Ok, earplug~ 0.3.0 is ready:
https://github.com/pd-externals/earplug https://github.com/pd-externals/earplug
Lucas (or any Windows dev), if you have time, could you make Windows builds for deken? Then Joao can give it a try...
On Jun 2, 2021, at 5:04 PM, Dan Wilcox danomatika@gmail.com wrote:
Nothing too major.
We haven't made a changelog yet but the changes in the current main branch form the last 0.2.1 tag (2013) are:
- update to build via pd-lib-builder
- loading the dataset txt file is now optional (needs double-checking)
- fixed hang when dataset file is not loadable
- post header only once on load and include version
- avoid crashes by making sure that the "dsp" method cannot be called from the patch
Some future optimizations in the pipeline are: https://github.com/pd-externals/earplug/pulls https://github.com/pd-externals/earplug/pulls
I will try to add a changelog and new stable 0.2.2 tag tonight. You could then bug a Windows dev to make a build for deken...
When we review the other optimizations, they would go into a 0.3.0 version.
On Jun 2, 2021, at 3:56 PM, pd-list-request@lists.iem.at mailto:pd-list-request@lists.iem.at wrote:
Message: 3 Date: Wed, 2 Jun 2021 13:33:11 +0100 From: Jo?o Pais <jmmmpais@gmail.com mailto:jmmmpais@gmail.com> Cc: Pd-List <pd-list@lists.iem.at mailto:pd-list@lists.iem.at> Subject: Re: [PD] Binaural w32? Message-ID: <3ffaeb84-291c-7e2e-981e-a1ca531f6dfe@gmail.com mailto:3ffaeb84-291c-7e2e-981e-a1ca531f6dfe@gmail.com> Content-Type: text/plain; charset=windows-1252; format=flowed
that is nice - although I can't really build anything in windows on my own. what are the improvements compared to the 2009 version?
I updated the earplug~ external to build using pd-lin-builder, so it should be easy to build on Windows:
https://github.com/pd-externals/earplug/tree/main https://github.com/pd-externals/earplug/tree/main <https://github.com/pd-externals/earplug/tree/main https://github.com/pd-externals/earplug/tree/main>
It's not the best but we use it for a project at work. It can use some additional optimizations and the dataset is for 44.1k so you are in luck. For our project, we are likely to introduce resampling for other sample rates so it's more accurate beyond 44.1k...
Dan Wilcox @danomatika http://twitter.com/danomatika danomatika.com http://danomatika.com/ robotcowboy.com http://robotcowboy.com/