Here's an excerpt from 10.2.3 release document. This may be a good news some peple here...

http://docs.info.apple.com/article.html?artnum=107263

...
Audio Enhancements

* Addresses a potential issue in which sound from some USB audio devices may "stutter" after waking from sleep.
* Addresses an issue in which iSub, when used with some third-party USB speakers, could produce static.
* Improves default reverb setting for applications which use QuickTime Musical Instruments.
* Addresses an issue in which audio recorded from a USB device with a sample rate of 48k or 96k may include distortion.
* Improves results when recording from a USB microphone with a sample rate of 11.025k or 22.050k.
* Addresses an issue in which a MIDI device may unexpectedly fail to be recognized, when several devices are connected.
* Refines MIDI device support by removing compatibility for MIDI devices that are connected via a serial port (which Apple has not shipped with machines for several years).

... and I guess it could be a bad news for some people if this is true:

* Refines MIDI device support by removing compatibility for MIDI devices that are connected via a serial port (which Apple has not shipped with machines for several years).

<TRANSLATION>: "So screw everybody who bought one for hundreds of dollars based on fraudulent outer-packaging misrepresentation: 'Compatible with Mac OS.' Oh, yeah: and screw the third-party companies who pissed us off and made serial port cards to replace what we had amputated. You will go *our* way, no matter what it costs you, and we don't even need a cattle prod to make you go our way! BWA-HA-HA-HA-HA!" </TRANSLATION>

Not being one of the Apple screwees in this particular box-car load, I am no less disgusted. It took a particularly devious marketing mind to couch this statement in the pathetic lie that this somehow "refines MIDI device support." No it doesn't: it *reduces* MIDI device support by cutting out owners of serial MIDI devices, period. Just what kind of idiots do the idiots who write this tripe take the public for?

I didn't get the "removing compatibility for MIDI devices that are connected via a serial port", does it mean third party devs will NOT be able to build drivers, or the such? I mean, from a developper point of view, did Apple put something in the kernel/servers that locks the CoreMIDI to NOT communicate with old serial/printer ports?

RIP to my good old MOTU Express XT... Which works even under windows, BTW...

Anyway, the french 10.2.3 update is causing a minor problem over here (ie in France ;-) : i-tunes update (3.0.1) is still here after an upgrade, even if installing it over, and over again. What about you UK & US folks?

Bye.

I don't have the iTunes problem here on my machines.

Ass for the serial/midi driver, I think they removed it because it was unmaintainable. it doesn't stop someone from making his/hers for his/her serial device, in whatever form it comes in.

However, the serial/midi driver was made in the first place for *apple* serial ports, wich were FAR from standard anyway.

Buserror, I don't understand what you said.

If you're going to say, "It's "black," you can't turn around in the next sentence and say "It's white." What does "unmaintainable" mean? If it's "unmaintainable" for Apple, how is is "maintainable" for some 3rd party? How does it *not* "stop someone from making his/hers for his/her serial device"? How are *they* going to "maintain" it if Apple is providing no system-level support for it, no "hooks" for it in Core MIDI services?

I know for a fact that 10.2 included "AppleMIDISerialDriver.kext" which is now absent from 10.2.3. If it was "not maintainable," why was it there in the first place? And why is it now gone--other than the whims of the Apple god-geeks? (While waiting for clarification, I'll just look the other way and whistle, and pretend that Apple *isn't* in the royalties line for firewire. and pretend that there isn't now going to be a ramp-up in firewire MIDI devices hitting the market. )

Look at the name "AppleMIDISerialDriver.kext" Means it was made for apple serial drivers. And there aren't many of those left.

The problem that this driver causes is that even if it MIGHT work on a third party serial card/adapter, apple cannot affort to guarantee it, so instead of letting it be used that way *and carry the blame if it doesn't work* they decided to pull it.

IF a third party vendor wants to make a MIDI plugin for their serial adapters, NOTHING stops them, it's in fact fairly easy to do; and since it will be tailor made for their card, theres no problem of compatibility.

Thanks for the...clarification.

I would have thought that the name "AppleMIDISerialDriver.kext" meant that it *was* a serial driver, not "made for apple serial drivers." So Apple is making drivers for their drivers now?

a MIDI plugin is just a layer on top of something else. It can be on top of a serial driver, USB, firewire, PCI etc. But it has un underlying "provider"

Do not throw your serial Midi interfaces.
I have the Alpha version of the MIDI driver for the Stealth port

You want to test it write to : Blanc.cyril@wanadoo.fr indicating your Midi config

Your obligatio will be to report of your test.

Already test with 2 MTI II Motu works !

If we've now somehow started talking about "a MIDI plugin"--whateverTF that is--I know one thing: when to quit.

How's the weather out your way?