Re: MIDI and Sleep
Re: MIDI and Sleep
- Subject: Re: MIDI and Sleep
- From: Kurt Bigler <email@hidden>
- Date: Wed, 06 Mar 2002 02:49:46 -0800
>
On Tuesday, March 5, 2002, at 12:42 AM, Kurt Bigler wrote:
>
> On OS 9 I always had the problem with either the MOTU Fastlane or the
>
> MidiSport 2x2 that if the computer was put to sleep, the interface had
>
> to be unplugged and re-plugged (to reset the USB bus?) after wake-up.
>
>
>
> Now under OS X it appears that the replugging is no longer necessary,
>
> but my MIDI applicaton must be quit and restarted. Is there anything I can
>
> do in my application to improve its behavior in this regard? My OS X testing
>
> is with the MOTU Fastlane because the MidiSport 2x2 does not work at all
>
> for me under OS X.
on 3/5/02 1:24 AM, Kurt Revis <email@hidden> wrote:
>
(I know the MOTU Fastlane works fine when you hot plug/unplug it, but
>
the MIDISport does not. Maybe that is related, maybe not.)
It is possible I never tried the MIDISport without hotplugging it. I am
used to hotplugging all the time because I never restart my computer except
after crashes or to switch betw 9 and 10, so I always replug after sleep as
a matter of course.
>
BTW, the MidiSport 2x2 works (mostly) for me. Midiman released new
>
drivers a week or two ago--have you tried them yet?
>
Also--are you running Classic while you are doing this? Maybe Classic
>
wakes up and grabs the MIDI devices before CoreMIDI can get to them, or
>
something.
With the new driver I got the Midiman to work for part of a minute if no
hotplugging was involved, and if Classic was NOT running. Classic is not
running any midi applications, however the USB drivers are present and
probably get loaded.
Is there some way to tell Classic to exclude certain extensions that should
be included in a regular OS 9 boot? (Apparently Apple has some way to do
this.)
Curiously in the current configuration, running my MIDI app causes Classic
to "disappear" immediately. However it does not disappear quickly enough to
avoid the problem of MIDI not working. But if I quit my app and restart it
(not having restarted Classic) then MIDI functionality returns for my app.
In any case the console.log looks like this:
>
MIDISPORTUSBDriver init
>
NewDevice returning 406320
>
creating new interface runner in USBMIDIDriverBase::Start
>
~MIDISPORTUSBDriver
>
Mar 6 02:17:17 localhost Classic[306]: *** The BlueBox just crashed! ***
>
>
Mar 6 02:17:17 localhost Classic[306]: ERROR = SystemCrash(9)
Even without Classic involved the MIDISPORT is still very unreliable. I get
missing note-offs constantly, and the app stops receiving MIDI within 30
seconds of playing. Restarting the app restores MIDI again, but not for
long. I have none of these problems with the Fastlane. Pretty much no
stuck notes ever with the Fastlane.
on 3/5/02 2:55 PM, Kurt Revis <email@hidden> wrote:
>
I can replicate this behavior with the MOTU Fastlane 2x2 on two
>
different machines. Put the machine to sleep, wake it up again, and the
>
interface will no longer send MIDI notes.
>
>
However, the Midisport 2x2 still works fine in this scenario, so
>
CoreMIDI is probably not at fault here. (Unless there is a CoreMIDI or
>
USB bug that the Midisport driver works around and the MOTU driver
>
doesn't -- hey, anything's possible.). I don't suggest that you change
>
your app to do anything special, just report the bug to MOTU and hope
>
they fix it.
>
>
(If you are desperate in the short term, you could investigate calling
>
MIDIReset() upon waking up--that may or may not have some effect. But
>
that's a really bad idea to do in shipping code.)
No, I don't have a dire need for a code workaround for the sleep problem
since I am used to replugging. I was more concerned whether I was doing
something wrong (or nonoptimal) in my app to cause this.
-Kurt Bigler
_______________________________________________
coreaudio-api mailing list | email@hidden
Help/Unsubscribe/Archives:
http://www.lists.apple.com/mailman/listinfo/coreaudio-api
Do not post admin requests to the list. They will be ignored.