Help with Dropouts

This is a reply to Laurence who picked up on an issue I mentioned in another thread. To be clear, I do not think at the moment that this is a problem directly related to C6 (which has reduced the incidents of the issue decribed below considerably - though I still have it) and seems considerably more stable than C4 from which I’ve upgraded.

Running C6 (64-bit) on a Win7-64 bit Home Premium. The PC was out-of-the-box clean before C4 installed originally. No other applications have been installed on the machine but have recently upgraded to C6.

When C6 kicks off there are no other apps running (that I’ve started) but, of course, a bunch of Windows services run themselves at start-up. A number of them are now disabled that were obvious as “not needed” such as Windows Update (done manually when needed).

Disabled the network interface and the on-board sound card (in the BIOS).

Execution preference given to “Background Processes”.

64-bit Yamaha FW Driver V1.6 installed. Physical firewire chain goes:-

Yamaha N12 <----------> Motif XS8 <----------->PC

Motif XS set as clock master in FW control panel. Latest version of Firewire firmware installed on N12 and Motif XS (ie the firmware that replaced the original mLAN).

N12 and Motif XS extensions installed to C6.

Logical VST connections are all from/to N12 (eg Stereo Out mapped to N12 ST L/R; Stereo In mapped to N12 Rec L/R etc). No Motif XS channels are explicitly mapped as VST connections on this C6 page.

Motif XS8 VSTi (64-bit version) configured to use MOTIF XS MAIN for the audio (and similarly for MIDI). VST instrument track for Motif XS8 mapped to “Stereo Out” as its destination.

Motif XS configured to “Monitor with PC”.

N12 physical MIDI in/out are connected to Mackie Control and configured as such in C6 Devices.

So, apart from when the Mackie Control receives/sends MIDI over it’s physical connection to the N12, all MIDI and audio signals are sent via the Firewire bus.

I thought at one time that the occasional dropouts might have been due to excessive MIDI traffic to/from the Mackie Control. Two reasons for not now thinking this is the problem:-

  1. The dropouts are not related to periods when there are high levels of automation in the mix (which would involve high levels of MIDI traffic to the Mackie Control);
  2. Disabling the Mackie Control doesn’t stop the dropouts.

VST Performance indicator is barely moving off the bottom of its gauge before, during or after a dropout occurs.

Dropouts seem to start happening after the system has been running for a few hours. Shutting the whole thing down and restarting everything will stop it happening for another couple of hours (memory leak?). When it starts, the dropouts (both audio and missing MIDI NOTE OFFs) will happen equally when playing the project and when exporting it (real-time due to the Motif XS) to a WAV or MP3 file. On rare occasions all audio output ceases completely. I’ve discovered that switching the master clock in the FW Control Panel to N12 and back again to MOTIF XS will sometimes, though not always, restore the audio. MIDI traffic continues uninterrupted when this happens.

So, that’s where I am with this. I’ve downloaded the Latency Testing tool (thanks for the link) and I’ll come back later with the results of that test.

Anything you, or anyone else, can offer as a suggestion to resolve this would be very gratefully received :smiley:


Have you done the Win7 Firewire driver change shown here
Many firewire devices don’t like the default Win7 Firewire drivers (including my Motu).
The above fix is found all over the net, and helps for many people.
Sorry if you’ve been through this already !!

OMG!! It works! I’m tyring to type while dancing round the room. How could I have been so stupid?!!?

Thank you.

I’ve just spent a couple of hours thrashing the FW interface with the biggest projects I could find. I made it play 32 audio tracks with loads of inserts and a similar number of MIDI tracks with dense automation. Then I made it download a huge file across my network while still playing. Not a single glitch on the output. The cursor jumped occasionally at particularly dense traffic moments but not a single dropout or ciphering MIDI note.

My gratitude should be flooding towards you :smiley:

BTW I ran the Latency Testing utility recommended by Laurence in a previous post and it reports a maximum of below 500 microseconds (I think that must be around half a millisecond). I’m guessing that’s not too bad.


…amazing how well it all works when it all works eh ???
Glad I could help.