@Dots Hi, The crashing issue(Pd: signal 4) seems to happen on some Linux machines.
A couple of Linux users have already reported this issue and I'm planning to fix this as soon as possible.

The problem is I don't have the crashing issue on Ubuntu(16.04 and 17.10) installed on my Macbook Pro so I'll need to try this on other device to confirm/fix the bug.
I'm planning to try it using my old Acer laptop next week.

By the way, I have a user who also reported the crashing issue on Ubuntu 17.10 and he told me he could fix it by selecting NVIDIA GPU and rebooting so maybe you can try this if you haven't.

Thank you again for reporting this and I will update you as soon as I fix the bug.

The brand new Camomile release (1.0.1) is available!

Camomile is a plugin with Pure Data embedded that offers to load and to control patches inside a digital audio workstation. The plugin is available in the VST2, VST3 and Audio Unit formats for Linux, Windows and MacOS. Download and information on the website: https://github.com/pierreguillot/camomile/wiki.

Read carefully the documentation. Feedback are more than welcome!
For feature request and bug report, please use the issue section on the Github repository if you can:
https://github.com/pierreguillot/camomile/issues.

An overview of the main changes:
v1.0.0

• Use libpd instead of my personal wrapper.
• Use TLS approach of Pd to manage thread concurrency issues.
• Use a text file to define the properties of the plugins.
• Generate plugins with the patches included.
• Separate the GUI and the parameters' definitions.

v1.0.1

• Fix thread concurrency issue that occurred when selecting a program (#77).
• Fix stack overflow issue due to concurrent access to the Pd's stack counter (#69).
• Update documentation for VST2/VST3/AU generation on MacOS to display the name of the plugins in Ableton (#75).
• Improve the whole documentation (#72) and start "How to Create Patches" (#73).
• Add more warning when there are extra arguments in parameters' methods.
• Add support for "openpanel" and "savepanel" methods.
• Update examples Bulgroz, AlmondOrgan, Castafiore, MiniMock.
• Start/Add support for patch description in the text file (#74).
• Start/Add support for patch credits in the text file (#74).

Further information on the previous versions and the changes on this topic.

Hello all,
I writing a patch that uses a simple shell Applescript app I made that closes a pdf file and opens a new pdf file in Preview mode on an Apple.

When the pd switches to the Applescript to close and then open pdfs, there is a digital click and pause in the smooth running of the program. Because this is obviously undesirable, I am wondering if I can do something to alleviate this problem.

Is it the Shell command?
Is it the nature of using Shell?
Is it that the computer leaves pd as the active window for a brief moment?
Is it that Preview is closed then opened again?

I can add the patch I created if that is needed to get to the bottom of this, but am wondering if there are any quick fixes to expand the dedication of the CPU to pd to avoid digital artifacts in performance.

Hello all,
My laptop crashed. When I changed to my new (Well, older laptop) I tried to load on everything but there are some issues. Any help with these would be greatly appreciated.

For this one piece I was working on, I was using pd-extended. Yes, I know it has been obsolete for a few years now. I switched to vanilla for a new piece I was working on.

Now I am back to the other piece I was composing.

A couple of frustrating things are happening. I was trying to use pd-extended and using |counter| and send the number of counter to }pack} but there was an error that stated that the value expected a float but received a bang.

Then I opened my patch in .047-1-64bit.app. and seeing some things that I am am trying to be most efficient in fixing. Here are a couple of them:

|unsig~| does not work in "regular" pd? Is there a different object I should use? Is there really no equivalent or am I doing something wrong? (I was using in pd-extended |unsig~ 50|

|Average| does not work in "regular" pd? Is there a different object I should use? Is there really no equivalent or am I doing something wrong?

Maybe a general question: What should I do? Start over? Recreate my patch? Try to fix everything in the old pd-extended? Go to the most recent version of pd?

Thank for any and all suggestions,
William

|pvu~| could not be created either. Again, am I doing anything wrong?

source : https://twitter.com/hellocatfood/status/932981517189632012

we really want to have proper w32-builds back (using VisualStudio) -> https://www.bountysource.com/issues/3383537-w32-build-msvc

since there is no QuickTime on OSX/64bit, we need a new video-capturing backend, probably based on AVFoundation -> https://www.bountysource.com/issues/4185679-native-video-capturing-on-osx-64bit

since there is no QuickTime on OSX/64bit, we need a new film-reading backend, probably based on AVFoundation -> https://www.bountysource.com/issues/4185729-native-film-reading-on-osx-64bit

since there is no QuickTime on OSX/64bit, we need new imageloader/imagesaver backends, probably based on AVFoundation. -> https://www.bountysource.com/issues/6492784-native-image-reading-writing-on-osx-64bit

@whale-av Yes as I mentioned I tried all those ancient tricks on puredata.info and no dice. Some of those apt-get package suggestions have been replaced in Sid so there's some ambiguity there but there's reason to doubt that's the issue because it's not just one distro or desktop environment (and yes, I even messed with all the desktop font configs I could find).

Did you read those two messages? I thought these passages spelled it out the issue rather succinctly:

When I switched fromUbuntu 14.04 to 16.04, boxes grew horizontally by one pixel per character for patches with fontsize=10.

When you switched to 16.04, you started using an updated font
engine to render the fonts. As far as I can tell there is no set of rules
or constraints on the font stack developers to keep font sizing algos
consistent across time. I do know that Windows, OSX, and the old
Gnu font stack sizing algo (i.e., 14.04) were within a pixel of each
other when measuring the width of a 31-characterfor string of
DejaVu Sans Mono at a particular pixel size. The newer Gnu font
stack rendered a string that was seven pixels wider at the same
pixel size.

and:

On further digging, I think the culprit is a slightly different tk scaling value:
Tk 8.4 -> scaling: 0.999016715831
Tk 8.6 -> scaling: 0.9990167158308751
Looks like a precision/rounding issue after all w/ 11 digits versus 17.

There's more later when they start talking about which app should be fixed Pd or tk, parameters to change before compiling to defeat the rounding, how to identify this specific error as opposed to all the other font problems etc. but no definitive solution yet.

It should be noted that this is a Vanilla-only issue. All the details on the other Pd distros relative to this problem can be found in that thread.

Hello,

I migrated from Pd Extended to Vanilla (0.47-1) a few days ago. I've tried to familiarize myself with the handling of externals and the Deken manager, which seems to slowly become more understandable the more I work with it.

But today I've run into some odd issues I never experienced with Pd-extended.

– First of all, the console is a mess of "tried to load, failed" and "tried to load, succeeded" at every start up, which is understandable, but I can't seem to figure out what to take from it. Sometimes I get both the "failed" and "succeeded" version of a message one after the other, regarding the same library/external. Sometimes it will claim to have failed trying to load a given external, but I can call up the objects just fine without having to [declare] or use the [name/*object] initialization.

I would assume that I have made too much of a mess trying to add the paths to "path" and "start up". So I guess it's trying to look for certain things in subfolders where it's not? The issue is that everything seems to load up correctly, but I still get a ton of misleading console messages.

– This one is odd. After having loaded up Pd and scrolling up through the console's many, many messages, I will get to the "credits" message of the zexy external, but it will lag and freeze for a moment before reaching it, sometimes multiple times before allowing me to scroll by it smoothly.

– I tried to continue work on a patch, a 4 channel random sample player. The gui elements will sometimes freeze after pressing play and running everything, which renders all gui elements unresponsive, though I can still operate the patch, but just with no update. From here on out I can't close any patches or windows, nor can I go into patch mode, I simply have to close down Pd and open it again.

I understand that many gui elements can be taxing, but I really haven't got a lot of them going on. I have worked on patches far more graphically taxing in Pd-extended without any problems what so ever. In fact, I've never had any issues like this with Pd-extended.

There is some [metro] objects banging at 100 ms intervals to update the 4 individual audio arrays, but stopping those didn't seem to help a whole lot.
Everytime a sample has ended, it will bang and choose a new sample for immediate playback, sometimes the whole patch system seemed to stop, seemingly as if not being able to handle the stream of information. Again, no issues in Pd-extended.

I hope you'll be willing to help me out here. I suspect that the first order of business is to clean up the "path" and "startup" menus to clear up the system, but I'm unsure about how I would go about that.

Thank you
M.

Purr Data is finally released!
Purr Data inherits the goodness of Pd-l2ork and runs on Gnu/Linux, Windows, and OSX. Infinite undo, enhanced editing and 2d drawing, and most of the the externals from Pd-extended (plus more from Pd-l2ork).

[Edit]
2.2.4:

• fix [wrap~] inconsistency
• port mouse events from [hcs/cursor]
• fix bug with dollarsigns in struct names
• add some "legacy_mouse" classes to support externals that report mouse events: "legacy_mousemotion", "legacy_mouseclick" and "legacy_mousewheel"
• fix bug loading default libs on OSX

2.2.3:

• fixed overallocation bug that made loading soundfiles into arrays take up unnecessary amount of memory (4x-8x overallocation)
• fix some utf8 string handling routines
• port fixes for [soundfiler] from Pd Vanilla
• fix bug with "$@" that could cause invalid reads in some cases
• fix [initbang] functionality
• fix crasher in [pdinfo]

New in 2.2.2:

• fixed usability issues with the dropdown object

New in 2.2.1:

• fix stray GUI bugs
• fix some external libs that weren't building under Windows
• fix some OSX key event bugs

New in 2.2.0:

• fix GOP sizing algo to accommodate small GOP abstractions
• turn off anoying Gnome-keyring popup that appears on some systems
• bump nw.js GUI toolkit to 0.22.1
• fix arrow-key navigation-- use arrow keys for scrollbar navigation in runmode, turn off in editmode
• ignore iemgui label in bbox computation for "-legacy" mode
• fix OSCx library on arm builds
• guard against more out-of-order messages from Pd to GUI
• clean up and simplify build instructions
• leave enough space in GOP abstractions to accommodate all xlets
• fix search path bug
• update some of the documentation, removing old "messageoddness" patches that no longer apply to Purr Data
• update unauthorized library
• clean up and simplify the repository code/structure

New in 2.1.2:

• minor bugfix in the console (duplicate messages weren't printed after clearing)
• some minor fixes in the default startup configuration
• some cosmetic changes in the accompanying README and license information
• various bugfixes in objects and internal API functions (message boxes, dropdown atom, loadbang-related issues)
• updated lyonpotpourri to the latest 3.0 version to fix various 64 bit issues on Linux and macOS
• saving the preferences is much faster on the Mac now
• new animated "About Pd-L2ork" popup; also, the proper version is now shown in the generic "About" box on the Mac

New in 2.1.1:

• normalized range for [bendin] (keep old behavior under legacy flag)
• added "mouseenter" and "mouseleave" events for data structures
• fixed "Recent Files" under Windows
• cleaned up documentation in repo
• fix for [midiclkin] (#255)
• added "l2ork_version" message for [pdinfo]
• make loader search order the same as Pd Vanilla
• fixed cord inspector font size
• silence spurious error when autopatching a signal object
• added a [dropdown] object for choosing a value for an atom box (interface not stable yet)
• made gatom resizable by click-dragging in edit mode
• added "<ctrl-mousewheel>" for zooming
• added solarized and inverted solarized gui presets
• fixed mycanvas stroke color updates
• improvements to mode 4 of intelligent patching
• fixed "<Delete>" not deleting a selected object on some systems

New since rc5:

• fixed display bug with [vu] on graph-on-parent canvas
• fix documentation for timer-help.pd
• fix "open" method for [ggee/image]
• default to the user's home directory on OSX app bundle
• open the help browser file browser in the doc folder
• add a canvas "Print" menu item (under "File" menu)
• allow option to save the zoom level with the canvas
• fix a data structure crasher

please report lots of bugs to

https://git.purrdata.net/jwilkes/purr-data/issues

Binaries:

https://github.com/jonwwilkes/purr-data/releases

Whoops! Missed this post, just found it by accident. Don't know if the OP gave up but might as well respond for the archive and mention that I finally filed a bug on this issue.

The known problem here -- as I have stated in other threads -- is probably not a midi clock issue but Active Sense messages (the Yamaha PSR-E323 midi implementation chart indicates that these are implemented in that device). Active Sense is part of the System Realtime group of commands in the midi spec (also includes clock, stop, start, and continue).

If configured correctly, Midi Patchbay on the MAC will stop these messages, but it can only be configured to block the entire System Realtime group. If the other messages are needed, a more selective option might be Midi Pipe, thou I have not yet tested this. It can be downloaded from here:

https://www.macupdate.com/app/mac/10541/description

See also these threads:

http://forum.pdpatchrepo.info/topic/9565/midi-issue

http://forum.pdpatchrepo.info/topic/9501/filter-out-midi-clock-information

http://forum.pdpatchrepo.info/topic/9635/does-midirealtimein-still-work

I'm only starting to deal with AVL2016 myself -- it's a completely new platform for Glen & the gang and there are still a few issues they have to work out with it before it will be up to par with previous releases. This is why I still haven't deleted my "Diehard" install (still good for audio/midi).

I can say that with "Diehard" and the other previous Debian Squash AVLs Vanilla always compiled, installed and worked without a hitch. I never installed L2ork in AVL (I have a Ubuntu/KX Studio partition for that and I'm mostly into Vanilla anyway) but others on this forum had no trouble with it after replacing the jack module. Neither Pd version was affected by the absence of Pulse Audio.

As I see it, the options for AVL2016 and Pd are this right now:

--Vanilla: Now includes the deken extension finder as standard, so adding all the libraries you need from Extended is much easier. Problem so far: I tried both compilers in 2016 and still get failed graphics renderings (i.e. gui bangs show up as blank boxes etc). Automatic loading of the system font does not work and I can't fix the issue with any of the font startup flags. This causes tightly-spaced patches to get messed up. There is another Sid-based multimedia distro -- i/o linux -- that has the latest Vanilla release in the repos. On i/o linux the graphical errors are not present, but the font issues are the same. I have not yet tried installing that repo package to see if gives the same result on AVL 2016.

--L2ork: Options as described in my last post OR you could also try the new PurrData version that is replacing it soon. It's not quite ready yet, but close. It is possible that the graphics issues are related to the outdated GUI and since the point of PurrData is to replace this, it may be a way to get past those issues. There's posts on here with links to the RC4 PurrData test release, thou you may have to compile it as I don't see any pre-compiled versions for Sid.. There is still the issue of the jack dependencies.

Let me know how you make out as we are in the same boat at the moment.