• Registered: 2011-04-07
  • Posts: 4

Topic: FF800 drivers blocking "3rd party" MIDI-OUT ports

Hello everybody!

Bit of a legacy system here, with
ASUS P4C800E DeLuxe Board
LaCie FW800 PCI card
2 x Fireface 800
...

all working fine - except:

as soon as RME drivers are installed - after restart -
Windows (both XP Pro SP2 or Server 2003, same symptoms)
sees only MIDI-Output #1 (out of 8) of my MOTU MIDI-Express XT 8x8 Interface
AND
the editor software of my ROLAND VG-99 won´t communicate with the device
anymore (using midi via USB for that,
"VG-99 Control" no more showing up in the editors MIDI-OUT-Setup-list)
AND
my ALESIS Trigger I/O USB MIDI-devices OUTPUTS are also no more available.

Uninstalling the RME drivers, everything comes  back to normal immediatly,
with the MIDI-Express showing all its 8 OUTS, and the VG-99 editor happily exchanging
data with its device (with "VG-99 Control" back in the MIDI-OUT-Setup-List)
and the Trigger I/O´s USB-MIDI-OUT also back in every list.

(MIDI-Inputs are never affected, always there as supposed to)

reinstall RME-drivers - aforementioned MIDI-OUTs disappear again.
uninstall RME-drivers - again all comes back.
I tried the legacy drivers from Driver Archive back to 2631,
but the driver version doesn´t make any difference.

All this on a fresh install of Windows btw.

Things were working better alongside a while ago, in those days I had only 1 Fireface
with a LaCie FW400 PCI controller.

This Either-Or - situation is somewhat uncomfortable, and since the
3 affected devices are quite different (MOTU via Parallel-Port connection,
ROLAND & ALESIS via USB connection), i am afraid that any other MIDI-device
would run into similar trouble...

Any ideas?

2 Reply by MC

  • MC
  • MC
  • Administrator
  • Offline
  • From: Germany/Thailand
  • Registered: 2007-04-06
  • Posts: 24,847

Re: FF800 drivers blocking "3rd party" MIDI-OUT ports

Search the web or this forum - this is a limitation of XP, no way around it except using Windows 7.

Regards
Matthias Carstens
RME

MC's Website

3 Reply by Sangeet (edited by Sangeet 2011-04-08 02:09:46)

  • Registered: 2011-04-07
  • Posts: 4

Re: FF800 drivers blocking "3rd party" MIDI-OUT ports

Thank you Matthias for the quick reply!

Before posting, I did search the web and this forum,
and therefore was aware of that "NT-MIDI-bottleneck".

It just didn´t really fit with the symptoms I´m observing here...
Meanwhile, I did some more "research", with a fresh
"vanilla" XP SP2 install (no audio/multimdia whatsoever stuff
installed yet):

- all (MIDI-related) drivers can be installed at the same time,
here it only matters which device is powered up or not;

the scenario explained in my post above is this:

[img align=M]http://www.sunderchandana.net/lgf/BayStuff/midiPics_web/FF1-2-MOTU.jpg[/img]

Both Firefaces active, MOTU MIDI Express allowed to show
1 of its 8 OUTs, noone else admitted.

- switching off the MOTU (and rebooting),
the ROLAND VG-99 OUT steps in:

[img align=M]http://www.sunderchandana.net/lgf/BayStuff/midiPics_web/FF1-2-VG_noIO.jpg[/img]

- switching off the VG, now the ALESIS trigger thingy OUT
(using generic Windows driver,
therefore listed as "USB Audio device") is allowed to show up:

[img align=M]http://www.sunderchandana.net/lgf/BayStuff/midiPics_web/FF1-2-IO.jpg[/img]

- so, as long as both Firefaces are running,
it so far is a narrow bottleneck...

... until I switch the second one off. Now I´m getting this:

[img align=M]http://www.sunderchandana.net/lgf/BayStuff/midiPics_web/FF1-all.jpg[/img]

Although the "Windows reg entry bottleneck theory" may still hold up,
this makes me suspect there might be some sort of a hidden driver conflict, as:

- 1#, there appears to be some hierarchy:

1. active RME devices OUTs are always available, thus on top of the pile.

2. next, the old MOTU Parallel Interface - if on - is then allowed to show 1 OUT (of 8).

3. 3rd here is the ROLAND device, which in turn has priority over
4. the ALESIS/Windows generic driver, which is the weakest of that bunch.

- 2#, so if both my Firefaces are running, out of 3 other devices 2 get their OUTs
knocked out, while the "strongest" of them is allowed one OUTPUT.
This seems to be a bit much, if we just try to blame it on the Windows NT thing alone, no?

What I´m trying to suggest (politely) is that the RME MIDI drivers
- at least when more than one FF involved -
could perhaps behave a little more "3rd party MIDI friendly".

- I btw. found some kind of whacky way around all this, by switching the 2nd FF off,
which results in maybe 15 secs. of general "slow-motion" behavior
(not of me, of the computer...), then all missing OUTs magically reappear,
after which, when switching it back on, I would eventually see this:

[img align=M]http://www.sunderchandana.net/lgf/BayStuff/midiPics_web/FF1-2-all.jpg[/img]

which on my system means MIDI-OUT heaven...
This state is sometimes stable enough to remain if I restart an application,
sometimes not. And it´s of course lost after reboot...
(go back to first screenshot...)

Sorry for this endless (perhaps boring) rant about trouble around a legacy OS on
a legacy computer... but as some folks still might stick with XP for a while...


greetings

4 Reply by MC

  • MC
  • MC
  • Administrator
  • Offline
  • From: Germany/Thailand
  • Registered: 2007-04-06
  • Posts: 24,847

Re: FF800 drivers blocking "3rd party" MIDI-OUT ports

It seems to me you searched (or found) the wrong topic. This is not about the 10 entry limit, this is about a limit of WDM devices and random/undocumented switching off of MIDI ports when the number of WDM devices gets critical. But good you found some kind of workaround for your setup.

Regards
Matthias Carstens
RME

MC's Website