Hey guys,

I put 10.5 on my mac a few weeks ago, and since then Pd-extended's behavior has been very erratic.

When I save patches, Pd crashes.
When I load the documentation browser, Pd crashes.
I really can't do anything.

Has anyone seen similar behavior? The new 40.3-extended nightly builds are not any better.

It does not always crash upon first save, but most subsequent saves cause a crash.

I've attached my osx crashlog as txt. Any insights?

http://www.pdpatchrepo.info/hurleur/pd-crashlog.txt

Hey guys,

Is there any quick way to grab the size of an array? I can't find this documented anywhere in the Pd docs. The benefit here would be it would be much easier to build a generic sampler that could loop any sample of arbitrary length and not have to set that manually.

This is the worst question ever but stay with me for a sec. Is there any way to dress up the Pd windows in linux? They look really sexy on Mac OS X. For example, compare this Netpd screenshot from wikipedia to my attached Pd window in mac os x.

http://en.wikipedia.org/wiki/Image:Screenshot10.png
vs.
attachment below.

I guess what I mean is can you remove the X scroll bars for Pd windows?

http://www.pdpatchrepo.info/hurleur/pd_mac.png

Hey all,

New processors are crazy fast, and ram is cheap. Anyone can get a system with a dual core 64bit processor and 4GB of ram for under $1,000. The question I pose to all of you is, why is 64bit development for linux lagging so far behind? I can't even get Ubuntu Studio for a 64bit box.

Has anyone here had success compiling Pd (and/or) the extended packages for a 64bit Linux environment? If so, I'd like to compile everyone's feedback and post a procedure for others to follow. My intention is to support Pd x86_64 builds. Thanks!

Hey Guys,

I've noticed, consistently, that if I select my MOTU 828mkII as both Audio In and Audio Out in PD .39extended (newest build from Hans in daily builds, RC4) my CPU usage jumps to almost double as compared to CPU usage with my Built In audio as Audio In and Audio Out. I have not seen this behaviour in any other Mac OS X application.

I've found a way around this.

If you're a Mac OS X user, go ahead and download Soundflower from Cycling74's website: http://www.cycling74.com/downloads/soundflower
Soundflower mirrors Jack in function, but is only available for OS X. It creates a virtual 2 channel and 16 channel interface in your OS so you can route audio between applications and out the soundcard.

To work around this CPU usage bug with firewire interfaces, select Soundflower 2ch or Soundflower 16ch as your default Audio Out / Audio In within Pd. Then within the Soundflowerbed menu, select your firewire interface as your soundflower output. I don't know why this works, but it does. You can now decrease your CPU usage up to about 2x when using a firewire interface. Screenshots follow.

Builtin audio OR soundflower as output while playing the 7. datastructure sequencer demo

Firewire as audio output while playing the 7. datastructure sequencer demo

Soundflower config

Has anyone else using OSX or Windows noticed this? When I use my 828mkII as an audio out in PD, my CPU usage doubles versus what it is when using the onboard audio. I have OSX 10.4.9 and running 0.39.2RC1 extended and don't experience this phenomenon with any other softwares. Any insights?

the "make" step of Pd from source throws this error consistently...

collect2: ld returned 1 exit status
lipo: can't open input file: /var/tmp//ccEMgBGH.out (No such file or directory)
make: *** [../bin/pd] Error 1

and I have yet to be able to get PDP/PiDiP to work with Pd-extended under OSX. Just of show of hands, is this to be expected, is there something obvious that people at this stage are usually missing? I love my mac, but linux is looking really good right about now.

Hey PD Community,

I am looking to move from OSX to Linux as my primary music and coding OS. I was hoping to be able to get a feel for what hardware everyone here is using, especially graphics hardware. I realize the state of linux graphics drivers right now is a sore spot, but I'd like to have a halfway decent system. I hear the X800 is great with open source drivers.

Any input you can give me or experiences you've had in this arena are greatly appreciated. Any other 2cents are great too....

If you're looking to build a sampler in the classical sense, this will be more difficult.

However, much of what you're describing can be accomplished most easily in realtime. Destructive edits such as "normalize" can be done using standard tabread4~ messages. Read the help files for tabread~, tabread4~, phasor~, and vline~. With those components, you can build your basic sampler very easily.

As for destructive resampling for timestretching, reverb, and delay, this could most easily be accomplished by doing it in realtime, sampling the output, and using that resample as the sample input.

Let us know of your progress. When I get home I'll post a few of my basic samplers for you to look at.

after hitting help, now it crashes when I try ti save.... i guess its still no different.