Skip to content

Conversation

@etbusch
Copy link

@etbusch etbusch commented Jul 8, 2023

This PR uses the std python concurrent.futures module to generate individual glyphs in parallel (for as many threads as there are on the machine) in order to speed up total generation time, as openscad is notoriously single-threaded and very slow.

It shortened generation time on my 4-core-8-thread laptop from ~15 minutes to under 3.

@settinger
Copy link
Owner

Very cool! I am going to test this out as soon as I can. What version of OpenSCAD are you running? The most recent stable version (2021.01) or one of the nightly builds with fast-csg ?

Also, if you don't mind me asking, when/how did you find the typeball project? I hadn't given it much thought in ~3 months and suddenly this week it's seeing a bit of attention, ahaha!

@etbusch
Copy link
Author

etbusch commented Jul 8, 2023

I am running OpenSCAD version 2021.01. I was unaware of improvements in the nightly versions, and will check that out too.

Also, if you don't mind me asking, when/how did you find the typeball project?

I read the Hackaday article back in April and am just now able to access a friend's resin printer, so I pulled it down and started playing with it. No idea why anyone else is just piling on now, but it's an interesting coincidence!

@Hubcapp
Copy link
Contributor

Hubcapp commented Jul 8, 2023

I found it this week because I just recently got my typewriter repaired, and was already interested in the concept of 3D printing a selectric typeball. :-) I have to agree it's just an interesting coincidence and not an algorithm surge.

P.S. @settinger You'll be pleased to learn that somehow the alignment on my typewriter magically fixed itself. I couldn't find the belt that Dave mentioned, but he must have been right that some belt got kinked, and I guess un-kinked itself after a few days of use.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants