Thank you for your donation!


Cloudsmith graciously provides open-source package management and distribution for our project.


BruteFir integration
#11
CamillaDSP would be the best choice but it's not clear how much work would be needed to integrate it. Maybe another dev will volunteer to do it.

What can't u do with the 4-band parametric? I think recording and mastering studios typically use a 4 or 5 band PEQ but I haven't kept up with the DAW space for a while now so ...
Enjoy the Music!
moodeaudio.org | Mastodon Feed | GitHub
Reply
#12
(10-03-2020, 01:34 PM)Tim Curtis Wrote: CamillaDSP would be the best choice but it's not clear how much work would be needed to integrate it. Maybe another dev will volunteer to do it.

What can't u do with the 4-band parametric? I think recording and mastering studios typically use a 4 or 5 band PEQ but I haven't kept up with the DAW space for a while now so ...

You may be aware of Jaako Pasanen's AutoEQ project for automated equalization of headphones based on measurements available on the net or using one's own measurements. As he says in the readme:


Quote:Keep in mind that parametric eq accuracy depends on the number of filters available. Usually 10 filters produce very good results but as little as 5 can be good enough.

Now four PEQs are certainly better than nothing but in this case more is definitely better. I have a MiniDSP Ears headphone measuring rig and a MinDSP 2x4HD that can use 10 PEQs on both the input and output and I assure you a 10 band EQ can sound a lot better than a 4 band EQ, especially on 'difficult' headphones. On some 'phones I could use even more than 10 bands.

So why not just use the MiniDSP you may ask? Well, I am! But I would prefer to use a different DAC, plus the 2x4HD isn't the prettiest device to have sitting on my desk. Also the 2x4HD has only 4 preset slots and changing presets is a bit clunky. So being able to do the full EQ in software on the Pi (preferably in Mooode) would be most welcome.
Reply
#13
Thats a cool project. The only EQ he mentions for Linux is one based on the Pulseaudio system. moOde uses ALSA directly and so not compatible with Pulse.

The PEQ used in moOde is from the CAPS project
http://quitte.de/dsp/caps.html#EqFA4p

Maybe a dev will volunteer to modify the sources and add some bands. I could then add them in the UI.
Enjoy the Music!
moodeaudio.org | Mastodon Feed | GitHub
Reply
#14
(10-04-2020, 01:15 PM)Tim Curtis Wrote: Thats a cool project. The only EQ he mentions for Linux is one based on the Pulseaudio system. moOde uses ALSA directly and so not compatible with Pulse.

The PEQ used in moOde is from the CAPS project
http://quitte.de/dsp/caps.html#EqFA4p

Maybe a dev will volunteer to modify the sources and add some bands. I could then add them in the UI.

I picked up the gauntlet and based on CAPS created a 12 band parametric EQ.
It is based on the EqFA4p and is called EqFA12p. Configuration is just like EqFA4p but with more bands.

EqFA4p exists out of a 4-way parametric parallel Mitra-Regalia filter.
EqFA12p exists out of a 3 times a 4-way parametric parallel Mitra-Regalia filter in serie.

The code branch can be found here:
https://github.com/bitkeeper/caps/tree/f...ametric_eq
Reply
#15
The CAPS project EqFA4p filters makes use of a bw (bandwitdh) parameter instead of the Q factor.
In the past a few asked question about the exact meaning of bw, but I didn't see the answer.
Did anybody make progress on that 

It look like it doesn't match at the classic form as in:
    bw = fc /Q. Where the bandwitdh is the difference in the upper and lower frequentie at -3 db from fc.

Attached is a measurement with:
fc= 1kHz, gain = 6 db, overall gain -3db  and a bw parameter of 0.5 and 1.

It seems that the EqFa4p bw parameters more looks like the classical bw/2.
Pratical that means that Q can't be higher then 4 (lower limit of the filter is bw=0.125)

If we do the math the following table gives an identication about how Q relates to bw parameter of eqfa:
Code:
fc=1000.00Hz Q=8.00 bw=0.12 fl=939.5 fh=1064.5 fd=125 bw_eqfa=0.06
fc=1000.00Hz Q=4.00 bw=0.25 fl=882.8 fh=1132.8 fd=250 bw_eqfa=0.12
fc=1000.00Hz Q=2.00 bw=0.50 fl=780.8 fh=1280.8 fd=500 bw_eqfa=0.25
fc=1000.00Hz Q=1.00 bw=1.00 fl=618.0 fh=1618.0 fd=1000 bw_eqfa=0.50
fc=1000.00Hz Q=0.50 bw=2.00 fl=414.2 fh=2414.2 fd=2000 bw_eqfa=1.00
fc=1000.00Hz Q=0.25 bw=4.00 fl=236.1 fh=4236.1 fd=4000 bw_eqfa=2.00
fc=1000.00Hz Q=0.12 bw=8.00 fl=123.1 fh=8123.1 fd=8000 bw_eqfa=4.00

Now at least we can calculate the correct parameters from the AutoEq project to setup EqFa4p based filters, like the EqFa12p.
Reply
#16
(10-06-2020, 09:36 AM)bitlab Wrote: I picked up the gauntlet and based on CAPS created a 12 band parametric EQ.
It is based on the EqFA4p and is called EqFA12p. Configuration is just like EqFA4p but with more bands.

Thanks for this. I look forward to it making its way into Moode at some point. Sadly, compiling from source is beyond my Linux fu at the moment.
Reply
#17
(10-08-2020, 07:06 PM)bitlab Wrote: Pratical that means that Q can't be higher then 4 (lower limit of the filter is bw=0.125)

Hmm. That's a pity. While I try to avoid high-Q filters I do find I need a few filters above Q4 ...

There no way round this limitation I take it with the CAPS EQ as base?
Reply
#18
(10-20-2020, 12:24 PM)bluefuzz Wrote:
(10-08-2020, 07:06 PM)bitlab Wrote: Pratical that means that Q can't be higher then 4 (lower limit of the filter is bw=0.125)

Hmm. That's a pity. While I try to avoid high-Q filters I do find I need a few filters above Q4 ...

There no way round this limitation I take it with the CAPS EQ as base?

We could try out what happens if we cross the borders. 
Maybe it is only used to limit the inputrange and not a technical reason.
Reply
#19
I have increased the range of freq from 20-14000 to 10-18000 and from bw to 0.06-16 of the EqFa12p filter. 
Validation is shown in the attached image.

@bluefuzz is the Q of 8 enough ?


Attached Files Thumbnail(s)
   
Reply
#20
(10-08-2020, 07:06 PM)bitlab Wrote: The CAPS project EqFA4p filters makes use of a bw (bandwitdh) parameter instead of the Q factor.
In the past a few asked question about the exact meaning of bw, but I didn't see the answer.
Did anybody make progress on that 


I may be completely wrong, but I seem to remember reading on some audio forum about the bw parameter in EqFA4p simply being the bandwidth in octaves (there are online converters for converting Q factor to bandwidth in octaves). Is that possible?

PS: @bitlab thanks for your great work!
Reply


Forum Jump: