<JakeSays>
so has there ever been a discussion on using occt for the preview rendering?
<peepsalot>
its been brought up before, but doesn't get any further than discussion. it would likely require a large number of dev hours and we're short on manpower
<JakeSays>
but would it work?
<lf94>
I don't see how it could work, I think it's fundamentally different?
<JakeSays>
that's what i'm asking. iirc occt is brep based
<JakeSays>
but man it sure renders nice
<peepsalot>
yes i think it is brep which is why it would be a huge change, but iirc for example FreeCAD has a plugin or something which supports some subset of OpenSCAD. i've never tried it so i don't know first hand, but apparently it isn't able to minkowski or hull, and maybe one or two other operations
<JakeSays>
i tried that plugin. hard fail
<JakeSays>
there's an openscad'ish system based on python that uses occt
<JakeSays>
what's the name of the toolkit currently used to render preview?
<peepsalot>
OpenCSG
<JakeSays>
ah that's right
<JakeSays>
hmm. project looks dead
<peepsalot>
seem kinda common for project which are done as an academic thesis. the original developers move on to other things
<lf94>
ok day 3: easy
<lf94>
thank god
<InPhase>
lol
<InPhase>
I thought you might find that one gentler. :)
<lf94>
I don't know the math which is used to determine the lengths of the cube you are using to intersect
<lf94>
but I'm eyeballing it
<InPhase>
There's a theory description in the comments, if you want to see that without the code. :)
<InPhase>
It explains some fractions used.
<lf94>
On the last ball
<lf94>
Then I'll look
<InPhase>
Coming up is a mix of easy, hard, and some that will probably explode the SDF engine. But we'll see. :)
<lf94>
Ok, done
<lf94>
Coloring
<lf94>
color(rands(0,1,3))
<lf94>
halp
<lf94>
Generates...random numbers between 0 and 1, returning 3 of them?
<lf94>
I'll have a "naive" and "refactored" version
<InPhase>
Yes.
<InPhase>
Day 7 will be interesting, because I admit I was also confused when I first saw the image for that one, and I had to look at the source to see how it was done. It presents the optical illusion of being much more difficult to do. :)
<InPhase>
lf94: So by "slow", was it prohibitively slow? Or would it have completed if you gave it a few hours to render?
<InPhase>
lf94: I have a suspicion that Day 24's, which took 45 minutes to render the animation of in OpenSCAD, could actually render in real time with an sdf approach. Some of these performance things will be inverted.
GNUmoon has quit [Ping timeout: 276 seconds]
arebil has joined #openscad
<lf94>
It was extremely slow at 100 unions
<InPhase>
Ah, only 100.
<InPhase>
Well I guess you might still be able to do tomorrow's with a little crunching along. :)
<InPhase>
Day 6 is where I will be very curious... It looks super simple and clean with no funny business, but if 100 unions give you trouble, then you'll have to think very carefully about how to design it to avoid this. I think it's possible to make efficient though with a wee bit of cleverness. I saw Inigo Quilez once use a technique that would work well for Day 6.
<InPhase>
I'll let you try it, then we can discuss, and then maybe that will inspire an efficient solution to day 2. I'll just leave this comment in the log as a reminder for myself. :)
GNUmoon has joined #openscad
arebil has quit [Quit: My keyboard has gone to sleep. ZZZzzz…]
<teepee>
yeah people trying to find out if it's possible to earn money that way. so far that has been difficult
<cbmuser>
knielsen: yay ;)
<cbmuser>
knielsen: did you get your GPG key added to the maintainer's keyring?
<cbmuser>
after that, I guess I have to give you upload permissions for openscad
<cbmuser>
and probably for lib3mf so that you can maintain the package together with teepee
<teepee>
not sure there's going to be an update for lib3mf, we may need lib3mf2
<cbmuser>
ah
<cbmuser>
teepee: btw, you are unreachable on Matrix again ;)
<teepee>
oh?
<teepee>
I guess it's just the notifications not really working :(
lastrodamo has joined #openscad
<knielsen>
cbmuser: yes, it's great! There's currently a request pending to get the key added to the maintainer's keyring, I assume it will be handled by the appropriate team as time permits, and then things should be ready
ochafik has joined #openscad
ochafik has quit [Ping timeout: 256 seconds]
ochafik has joined #openscad
ochafik has quit [Remote host closed the connection]
<teepee>
although a number are just the same curve formula over and over again making it more a tedious thing but still cool stuff
<Jack21>
nice but seems much more work as using a different program
<teepee>
that's part of the point I think :)
<lf94>
yep after reviewing them all, they are 99% disappointing
<lf94>
all just lines / curves with constraints
<lf94>
The Van Gogh and 3D scene are the most impressive
<Jack21>
and the car model is from scatchfab ( 1st in 18 yr)
<Jack21>
but i like that you can switch eyerything on and of
ochafik has joined #openscad
<Jack21>
reminds me to that guy writing over 3000 lines Arduino code for some art work when i told him about functions and loops so he don't need to repeat the same stuff on and on
<Jack21>
s/don't/doesn't
<buZz>
lol
<buZz>
'WHY DIDNT YOU TELL ME THIS 2 WEEKS AGO'
<buZz>
:)
<lf94>
lmfao.
JakeSays has quit [Ping timeout: 268 seconds]
<Jack21>
you would wonder this is normal for most people - at work i was wondering why they need 50× the time (i needed 10 min they several days) for making a simple spreadsheet - well then i found out that they filled every cell manually. No fill down and no formulas to calculate the content.
<Jack21>
it is like living in two parallel worlds
ochafik has quit [Ping timeout: 252 seconds]
GNUmoon has quit [Remote host closed the connection]
GNUmoon has joined #openscad
tcurdt has joined #openscad
<tcurdt>
I looked at the two open issues about the M1 builds - but I am not sure I read them correctly. Is there or isn't there a snapshot build yet?
ochafik has joined #openscad
ochafik has quit [Remote host closed the connection]
ochafik has joined #openscad
tcurdt has quit [Quit: Leaving...]
PovilasCNC has joined #openscad
ochafik has quit [Remote host closed the connection]
PovilasCNC has quit [Read error: Connection reset by peer]
Jack21 has quit [Ping timeout: 256 seconds]
Jack21 has joined #openscad
tcurdt has joined #openscad
<tcurdt>
since the github issue gets a little OT I thought I just show up here instead :)
<teepee>
.wb
<tcurdt>
so - you are concerned about having auth on the file server or at github?
<teepee>
write token on the file server that would allow access to the github repos
<tcurdt>
to iterated the releases?
<teepee>
to download the releases
<teepee>
iterating builds and releases via API works with a read token, that's fine
<tcurdt>
why create a github release and just mirror the build from there?
<teepee>
but I can't download the build artifacts, that needs some sort of full-access token as there's no dedicated setting for that
<tcurdt>
and you don't want the users to download the builds from github
<teepee>
that's just a huge amount of clutter, we want to build all branches
<teepee>
that would need yet another special cases for everything
<teepee>
no, because those build artifacts usually go away pretty soon
<tcurdt>
ah ... ok ... you mean builds in general ... not just releases ... ok
<tcurdt>
got it
<teepee>
releases needs some manual stuff anyway
<tcurdt>
well, you could have the ci push the artifacts
<teepee>
but the snapshots are normally fully automated, so I don't need to spent time on that on a regular basis
<tcurdt>
that's what I usually do
<teepee>
which means storing fileserver credential on github which is maybe a bit better but still annyoing
JakeSays has joined #openscad
<tcurdt>
I am not entirely happy with the security aspect of it either - but I've stripped that down. It's a special account that can only run exactly one command - nothing else.
<tcurdt>
and github now also supports oidc
<tcurdt>
which would be also an option
mhroncok has quit [Quit: Leaving.]
JakeSays_ has joined #openscad
JakeSays has quit [Ping timeout: 256 seconds]
<teepee>
I have to look up what that is later, need to finish some work stuff first...
mhroncok has joined #openscad
<InPhase>
Jack21: I used to run some lab sessions teaching very basic data analysis with a spreadsheet. And regardless of how many times or how strongly I emphasize "Don't manually calculate the numbers" and how many instructions I give them on entering the equations and the syntax for those, I always circulate the room and find somebody sitting there with a calculator typing out calculations and manually
<InPhase>
entering the numbers, which I usually discover because they are way behind everyone else.
<InPhase>
Jack21: At least by the end of the process everyone gets it and sees how its faster, because I designed the activities to be excessively tedious if you don't make the computer do the work. It's just fascinating how much resistance there is starting out to the mindset change.
<InPhase>
I started programming at 5, so the idea of expressing things in code got baked in early the way any natural language does. But when people encounter the concept for the first time as adults there's often an inertial wall up against it. Our schools need to be more uniformly integrating this into early curricula.
<Jack21>
haha i made that with pupils giving them 100 lines of date and it is an eye opener when they see that they can get analysis in just a few clicks for such a big list (for them 100lines is a lot) - but it is easy to forget how simple some minds work
<Jack21>
Ok that explains why your code is that amazing - Ü
<Jack21>
i always try to explain kids that math had nothing to do with numbers or additions - that it is the art of learning and building a mindset that is able to reasoning as it is an abstract of logical dependencies. And programming is just the next level above math with techniques to build more complicated structures to solve problems.
<Jack21>
who taught you this with 5? Your parents had computers (and knew how to use them)?
<InPhase>
It was 1985, and my father found some computer magazines in the store and got interested, so he bought a Commodore 64 and some programming books. Then once he figured out a few basics, he sat me down and taught me a few things. By the time we got to 10 PRINT "HI" 20 GOTO 10 I was hooked.
<InPhase>
But my questions rapidly outpaced what he had learned from playing so he just handed me the programming books and I read them.
JakeSays_ is now known as JakeSays
<InPhase>
Nobody bothered telling me that those programming books weren't for my age level, so I just pushed through and figured it out.
<Jack21>
i really wonder if that works for every kid - i am not so sure. And We ( by age) learned this in times were computers were not that idiot proof
<teepee>
not at 5 though, maybe 12 or so, not really sure
<teepee>
according to UK and BBC Micro, now Raspberry PI: yes
<teepee>
well not *every* but much bigger chance to get kids interested if they have the option
<InPhase>
Jack21: My mother aspired to be a school teacher, but at the time wasn't able to afford finishing college (she actually finished this when I was in college). But one benefit of that is she really dedicatedly taught me to read pretty young, so that helped a lot.
<Jack21>
when i got my first pc there was OS/2 installed what totally confused me as i only new windows/Dos from friends.. and had no one to help me with that
<JakeSays>
InPhase: how old were you in 1985
<InPhase>
5
<JakeSays>
bah
<Jack21>
Marissa Mayer ( ex yahoo CEO) once said - those things in live you do while not being ready for will giving you the most benefit for future development -
<InPhase>
There's a lot of truth to it I think.
<JakeSays>
so like running yahoo
<Jack21>
i am sure she learned a lot .. but she also was former GOOGL executive
<InPhase>
You might make a lot of mistakes doing it (like running yahoo), but in many cases only what comes after that series of mistakes gets worked out matters.
<JakeSays>
wonder what she's up to these days. haven't heard much about her since yahoo
<Jack21>
also not everybody is learning from their mistakes that much.
<JakeSays>
those who don't repeat them
<InPhase>
I sure wrote a lot of bad code as a kid. :) But I wrote things that worked, and rediscovered a lot of ways to not write code.
<InPhase>
That's a page with a lot of high promises and not really any explicit useful features described for solving the contact organization problem.
<JakeSays>
being the anti-social person that I am, contact organization isn't an issue for me. lol
<Jack21>
i am not trusting an AI company with my contacts
<InPhase>
I have far too many contacts with tons of duplicates and would love a tool to organize them. But it has to be useful to be worth my time to attempt, as I only contact a small fraction of my contacts.
<Jack21>
well .. just loose your backup and phone .. sometimes you have to rebuild from the ground
<Jack21>
JakeSays: maybe anti-social is a result of to many Muppets out there
<Jack21>
teepee how available was the Robotron? did you waited years to get one or could you only use them in schools?
<teepee>
quite expensive for normal people and generally not widely available, but I'm in Dresden which was the (or one of the) homes of Robotron, so they were sold in the "Centrum-Warenhaus"
<teepee>
A bit before I did send a postcard for the Z1013 kit which had a 1 year waiting list :)
<teepee>
IIRC it would have also meant driving quite some distance to get it, but in the end that never happened
<teepee>
brb, some shopping needed ;-)
<buZz>
oh dangit, i should have gone aswell
<buZz>
stores closed just 6 mins ago :P
arebil has quit [Quit: My keyboard has gone to sleep. ZZZzzz…]
<Jack21>
plastic can't have these low values without post processing
<JakeSays>
this one can
<Jack21>
looks like heat treatment on a hot plate .. how do they connect the leds ?
<JakeSays>
after they printed the left side, they just re-heated it with hot air and pressed the leds in to it
<Jack21>
well there is no magic so they need a binder and very fine powder - usually these filaments are extremely brittle
<JakeSays>
"Electrifi Filament is a non-hazardous, proprietary metal-polymer composite that consists primarily of a biodegradable polyester and copper."
<JakeSays>
sure would be a lot of fun to play with
<JakeSays>
but at it's price i'd have to make my own filament with their pellets to justify it
<Jack21>
they saying something about electroplating - this for sure makes it conductive