<gbruno>
[github] kintel synchronize pull request #5219 (Merge the "render-colors" feature into the "manifold" feature, update lots of tests) https://github.com/openscad/openscad/pull/5219
adcurtin has joined #openscad
adcurtin has quit [Client Quit]
ooxoo has quit [Read error: Connection reset by peer]
<gbruno>
[github] kintel synchronize pull request #5219 (Merge the "render-colors" feature into the "manifold" feature, update lots of tests) https://github.com/openscad/openscad/pull/5219
ooxoo has quit [Read error: Connection reset by peer]
ooxoo has joined #openscad
drfff has joined #openscad
drkow has quit [Ping timeout: 248 seconds]
drkow has joined #openscad
deathonater has joined #openscad
Smeef has quit [Killed (NickServ (GHOST command used by deathonater))]
deathonater is now known as Smeef
drfff has quit [Ping timeout: 248 seconds]
misterfish has joined #openscad
splud has quit [Ping timeout: 260 seconds]
cart_ has joined #openscad
splud has joined #openscad
ferdna has quit [Quit: Leaving]
misterfish has quit [Ping timeout: 252 seconds]
cart_ has quit [Ping timeout: 252 seconds]
cart_ has joined #openscad
hyperair has quit [Ping timeout: 252 seconds]
ccox has joined #openscad
ccox_ has quit [Ping timeout: 246 seconds]
hyperair has joined #openscad
mohammad1722 has joined #openscad
mohammad1722 has quit [Client Quit]
mmu_man has joined #openscad
J24k19 has quit [Quit: Client closed]
J24k19 has joined #openscad
Alexer has quit [Ping timeout: 252 seconds]
teepee has quit [Remote host closed the connection]
teepee has joined #openscad
teepee has quit [Remote host closed the connection]
teepee has joined #openscad
mohammad1722 has joined #openscad
mmu_man has quit [Ping timeout: 248 seconds]
mmu_man has joined #openscad
mohammad1722 has quit [Ping timeout: 256 seconds]
cart_ has quit [Ping timeout: 248 seconds]
abff has quit [Remote host closed the connection]
germ has quit [Remote host closed the connection]
J24k75 has joined #openscad
mmu_man has quit [Ping timeout: 252 seconds]
J24k19 has quit [Ping timeout: 256 seconds]
mtm has quit [Ping timeout: 252 seconds]
mtm has joined #openscad
guso78k has joined #openscad
<guso78k>
kintel thank you for your answers. I tried this morning to run scripts/maxosx-build-dependencies.sh but it finished immediatley because everything was built.
<guso78k>
so i tried to run it with-f force, but this is not an option because it fails right in the middle, it compiles over and over again with -f losing a lot of time
<guso78k>
whats the best way to safely replace all hombrew libs with native builds, i also i do need libpng for libfive and libjpeg for the textures.
<kintel>
guso78 if scripts/maxosx-build-dependencies.sh finished because everything was built, that probably means that you already ran it once
<kintel>
homebrew builds and native builds are done in separate locations, so there isn't really a way to replace them; they should ideally be able to co-exist
<kintel>
You enable the native builds by running 'source scripts/setenv-macos.sh', otherwise, homebrew builds are used
<kintel>
There is a chance that bugs in the build system may cause the native builds (or OpenSCAD) to pick up a mixture of libraries though. If you experience that, please open a ticket. If those are third-party bugs it can be tricky to fix; in that case uninstalling the homebrew packages in question is probably the easiest way out
<kintel>
Likewise, any failures to build using -f should be considered a bug and reported
<kintel>
If you need new libraries, homebrew is likely the best place to focus, but you may need to fix qscintilla first : (
teepee_ has joined #openscad
kintel has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
teepee has quit [Ping timeout: 260 seconds]
teepee_ is now known as teepee
<gbruno>
[github] kintel synchronize pull request #5219 (Merge the "render-colors" feature into the "manifold" feature, update lots of tests) https://github.com/openscad/openscad/pull/5219
<kintel>
If anyone has outstanding test changes, be prepared for some conflict management after the above merge
kintel has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
krushia has joined #openscad
misterfish has joined #openscad
<InPhase>
kintel: Good to know. I was about to make some test changes on a local branch, but got sidetracked, aiming to pick it up again in a week or so.
<InPhase>
(on quite a number of tests)
guso78k has quit [Quit: Client closed]
misterfish has quit [Ping timeout: 248 seconds]
kintel has joined #openscad
<kintel>
InPhase ..and any ideas on how to simplify the test framework are welcome. It's pretty hard to refactor with so many tests : /
<kintel>
Also FYI: One of my upcoming commits will be moving Manifold out of experimental and making the backend selectable in preferences + cmd-line option instead. My hope is that tests shouldn't be too affected by that after this change.
<kintel>
I think moving out of experimental is the right thing to do now, and rather manage remaining bugs between now and the next release
<InPhase>
kintel: I'll ponder in the back of my mind while working on it. I was planning on putting up a PR for modernized defaults of $fa and $fs. But the first round of this failed out hard because of some of the larger stress-test entries in the testing suite, which did not appreciate those defaults.
<InPhase>
I was also going to update the default color scheme in the process, since so many tests would have to be updated that it's probably worth bundling these. My local branch name is "pretty_render".
<InPhase>
A bit of dev chaos would be introduced by this, but I really do think improved first impressions are going to be important.
<InPhase>
We're very fast now, and should also be prettier up front.
kintel has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
kintel has joined #openscad
<kintel>
InPhase Just be aware that it's very hard (for humans) to manually verify two changes at once for hundreds of images :)
kintel has quit [Client Quit]
guso78k has joined #openscad
guso78k has quit [Ping timeout: 256 seconds]
ferdna has joined #openscad
<InPhase>
kintel: Yeah.
<J24k75>
didn't we use a diff of the images to spot differences?
<InPhase>
J24k75: That doesn't help much if you change the color and resolution.
<InPhase>
J24k75: It'll mark it as all changed. :)
<InPhase>
Which it should.
J24k75 has quit [Quit: Client closed]
J24k75 has joined #openscad
<J24k75>
InPhase maybe if using an edge filter and only compare these - and resample for resolution .. but sure it is more tricky .. ask an AI Ü
<teepee>
AI? that thing that just shut down half the world today?
<J24k75>
but the faulty update wasn't AI generated
<J24k75>
or?
<teepee>
but yeah, it probably was some broken driver stuff trying to break into the OS to access data it normally can't
<teepee>
we'll never know though
<teepee>
still, at any rate I think it's a stupid idea to give an AI system the power to shut down your systems
<J24k75>
but sure just a question of time.. just had some funny discussion with chatGPT and first it says 18 days then i ask for the result in 18 days and how it explain that - then it apologized and admitted it was wrong.
<J24k75>
you can't be mad - also it wouldn't change anything
<teepee>
now imagine giving that thing full automated access to your technical infrastructure
<kintel>
The CrowdStrike issue really sounds like some manager thinking "it's just a single file, what can possibly go wrong if we roll out to 100% of the population at once, instead of the usual 3% every day for a month?"
<teepee>
yeah, that's one of the surprises. at that scale it even sounds too much work to roll out at once
<teepee>
but then, hello centralization
<kintel>
..and "I'm going on vacation tomorrow, I want to close this ticket" :)
<kintel>
Heh, I feel like the general consensus on LLMs around here is that they're terrible for most tasks, but amazing at _some_ tasks. You just need to know which ones
<kintel>
(I should say "engineer consensus", as leadership consensus seems to be the opposite)
dustinm` has joined #openscad
<J24k75>
hehe
<kintel>
If anyone has too much time available: We should really fix our transparency rendering. It's getting stupid. Especially with manifold color support, where manifold's triangle order not being stable...