolution.
<InPhase>
Probably if we stabilize at Manifold we can just purge it, or make sure it's not used for any Manifold stuff.
rapha has left #openscad [WeeChat 3.7.1]
<teepee>
you can't *patch out* the recursion limit
guso78 has quit [Ping timeout: 245 seconds]
caveman has quit [Remote host closed the connection]
<teepee>
which has nothing to do with recursion limit
<teepee>
there are a number of things that need improvement / replacement
<teepee>
that is not possible with recursion limits
snaked has joined #openscad
greenbigfrog has quit [Ping timeout: 240 seconds]
peeps[work] has quit [Ping timeout: 246 seconds]
<InPhase>
teepee: "resolution" not "recursion". :)
<teepee>
ah, someone get me new glasses then
<InPhase>
Although the recursion limit can also be removed for Linux I think.
<teepee>
if you set a bigger stack it might work
<InPhase>
Linux has an unlimited stack option, which runs up to RAM size.
<teepee>
that 1 million limit should be replaced by a working cancel button at some point :)
<InPhase>
Well, RAM+swap if you want to get really crazy.
<InPhase>
Yes! A responsive interface would be a good modernization. :)
<teepee>
indeed
<InPhase>
Which is probably my fault. I've been repeatedly consulted in my day jobs for my experience making stable multithreaded gui programs, and I've done absolutely nothing about OpenSCAD on this issue.
<teepee>
talking of resolution limit, maybe a reasonable increase would make sense with todays hardware
<InPhase>
My first concern is that we should up the defaults for $fa and $fs before next release.
<InPhase>
Well, lower. Make better.
<InPhase>
I vote for $fa=1; $fs=0.4;
<teepee>
hmm, fixed or as config?
<InPhase>
Just new defaults. It's already changeable in every program that cares.
<InPhase>
I interpret not setting this as apathy.
<teepee>
or very rectangular design preferences
<InPhase>
lol
<InPhase>
I wouldn't really object to leaving it as a config option if someone feels super strongly about that, but I don't personally think it's necessary given the triviality of setting it in programs.
<teepee>
a separate limit parameter would be useful for people running server side stuff
<InPhase>
For example, command line can already set this with -D.
<InPhase>
If we really needed it before rather than after like -D maybe a general purpose -P for a pre-script context would be a bonus feature achieving the goal.
<InPhase>
That would have many uses, like setting defaults of values that could be overridden if and only if they're in a script. That probably has lots of pipeline usages.
<teepee>
and could be made sane by allowing only valid expressions
little_blossom has quit [Ping timeout: 276 seconds]