Hi Jack,
Yes, that was my mistake; I've now attached both the patch and the abstraction.
Thanks! Nick
On Wed, Apr 16, 2014 at 6:00 AM, pd-list-request@iem.at wrote:
Send Pd-list mailing list submissions to pd-list@iem.at
To subscribe or unsubscribe via the World Wide Web, visit http://lists.puredata.info/listinfo/pd-list or, via email, send a message with subject or body 'help' to pd-list-request@iem.at
You can reach the person managing the list at pd-list-owner@iem.at
When replying, please edit your Subject line so it is more specific than "Re: Contents of Pd-list digest..."
Today's Topics:
- Re: Pd and RME HDSP alsa vs jack performance (Peter P.)
- Re: PD List (GEM VideoSynth) (Jack)
Message: 1 Date: Tue, 15 Apr 2014 18:40:31 -0400 From: "Peter P." p8rpp@aol.com Subject: Re: [PD] Pd and RME HDSP alsa vs jack performance To: Federico Galland federicogalland@gmail.com Cc: pd-list@iem.at Message-ID: 20140415224030.GB8322@aol.de Content-Type: text/plain; charset=us-ascii
- Federico Galland federicogalland@gmail.com [2014-04-15 17:46]:
On Tue, 15 Apr 2014 16:07:03 -0400 "Peter P." p8rpp@aol.com wrote:
Hi list,
following up onto some earlier postings I had made about the current problems running Pd with any buffersize larger(!) than 5 using alsa, the mmap code and an RME HDSP Multiface on Debian Linux, here are some more results from tests I made recently:
Alsa only works with audiobuf smaller 6, otherwise causing totally garbled sound. With that short buffer settings it is unusable for everything more CPU intense than sine test tones though.
The only way to get Pd to run reliably under some load on my box using the HDSP is to use jack with 256 frames at 2 periods/buffer.
Now what is weird is that the built in soundcard of my box is able to do the same thing without problems with just 64 frames at 3 periods/buffer in jack.
Again with the HDSP, once I start adding Gem and let it render a single geo, I do get many audio dropouts unless I increase jack's buffers to a minimum 2048 frames at 2 frames/buffer.
I tested this using two different HDSP cards on two different Debian boxes, using the -rt flag, realtime and non-realtime kernels, optional callbacks and different sleegrain setting.
Somehow it seems very sad that one of the best pro multichannel sound cards under Linux seems to be unusable with Pd under Alsa reliably.
I would like to contribute to any ways of improving this situation and desperately hope that someone can comment on my findings.
thank you! best, Peter
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management ->
http://lists.puredata.info/listinfo/pd-list
I guess you have ruled out everything under this wiki
http://wiki.linuxaudio.org/wiki/system_configuration ? Thank you Federico, I looked at the page and most suggesstions is stuff I already considered.
Is the HDSP a PCI interface? If so, pay particular attention to the "pci
bus latency" part. Well it is a ExpressCard interface in my case, and I wonder if that qualifies as PCI device. The above pages mentions that PCIe devices are not affected by the irq priorities as PCI does.
best, Peter
Message: 2 Date: Wed, 16 Apr 2014 10:55:40 +0200 From: Jack jack@rybn.org Subject: Re: [PD] PD List (GEM VideoSynth) To: Nick Arner nicholasarner@gmail.com, pd-list pd-list@iem.at Message-ID: 534E458C.6060408@rybn.org Content-Type: text/plain; charset="utf-8"
Hello Nick,
Don't forget to reply to all. That could be interresting for other people on this list. The document you sent is now a patch (attached). But what can we do with it ? There is only a single object [video_synth] on the canvas. It seems to be an abstraction, right ? You have to send us the patch AND the abstraction. And where are other objects like [gemwin] ? We are talking about Gem, no ? ++
Jack
06 59 23 94 69 Andr?
Le 16/04/2014 01:23, Nick Arner a ?crit :
Hi Jack,
Thanks for the reply! I've attached the PD patch (it looks like the attachment went through this time)
Thanks, Nick
-- nickarner.com http://nickarner.com