On Wed, Dec 29, 2010 at 01:01:42AM +0000, errordeveloper@gmail.com wrote:
Greate to hear that vanilla compiles ok and starts up!!
I did have to patch the make files to use $(STRIP) instead of "strip" and I did some thrashing about in Makefile.am to get pd_LDFLAGS suck in enviornment setting for LDFLAGS that I don't feel supper confident about but, we can review that mess after it makes noises.
what version of GCC did you use?
gcc 4.5.2 20101105(prerelease) its what OE made as part of the angstrom build.
how floating point is handled ?
I bet its in software libs.
have you tried using USB audion instead?
no.
I just had been looking at the Pandora Board, which uses 2-core Cortex-A9 with FPU, thing to get it soonish to run my pd WSN project (http://wmi.new-synth.info).
Yeah the panda is nice. I have a BeagleBoard.
--mark
On Tue, Dec 28, 2010 at 10:14:44AM -0800, mark gross wrote:
I'm trying to get vanilla PD working on a beagle board using open embedded's angstrom-2010.x distro. I've hacked up a bit bake recipe (still a hack job) and the task-beagleboard-demo-image.bb to make a image that will actually run the PD GUI on a BB.
My problem now is with the audio. I can't seem to get PD's audio + midi test to make any more than queuer counter noises. I've played with using OSS and Alsa, and disabling pulse from the build.
aplay does play a 22050Hz mono-wave file.
I get the following errors from PD when I attempt to run it as root: using OSS I get clicks from the test patch. Using Alsa I get a spike in cpu use and things get really sluggish. and I get -- tried but couldn't sync A/D/A ...
I would like to get any debugging tips or advice possible to finish this project and post a howto (and perhaps add the puredata.bb recipe to OE)
Thanks!
--mark
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list