I am having a problem with VoiceMeeter Banana that appears to have popped up after the latest Win10 cumulative update (July 11). Every few seconds, the audio output of VMB (which I can monitor even with the HPSDR app closed)... in response to various audio inputs that I have tried... appears to drop out for a few tens of milliseconds. The effect is not periodic... but it does occur every few seconds. I can also monitor these audio sources without using VMB (i.e., by directing their audio output to my Focusrite 2i2 sound card instead of VMB)... and the dropouts are not heard. So the problem appears to occur only with VMB in the digital audio path.
The audio source (e.g. WSJT-X, FLDIGI, NCH tone generator) is fed into VMB via the Voicemeeter AUX input. The audio output is obtained from the VMB VAIO (B1) output. The output of Voicemeeter AUX channel is directed to B1... and the output of the Voicemeeter VAIO channel is directed to B2... thereby providing a two way connection between apps like WSJT-X and my ANAN-10's VAC 1 I/O, when VAC 1 is enabled.
All of the VMB sample rates (including the virtual channels) are set to 48000Hz.
In the Windows 10 "Sound" settings... all of the devices are set to a sampling rate of 48000 and 16 bits per sample (DVD quality).
I removed and reinstalled VMB... but that didn't fix the problem.
Has anyone seen this problem pop up recently?
Stu
Recent problem using VMB
- Tony EI7BMB
- Posts: 656
- Joined: Sun Apr 09, 2017 2:31 pm
- Location: Dublin
- Contact:
Re: Recent problem using VMB
Not sure if this applies Stu but I did have to redo my Ethernet settings after the last update, I originally thought I had a purely audio problem.
Re: Recent problem using VMB
Tony
Thanks for the suggestion. I updated all of the Ethernet adapter settings as recommended in this document:
https://helpdesk.flexradio.com/hc/en-us ... erformance
But... the VMB audio output dropout problem still remained.
I changed the VMB "MME Buffering" setting from 1024 (default) to 2048... and the problem appears to have gone away.
Thanks for the suggestion. I updated all of the Ethernet adapter settings as recommended in this document:
https://helpdesk.flexradio.com/hc/en-us ... erformance
But... the VMB audio output dropout problem still remained.
I changed the VMB "MME Buffering" setting from 1024 (default) to 2048... and the problem appears to have gone away.
- Tony EI7BMB
- Posts: 656
- Joined: Sun Apr 09, 2017 2:31 pm
- Location: Dublin
- Contact:
Re: Recent problem using VMB
Glad its ok now. Thanks for the link, handy reference
Edit: Out of curiosity I changed MME also to 2048 and even though I'm using ASIO I seem to be getting fewer clicks/pops when using JT65
Edit: Out of curiosity I changed MME also to 2048 and even though I'm using ASIO I seem to be getting fewer clicks/pops when using JT65
Re: Recent problem using VMB
Tony
Out of curiosity, have you tried selecting MME for the VAC 1 driver? I know that ASIO is recommended for lower latency... but, of course, latency isn't an issue with the digital modes.
I don't hear any "pops" (now that I have fixed the problem that I described earlier).
On transmit, even a brief dropout/glitch would show up on my 10 digit frequency counter (which is monitoring a sample of the RF output). During a one minute WSJT transmission (the counter is set to a 0.3 second gate interval) ... the display is steady as a rock... other than following the FSK pattern.
Stu
Out of curiosity, have you tried selecting MME for the VAC 1 driver? I know that ASIO is recommended for lower latency... but, of course, latency isn't an issue with the digital modes.
I don't hear any "pops" (now that I have fixed the problem that I described earlier).
On transmit, even a brief dropout/glitch would show up on my 10 digit frequency counter (which is monitoring a sample of the RF output). During a one minute WSJT transmission (the counter is set to a 0.3 second gate interval) ... the display is steady as a rock... other than following the FSK pattern.
Stu
- Tony EI7BMB
- Posts: 656
- Joined: Sun Apr 09, 2017 2:31 pm
- Location: Dublin
- Contact:
Re: Recent problem using VMB
Stu, I've not tried that yet but probably won't mess with anything until 6M season is over , lost too many hairs already
Re: Recent problem using VMB
It's very unlikely that it is an Ethernet problem.
See this thread about using LatencyMon to see if your system has problems processing real time audio streams.
73,
Scott
See this thread about using LatencyMon to see if your system has problems processing real time audio streams.
73,
Scott
Re: Recent problem using VMB
Scott
Here is what I obtained using LatencyMon
PC= Dell: I7-4790@3.60Hz, Win10Pro, NVidia GeForce GT 720
Running: PowerSDR, ANAN-10E via point-to-point Ethernet, Voicemeeter Banana, WJST-X, Internet Explorer via WLAN, plus miscellaneous other applications.
Everything working okay.
Power SDR main GUI CPU utilization reported: 8-11%
Stu
Time running: 0:19:11 (19 minutes and 11 seconds)
Current measured interrupt to process latency: 115us
Highest measured interrupt to process latency: 500us
Highest reported ISR routine execution time: 339us (DirectX Graphics Kernel)
Highest reported DPC routine execution time: 517us (NVIDIA Windows Kernel Mode Driver)
Reported total page fault count: 0
Here is what I obtained using LatencyMon
PC= Dell: I7-4790@3.60Hz, Win10Pro, NVidia GeForce GT 720
Running: PowerSDR, ANAN-10E via point-to-point Ethernet, Voicemeeter Banana, WJST-X, Internet Explorer via WLAN, plus miscellaneous other applications.
Everything working okay.
Power SDR main GUI CPU utilization reported: 8-11%
Stu
Time running: 0:19:11 (19 minutes and 11 seconds)
Current measured interrupt to process latency: 115us
Highest measured interrupt to process latency: 500us
Highest reported ISR routine execution time: 339us (DirectX Graphics Kernel)
Highest reported DPC routine execution time: 517us (NVIDIA Windows Kernel Mode Driver)
Reported total page fault count: 0
Re: Recent problem using VMB
Those are some great numbers, Stu.
73!
Scott
73!
Scott
Re: Recent problem using VMB
I started having problems too. Latency mon sees no problems. 2048 buffer size didn't change things. But dropping the Primary sample rate to 96000 did solve it.
de Mike W9MDB
de Mike W9MDB
Mike W9MDB