FusionDsp for Volumio3 beta

Please use this thread to report your experiences using FusionDsp and not pollute Volumio 3 RC1 thread
Thanks

This plugin is designed to apply different type of Dsp on Volumio using CamillaDsp
In order to install, go in IP/dev and enable 'plugin test mode’
And install from the store

Provides

  • A 15 bands graphic equalizer
  • Or a 2x15 bands graphic equalizer
  • Or a Parametric equalizer with :
    • Up to 50 bands (peaking, lowshelf, highshelf, lowpass, highpass, notch)
    • Equalizer scope for each band (L, R, L+R)
    • More than 3800 variant of headphones EQ from AutoEQ ! AutoEq
    • import for local EQ file (must be in /data/INTERNAL/FusionDsp/peq/)
  • 3 preset
  • Or a convolution filters (FIR) with autoswitch samplerate for filters
    • DRC-FIR to create filter with an impulse

For all

  • Progressive Loudness effect with threshold setting - sort of equal loudness curve
  • Auto gain
  • Toggle with/without effect
  • Separate volume level for left an right
  • 4 crossfeed for headphone (Bauer, Chu Moy, Jan Meier, Linkwitz)
  • Mono / stereo toggle
  • High quality resampling
  • tools to easily plays test files (pink noise, sweep) to help measurments

Some screenshots
15 bands Eq

2x15 bands Eq

Parametric Eq

FIR filters

Headphones Eq import from AutoEq

Crossfeed for headphones

And much more features!

2 Likes

I’m having a look right now
Edit. It seems to works fine for me.
Can you copy/paste values you are using please?
Tip : save values between each eq you add

1 Like

No. It may work with a specific compiled CamillaDsp for arm.
The problem is I don’t have a PI zero…

1 Like

Im trying to enter the some parametric eq settings for my headphones. there are 10 settings and 4 of those do not work.
low shelf 35Hz gain 2.5dB Q factor 0.71 so I select “lowshelf Hz,Db,Q” and enter 35,2.5,0.71 in the Eq box
others that come up with the error
“lowshelf Hz,Db,Q” 105,5.5,0.71
“highshelf Hz,Db,Q” 1300,6.0,0.71
“highshelf Hz,Db,Q” 10000,-5,0.71

this is with Volumiobuster-3.114-2021-10-15-pi and its installed 0.8 beta of the dsp

thanks

Ok I see the problem. Float value for Q is not allowed…
You can still use 1 instead of 0.7…
I’m going to fix it asap…
Sorry for that… It is beta testing :wink:

1 Like

@balbuze
If i were to use Rew for AutoEQ which option i should select so i can import the EQ file directly? Or is there an option for that, it’s fine if i need to just replace some file in the system for eq to take effect.

1 Like

you mean float values of less than 1 as Q values of 1.8 and 2.5 etc all work ok for the peak

but yes I understand the the whole beta thing.

thanks and great work by the way.

You have just to select your headphone in the list and apply. This as simple :wink:

1 Like

Yes, you’re right. Less than 1. As I said I try to push a version tomorrow… :wink:

2 Likes

Qobuz tracks stop playing before the end of the song and next song in the queue doesn’t start playing from the queue when DSP is enabled

EDIT: this might be a Qobuz integration bug as no problems when playing files from USB drive or using Spotify connect.

2 Likes

Hi. Yes this a known problem. The root cause is not clearly identified, but we are working on it.

Hi. You have to export from REW as.txt and place the file in /Internal/Fusiondsp/peq.
And from the plugin, in PEQ mode, import local file. You import a first file for left channel with scope set to left, then move the switch to ‘add’ and import the second file with scope set to right.
I will write a guide later…
Let me know if you need more information :wink:

3 Likes

It should be now fixed with FusionDsp 0.0.9
Let me know :wink:

3 Likes

Great Job :+1: All values seem to enter without error.

As you know if you boost a particular fequency you must reduce the preamp to negative of whatever the maximum boost value is, to avoid clipping.

there is no PreAmp setting per se, just channel gain for left and right. this sounds correct for what a preamp does but when you enable and disable the effects it doesnt change the gain back.

is this by design.

The plugin auto set global attenuation.
For instance, if you set a gain at 10dB an other at 12dB, then -12dB is used +2 dB margin due to possible phase rotation (yes, even with only negative value, clipping may occur due to that). So there is always at least -2dB of attenuation when FusiiDsp is enabled.
So you don’t have to worry about that.
Attenuation for left or right channel is added to global attenuation is not dependent from values set in Es.
:wink:

1 Like

Marvelous :slight_smile: auto pre-amp adjust. I can dial back the channel gain, I did think it seemed over attenuated, but that explains that.

You fancy a suggestion for the future? it would be good to be able to enable and disable easily. Not sure if a button on the main Volumio screen would be possible, as no other plugins do that, but would it be possible to have a GPIO button activate and disable the dsp, pretty much like “GPIO buttons” plugin control play, pause etc.

Again you are doing a great job…Thanks

2 Likes

Hi- I have a question and maybe this isn’t the right thread, so let me know and I’ll post elsewhere. Am I correct that the the dsp would be able to play an LP (assuming there was an analog to digital interface) if the dsp curve were adjusted by the RIAA curve? In your approach, does the user generate the room correction curves and then have those further adjusted by any EQ the user wants to do? Can another step be application of RIAA curves? If so then you would effectively have a phono preamp with the addition of an analog to digital interface and maybe a moving coil step up transformer? Anyway, this could open up even more applications for Volumio

Hello.
At this point, RIIA curve is not supported by the plugin. My guess is that your LP output is already corrected for that before sending signal to the Dac or analogue output.
I don’t plan to add such feature as it is marginal IMHO.
Then, any input in Volumio, if you have a Myvolumio plan, goes through FusionDsp.
In FusionDsp, you can load room correction filters, or even creating them by loading the impulse for both speaker in the room.

1 Like

Thanks for your response…the input would only be adjusted for RIAA if run through phono preamp…digital phono preamps like the parks audio puffin offer multiple curves and dsp adjustments, but then if the signal is going to be run through a room correction DSP, you are processing the signal multiple times and potentially degrading it. Software like VynilStudio and ChannelD system’s pure vinyl allow users to make RIAA adjustments to digitally recorded tracks…the tracks can are generated by connecting a turntable to a digital interface (many are the same that are used for microphone recordings). Anyway I thought I’d inquire because it seemed to me that calculating the end curve inclusive of RIAA and processing once would be the best solution…and it also seems trivial, which it may not be…Your efforts are greatly appreciated thanks