• SONAR
  • Weird dropout problem (p.2)
2016/08/16 10:02:51
JohanSebatianGremlin
Rob[atSound-Rehab]
 
Are you sure? MIDI is voltage and depending on how it's wired internally the signal might be present at the MIDI thru. A quick check unplugging the cable cannot hurt ...

Well I've never actually tried unplugging cables. But I do know that when I turn the keyboard off, the transport controls on the control surface stop working.
 
 
Rob[atSound-Rehab]
 
Have you ever tried LatencyMon? It's quite good at listing what causes a CPU spike ...

I'll check it out.


2016/08/16 10:21:38
Cactus Music
If your interace is USB buss powered, there is a setting that shuts down USB ports try that. 
 
But also load a simple project with bare bones and just the TTS-1. 
2016/08/16 11:15:47
JonD
JohanSebatianGremlin
I have already disabled every power saving feature I could find.


For every USB hub?  You can check by going to Device manager, and into the properties of each USB hub.  Under Power Management, the boxes saying to allow to save power (or something like that) should all be unchecked.
2016/08/16 11:17:48
JohanSebatianGremlin
JonD
JohanSebatianGremlin
I have already disabled every power saving feature I could find.


For every USB hub?  You can check by going to Device manager, and into the properties of each USB hub.  Under Power Management, the boxes saying to allow to save power (or something like that) should all be unchecked.


I don't recall doing that. I'll check when I get home tonight. Thanks.
2016/08/16 15:36:07
tlw
If you've not tried it yet, what happens if you disconnect audio interface and all MIDI Hardware and reboot, then load Sonar, point it at the on-board audio (if it exists) and create a fresh project using the stock "normal" template (or a project with just MIDI tracks pointed at nothing if there's no on-board audio) and play that?

That way if the problem recurs the source would seem to be something internal to Windows, a driver, the PC or Sonar.

Another thing that might be worth trying is to rename Sonar's aud.ini file to something like aud.old so Sonar creates a new one.

A final thought that occurs to me is it might be to do with the mouse driver or associated mouse preferences software. I'm wondering if the mouse is getting continually polled by its software to check if it's doing anything and after enough time a lack of response does something strange, like a huge PCI bus latency spike or there's a memory leak. The problem not occuring earlier than 4 minutes because every time the mouse is moved or clicked whatever's doing the polling resets itself in some way. Years ago I had strange freezes that took ages to track down by the "switch something off and see what happens" method. The culprit turned out to be a mouse driver with a memory leak that polled non-stop and gradually used up huge amounts of RAM.
2016/08/16 17:32:01
JohanSebatianGremlin
Its not just mouse activity that prevents dropouts though. If I'm recording a midi track or just playing/noodling on the midi keyboard, I can work all night and never have a single dropout. But once I'm to the point that I'm recording external instruments (not touching the midi keyboard or mouse during the take) or doing something like just listening to the mix/taking notes, I get dropouts after 4 minutes of no activity.
2016/08/16 18:06:24
tlw
Well, that blows my idea out of the water.

Tricky one this. OK, another possibly stupid thought occurs to me. Is the problem limited to Sonar or does playing audio through the interface from Windows media player or itunes also result in the death of audio? Does Sonar behave itself using a different driver mode (MME/WDM rather than ASIO)?

Anything relevant in the Windows system and application logs?
2016/08/16 21:09:54
JohanSebatianGremlin
Ding ding ding ding!!!!!
 
Looks like JonD might be the winner. Some, but not all of the generic and root USB hubs listed in device manager had the box to allow power management unchecked. I made sure all were unchecked and then loaded up sonar and a 4 minute 50 second project file and let it play. Played perfectly to the end, no drop out. 


I'll have to do more testing to be sure, but so far it looks like JonD nailed it. Well done sir and thanks to all who offered their time and knowledge to help with this.
2016/08/16 22:43:26
pinguinotuerto
So it was a power saving issue after all.
12
© 2026 APG vNext Commercial Version 5.1

Use My Existing Forum Account

Use My Social Media Account