FusionDsp : the complete Dsp center for Volumio3!

This is something I saw… Maybe I could adjust a buffer. But as I said, this a workaround. It has other issue such as it does not change sample rate of tracks… I’m working on something more ‘audiophile’.
I try to apply some tweaks tomorrow.
Thanks for feedback.

Happy new year to you too! I just upgraded from 0.1.18 to 0.1.19 and everything seems fine. I still don’t see the ‘speaker distance’ setting though :slight_smile:

1 Like

And also happy new year!
I set audiobuffer to 12MB and seems to work.
Or is my gigabyte nuc 3150 with Celeron N3150 and 8GB RAM too slow?
But no plugin and upscaling to 32Bit 192Khz in best quality works without any problems.
@balbuze I will try your next tweaks for sure :slight_smile:

Thanks for your feedback.
I’m wondering if I can edit manually camilladsp.yml to enable multichannel configuration. (even without UI part in Volumio)

It works flawlessly in Moode audio.
eg.
playback:
channels: 8

mixers:
stereo:
channels:
in: 2
out: 8

Hello

I’m having a problem with PEQ and high Q values.

Volumio 3.179 with FusionDSP 0.1.19 on a Raspberry Pi 4
I have a PEQ config created by REW which includes Q values higher than 15.
Example:

Equaliser: Generic
sep 22
Filter 1: ON PK Fc 48.90 Hz Gain -4.40 dB Q 9.223
Filter 2: ON PK Fc 59.40 Hz Gain -7.10 dB Q 5.009
Filter 3: ON PK Fc 68.40 Hz Gain -8.00 dB Q 19.914
Filter 4: ON PK Fc 80.30 Hz Gain -9.40 dB Q 17.968
Filter 5: ON PK Fc 100.0 Hz Gain -6.30 dB Q 26.510
Filter 6: ON PK Fc 116.0 Hz Gain -9.00 dB Q 19.420
Filter 7: ON PK Fc 137.0 Hz Gain -9.10 dB Q 18.821
Filter 8: ON PK Fc 142.5 Hz Gain -8.70 dB Q 16.509
Filter 9: ON PK Fc 159.0 Hz Gain -9.90 dB Q 33.368

When I load the file in FusionDSP, is works fine. I can see the EQs created in the UI and the camilladsp.yml file looks ok.
Example:

eq5:
type: Biquad
parameters:
type: Peaking
freq: 100.0
q: 26.510
gain: -6.30

Now when I hit apply, I get a message saying “Coefficient Q Must be a float [0,1-15] in eq5”
The EQ is working but if I try to make any change in the UI, for example adding a lowshelf, I keep getting this error for all the EQs with Q higher than 15.

Thanks and happy new year :slight_smile:

Hi!
When Eq is imported, no value checking for Q is done.
When you hit save, values range are checked.
For Q I set range from 0 to 15… Of course, I can change this range to met your needs.
I have to check CamillaDsp possibly.

1 Like

Great! V0.1.19 solves my problem I had with stepping through the queue (filled with webradios) reported earlier:

I´ve got one (last) topic on my wishlist :blush::
Can you add the higher order Highpass/Lowpass filters from CamillaDSP :

ButterworthHighpass(Hz, order) and ButterworthLowpass(Hz, order)?

Especially the Highpass is very useful as a subsonic filter.

1 Like

Ok, I’ll have a look!

3 Likes

I tested longer today still dropouts even with 12MB buffer.

In fact. I was talking about an internal setting… I need to push a new version.,

2 Likes

What’s the difference between Resampling in the standard playback options and in the fusiondsp settings? Is the resampling something that take place before EQ or after? I use currently the standard playback options resampling and want to know what is different when I switch from standard to resampling in fusiondsp settings?

Standard resampling is apply only for mpd.
Resampling in FusionDsp is applied to all sources, Spotify, BT, Tidal.
You have to disable standard resampling to use resampling in FusionDsp.

Okay thanks.
Awaiting your next update and hope the dropouts are fixed then it’s 100% useful for me.

Hello!
New version v 0.1.20

  • Q range is now [0.1-40]
  • New filters ButterworthHighpass & ButterworthLowpass
  • Some adjustements to reduce dropouts. (not sure)

Reboot after install
Let me know

3 Likes

Form me works great on x86, no dropouts now, flawless…thank you for your efforsts!
Just quick question, do you use attenuation? For settings gain -2.6 and Q 2.4 its like -2db?

Yes. When only negative values are applied, -2dB is set to avoid potential phase rotation.
When positive values are applied, attenuation is the max value detected+ 2dB
So in your case 4.4dB

1 Like

Thanks!!! :grinning:

Dropouts every few sec and this is with resample and also without resample unuseless for me.

Very often dropouts worser then 19.
http://logs.volumio.org/volumio/6NyMWWg.html

When I set buffer from 4mb to 12mb a bit better dropout every 10…60sec.