Well, on Linux it enables the Midi ports but it also forces capture/playback to 48k/96k at 24 bit. In my experience, the latter is what borks the USB communication as there isn't enough bandwidth on USB 1 for full stereo duplex of anything other then 16bit, 44.1k. The Edirol "advanced drivers" for OSX & Win can somehow handle that, but of course the Linux drivers don't and you get hard locks. These devices were designed before USB 2.0, so they Roland/Edirol were doing non-standard tricks ...
In any case, the advanced mode switch *does* work with Linux, provided you only want to either capture or playback, but not both simultaneously. I'd imagine it would work with 1 input and 1 output, but I bet that's not an option at the driver level ... On May 10, 2013, at 12:32 PM, Miller Puckette <[email protected]> wrote: > Thanks Julian and Dan - > > I've never known what that button did but now I guess i do - it makes the > device fail. > > M > > On Fri, May 10, 2013 at 11:52:05AM -0400, Dan Wilcox wrote: >> I can confirm that! Got nailed in a live show some years ago when the switch >> was accidentally in advanced mode. The USB bus would explode whenever I >> loaded a patch with audio in ... >> >> On May 10, 2013, at 5:04 AM, [email protected] wrote: >> >>> From: Julian Brooks <[email protected]> >>> Subject: [PD] Edirol UA25 on rPi (was) ESI Gigaport HD+ on RPI >>> Date: May 9, 2013 9:37:54 PM EDT >>> To: Miller Puckette <[email protected]> >>> Cc: pd-list <[email protected]> >>> >>> >>> Hi Miller, >>> >>> Just spotted this and remembered you mentioned an Edirol UA25: >>> >>> 'I've found that with my Edirol UA-25 I have to disable Advance mode in >>> order to make it work in full duplex with JACK.' >>> >>> From here: >>> http://www.raspberrypi.org/phpBB3/viewtopic.php?p=286222#p286222 >>> >>> Regards, >>> >>> Julian >>> >>> >>> On 27 April 2013 16:25, Miller Puckette <[email protected]> wrote: >>> I can't report about that but I have another compliant USB 1.1 interface, >>> Edrol UA25, which I've never been able to get to work with a pi. So I think >>> this is confirmation that that can indeed sometimes happen (if it were just >>> one of us perhaps it could have been a fluke). >>> >>> cheers >>> M >>> >> >> -------- >> Dan Wilcox >> @danomatika >> danomatika.com >> robotcowboy.com >> >> >> >> >> > >> _______________________________________________ >> [email protected] mailing list >> UNSUBSCRIBE and account-management -> >> http://lists.puredata.info/listinfo/pd-list > -------- Dan Wilcox @danomatika danomatika.com robotcowboy.com
_______________________________________________ [email protected] mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
