ok - totally noob-ish question -- so sorry! but ... I found pd about 4 years ago; used pd-extended then; and now I'm back with Andy Farnell's book in hand where he talks about "the enormous range of objects available in pd-extended" only to discover that extended is "discontinued" . So I installed the latest vanilla but I'm really not sure how to get the objects that I guess just aren't in vanilla?! Or even if I need them?! Anyone got any advice?

Since Pd-0.47, the deken-plugin is included into Pure Data itself. Deken, a minimal package management system for Pure Data externals, can be accessed from the Help -> Find Packages menu. To know more about Deken, see https://github.com/pure-data/deken.

Regarding the "or even if I need them" question, there is no real answer to that. Personally I like to try and stick to Pd Vanilla as much as possible, to make my patches as future-proof and as portable as possible. However, there are certain things that cannot be achieved with the vanilla version alone. So it all depends on the goal and scope of your project!

ok - thanks -- and I found the answer to my next "how do you decide who to trust" question on the github link. But I guess I need to download the package to find out what objects it contains?

@hayasidist This list might help, but usually you install an external library if you know you need the contents of it.

Thanks! very helpful list! -- but it's a bit chicken and egg - if I know only conceptually what I want (e.g. draw a square) I'd need to find a library that might have something useful -- recognising that the list you pointed me at does seem to give the source library, it would still be useful to have the ability to scan the contents list for libraries before installing them

@hayasidist One more thought for you.
If you absolutely need a particular group of externals then you might, if it has never been re-compiled for 64-bit for your system, need to run at the very least a 32-bit version of vanilla.
This will be particularly problematic on recent osx and for RPI stretch.
Often you will know because they simply will not be listed under Deken for your system.
Not always though. For example it is possible (not easy) to run GEM for 64-bit Pd under windows, but I don't think you will get it through Deken.
David.

@whale-av Thanks.. still slowly getting back into this after a 4 year lay-off....

@hayasidist If you want an experience somewhat like Pd-extended you might give Purr Data a try:

https://git.purrdata.net/jwilkes/purr-data#downloads

It ships with nearly all the external libraries that came included in Pd-extended.

As far as "who to trust"-- the source code for the externals in Purr Data was forked directly from Pd-extended. All changes have been merged in during our regular release process where we test for crashes, memory errors, etc.

@jancsika Thank you -- I stumbled across purrdata in this thread -- https://forum.pdpatchrepo.info/topic/10349/drawing-objects-simplified/2 -- at the time I ignored it (one more thing to be confused about!) but your link to git has clarified a lot for me. And the reason I got to that thread was to look into pd's drawing capabilities because ....

I'm looking to pd (+/- variants) to be able to help me create sound and video from something like a "piano roll". Ideally the piano roll has "only" the music but this is interpreted as both sound and video. The end point is to have a graphic score that can be "played": e.g. https://www.classicfm.com/discover-music/latest/graphic-scores-art-music-pictures/bergstrom-nielsen-towards-unbearable-lightnes/ (but I'll create the score from the music and then make it look as though it's being played!)

In the process of looking further into this I found this that was done with pd: https://upload.wikimedia.org/wikipedia/commons/5/52/Solitude.png which is close, but closer to a traditional score (it's time-based) than the more abstract notation I was aiming at.

@hayasidist Your second link looks very much as though it uses data structures.
It is not an easy concept, Chapter 2.9 here........ http://puredata.info/docs/manuals/pd/x2.htm might get you started. Then work your way through this...... tutorial_scalars.zip
There is more help in the Pd doc folder....... \pd\doc\4.data.structures
Take a look at 07.sequencer.pd in that folder for more.
Actually, looking at that patch first will give you more inspiration.
Data structures are a steep learning curve, but worthwhile, and maybe the best route to what you are looking for.
David.

thanks. I started with 07.sequencer.pd -- I'm beginning to understand the relationship between top-level template and the stuff that follows it... not quite sure yet how pd knows about x, y, width for the amp and voice templates ... but I'm working on it!

@hayasidist Below is an example patch that uses data structures with the Purr Data [draw] object to render part of the "Unbearable Lightness" score...

lightness.pd

As you can see, each drawing instruction can be handled separately, animated, etc. I'm just using a quick-and-dirty way to render by using a lot of [draw] objects with a single scalar. But it's also possible to use multiple scalars with variables for colors, points, etc. Or even a scalar with an array field.

@jancsika sorry for my noob question (the file threw a hissy fit when I gave it to pd vanilla) -- do I need purr data installed or is the draw object something that I can grab from somewhere and "install" for pd vanilla?

It's a core part of Purr Data so you'd have to use Purr Data.

ok - thanks.