EQ in LMS/pCP?

[charleski]

piCorePlayer also uses LMS 8.3.0, so it's not that. Can you zip up the custom-convert.conf file you have and attach it?

 

[hoverdonkey]

Thank you.
Yes, I always restart the Logitechmediaserver service in DietPi. I’ve tried your suggestion “# F” without “T…”. That doesn’t make a difference. I doubt the problem I encounter has anything to do with the custom-convert.conf file itself. It is recognised, but rejected. The “flac ()” on my system must be the problem, I suppose. Error message I see in the log is “Error: Didn't find any command matches for type: flc”.

You are on piCorePlayer? I’m on DietPi running LMS 8.3.0. Maybe there are differences.

I'm using DietPi. My previously posted conf file contents work.

Dare I say it.. check the indents again now you have fixed the FILE typo?

 

[willefg]

piCorePlayer also uses LMS 8.3.0, so it's not that. Can you zip up the custom-convert.conf file you have and attach it?

 

[willefg]

I'm using DietPi. My previously posted conf file contents work.

Dare I say it.. check the indents again now you have fixed the FILE typo?

I tried all sorts of indents including the ones you described, but nothing changed. Last thing I tried was taking away the space between “# FT..” to ”#FT…”. That did change the “flac ()” into “flac (I,F)” but then the logs show it was rejected because of “T”. It’s definitely something in custom-convert.conf because that’s what creates the “flac ()” entry.

 

[charleski]

Bad news I'm afraid. I uploaded the exact custom_covert.conf file you attached to my LMS, and it seems to work on pCP with (F) showing up:

(I'm currently pottering around trying to get camilladsp working, which is why the others aren't disabled.)

Maybe try

# EQ setting for Sennheiser HD 650 (oratory1990)
flc flc * dc:a6:32:2f:9c:16
        # IFT:{START=--skip=%t}U:{END=--until=%v}
        [flac] -dcs $START$ $END$ -- $FILE$ | [sox] -q -t wav - -b 24 -t flac - gain -29.3 equalizer 20 1.1q +4 equalizer 97 0.7q -2.5 bass +5.5 105 0.71q equalizer 215 1.1q -1.8 equalizer 1300 1.5q -1.5 treble +3 2000 0.71q equalizer 2700 3q -1.3 equalizer 3300 2.7q -3.3 equalizer 5350 3q -2.5 treble -3 11000 0.71q

or

# EQ setting for Sennheiser HD 650 (oratory1990)
flc flc * dc:a6:32:2f:9c:16
        # FT:{START=--skip=%t}U:{END=--until=%v}
        [flac] -dcs $START$ $END$ $FILE$ | [sox] -q -t wav - -b 24 -t flac - gain -29.3 equalizer 20 1.1q +4 equalizer 97 0.7q -2.5 bass +5.5 105 0.71q equalizer 215 1.1q -1.8 equalizer 1300 1.5q -1.5 treble +3 2000 0.71q equalizer 2700 3q -1.3 equalizer 3300 2.7q -3.3 equalizer 5350 3q -2.5 treble -3 11000 0.71q

To be honest I'm not at all sure what the '--' in the flac commandline means, but it might be a knob worth twiddling.

 

[mctron]

Is it trivial to add support for Spotify streaming via LMS (which uses Ogg Vorbis)?

 

[charleski]

Is it trivial to add support for Spotify streaming via LMS (which uses Ogg Vorbis)?

I see there are a bunch of 3rd party Spotify plugins for LMS. No idea how easy they are to use.

 

[mctron]

I see there are a bunch of 3rd party Spotify plugins for LMS. No idea how easy they are to use.

The Spotty plugin itself is fine! Just wondering about the EQ part. Assuming you would just need to put an entry for ogg with the proper decoder as the Spotify implementation uses ogg, but I'm not sure

 

[charleski]

Since sox can decode ogg directly you just need something like
ogg flc * *
# IF
[sox] -t ogg $FILE$ -t flac -C 0 - [sox switches to define your EQ ]

 

[willefg]

Bad news I'm afraid. I uploaded the exact custom_covert.conf file you attached to my LMS, and it seems to work on pCP with (F) showing up:
View attachment 258711

(I'm currently pottering around trying to get camilladsp working, which is why the others aren't disabled.)

Maybe try

# EQ setting for Sennheiser HD 650 (oratory1990)
flc flc * dc:a6:32:2f:9c:16
        # IFT:{START=--skip=%t}U:{END=--until=%v}
        [flac] -dcs $START$ $END$ -- $FILE$ | [sox] -q -t wav - -b 24 -t flac - gain -29.3 equalizer 20 1.1q +4 equalizer 97 0.7q -2.5 bass +5.5 105 0.71q equalizer 215 1.1q -1.8 equalizer 1300 1.5q -1.5 treble +3 2000 0.71q equalizer 2700 3q -1.3 equalizer 3300 2.7q -3.3 equalizer 5350 3q -2.5 treble -3 11000 0.71q

or

# EQ setting for Sennheiser HD 650 (oratory1990)
flc flc * dc:a6:32:2f:9c:16
        # FT:{START=--skip=%t}U:{END=--until=%v}
        [flac] -dcs $START$ $END$ $FILE$ | [sox] -q -t wav - -b 24 -t flac - gain -29.3 equalizer 20 1.1q +4 equalizer 97 0.7q -2.5 bass +5.5 105 0.71q equalizer 215 1.1q -1.8 equalizer 1300 1.5q -1.5 treble +3 2000 0.71q equalizer 2700 3q -1.3 equalizer 3300 2.7q -3.3 equalizer 5350 3q -2.5 treble -3 11000 0.71q

To be honest I'm not at all sure what the '--' in the flac commandline means, but it might be a knob worth twiddling.

Thank you for all your time and effort!!

I tried both options but nothing. My FLAC for SoX stays at flac(), unfortunately. When I look at the logs it does mention:

Slim:layer::TranscodingHelper::_getCapabilities (197) Capabilities for flc-flc-*-dc:a6:32:2f:9c:16: syntax error in FT:{START=--skip=%t}U:{END=--until=%v}

Later it sees the EQ settings but they are immediately rejected. Looks like there’s a difference in settings regarding File Types and SoX between the LMS distro in DietPi and the one used in pCp.

CamillaDSP directly on the player instead of the server would definitely be better for me. Especially if I could use the CamillaDSP GUI and change setting without SSH access.

Let us know how you get on with it. I’ve seen on LMS forum that more people are working on Parametric EQ within LMS or Squeezelite players. Would make LMS more of a Roon killer than it already is.

In the meantime, if anybody can shed a light on this SoX EQ issue: thanks in advance…

 

[charleski]

CamillaDSP directly on the player instead of the server would definitely be better for me. Especially if I could use the CamillaDSP GUI and change setting without SSH access.

Let us know how you get on with it. I’ve seen on LMS forum that more people are working on Parametric EQ within LMS or Squeezelite players. Would make LMS more of a Roon killer than it already is.

I started out trying the script and hack from here. But the python script never really worked for me (and the instructions had me running around in circles ...), so I switched to using the method described here (which is alluded to in the first link). It took several hours (mostly because I was wasting time with the python script), but it's working fine now.

Since Camilladsp requires a fixed input samplerate, these methods automate a process in which it's restarted and set to the correct rate whenever it changes. I had to switch to the 32bit version of pCP for this, but I think someone's working on getting it to function on a 64bit kernel.

The big advantage of using camilladsp is that you get proper gapless playback, which is the whole reason I switched over to LMS in the first place as all mpd/ffmpeg systems are broken in this regard.

 

[MCH]

CamillaDSP directly on the player instead of the server would definitely be better for me. Especially if I could use the CamillaDSP GUI and change setting without SSH access.

You can change many things (filters, levels, inputs, whole configs....) in camilladsp at the flick of a regular remote control switch using the pyCamillaDSP python library. You just need some fiddling with python....

 

[mctron]

CamillaDSP directly on the player instead of the server would definitely be better for me. Especially if I could use the CamillaDSP GUI and change setting without SSH access.

If you are willing to use AirPlay Bridge workaround via Moode I mentioned in post 38 you can have all of these things and do EQ on the client. Moode can apply CamillaDSP same as PEQ you just need to configure it (many tutorials for that of course).