levelling volume of music on NAS

On dbPoweramp you need to rip with the replay gain DSP. I know on a Sonos system only FLAC supports replay gain on playback. So I would first check it is supported be the client (UnitiQute) in this case and which file format.

Unfortunately Graham the UniQute doesn't have a volume knob.

Thanks Mike, from looking at the dBpoweramp help I think volume normalise is what I'm looking for, which does not require Id tags.

are there any other ways of doing this?

particularly with the tracks I've already ripped?

SJB

is no-one else having a issue with the volume differences between a 1989 cd and a 2010 cd when both occur in a playlist?

If so I must be missing some setting?

Can UniQute and siblings read a "replay gain" ID tag?

SJB

As someone yet to even rip his frst disc via a Vortxbox, this is something which had never occurred to me

As far I know when ripping, the Vortexbox sets the FLAC Replay Gain tag to normalise the volume across rips. If you have imported the files, as I did from another ripper then you simply type at the command line and hit return

nohup apply_replaygain.pl --processall --info /storage/music/flac > /dev/null 2>&1 &

This will automatically go through every FLAC file in your music folder and recalculate the correct value for the Replay Gain setting to normalise the Replay Gain setting. The active part of the command is in bold. The "nohup" means the program will still run if you log out, "> /dev/null" bit just tells it to shut up about what its doing otherwise it'll write it all over the screen (i.e. your terminal window from your Mac in to the VB; assume you can run telnet from Windows too, but not sure), the "&" at the end tells it to run in the background so you can still use your VB.

Of course, it assumes the music player will recognise the Replay Gain tag. Sonos does. I think the UQ does too, as I've not heard anything that makes me think otherwise such as an album suddenly sounding much louder than expected.  

However if you rip with the VB then most of this is done without you ever knowing or needing to know. This is simply a way of re-applying it to tracks with the wrong settings that were ripped elsewhere. 

Sorry, I don't know how to do this for any other NAS or non-Apple environment, but I think dBPowerAmp can retrofit it for PC users. 

Well I hit the support email with the following

Hi,

I use dBpoweramp ro rip my music that is streamed to my UniQute.

There is a setting (ID Tag) called replay gain (similar to sound check
on IPOD) - does UniQute process this TAG (I use FLAC) and make my 1990
CDs sound a bit louder when they come in a random selection?

Thanks


and got this reply

Hello

Sorry no it does not

Regards



Steven Hopkins
Customer Relations Manager

which answers my question.


I'll just have to stop putting Leonard Cohen songs on my playlists!

SJB

Is this still the case or with UServe now ripping to flac as well as wav can current Naim streamers read replay gain tags?

One of the beauties of streaming is random playlists, years, genres etc. and with my SBT>nDAC all these are played at a similar volume, whereas my UniQute gives a lecture on the loudness wars.

SJB

It does however have a remote control with volume control on it.  I sit next to mine.

The 'volumes' on the CD rips on my NAS vary greatly but a couple of clicks on the remote volume sorts it quick enough.

Hi everyone

Just on a sort of related note. A couple of years ago all my music was in mp3 format (not anymore since I got the UQ as I re-ripped them) and I ran a program to level out the volume of all the tracks in my library having the same problem with volume  levels, It did just that. The problem was that when I went to listen to something like Pink Floyd's The Wall where some tracks are intentionally quiet and some loud it was a complete mess.

Trust me there's no point having gapless playback if the track which follows jumps a couple of decibels. Just a word of caution.

Russell

Do not use Normalise. This physically processes the data so your rip will not be the same as the CD.

Use ReplayGain - I do this after ripping with MediaMonkey. This tags the music with the gain needed.

Use the option to apply gain based on album and not track to avoid Sorren's issue.

And do not use Sound Check on an iPod because it kills the dynamics.

I agree, I do not use normalising.. It messes with the PCM samples, and as noted earlier can wreck the dynamic and feel of an album, not to mention causing level jumps on gapless playback which undermines the benefit of using a quality playback system such as Naim.

if you want a more consistent sound level it is more effective to compress the audio rather than normalise, but that is more complicated, and significantly messes with the audio... probably why I think it's best to leave it to the sound engineer on the master.

BTW normalising a low level file can amplify quantisation errors (the digital equivalent of hiss), and make the audio sound less clean.

Normaling is most appropriate for already highly compressed individual 'single' tracks where there are only very small differences between full scale peak data values between the tracks. 

Simon

I resurrected an old thread rather than starting a new one. It is "replay gain" that I use as I rip which is a tag that the squeezebox can read and react to accordingly. I use the "smart gain" setting on the Touch which seems to work well in all scenarios. 

I'm wondering whether the latest Naim streamers can read these tags?

SJB

I doubt it, as it is non standard - and does compromise sound quality ultimately. If Naim did support there would almost certainly be a UI option to disable.

Simon

Another way to support this is to round-off the replay gain value and nudge the (digitally-controlled analogue) volume accordingly.  I use album-wide gain rather than track gain, largely because many of my albums cross-fade between tracks.

i don't own a qute but surely it has a volume control i thought it was the logo? Surely it comes with a remote anyhow?

for all the efforts to rip files, any form of volume control post rip is taking you away from a proper linelevel  out that is to say you will be processing the sound. This is the same for say itunes, at max volume its line level, anything below its being processed just as surely as sending it through a graphic equaliser.

I agree with garyi - if a track Is louder on my qute I just lower the volume.  It took me ages to rip all my CDs as perfect rips I wouldn't want to compromise them at the source?

I have to agree -- do a bit-perfect rip, and then adjust from there during playback.

On the Squeezebox you can turn it off if you are really critically listening and want to hear those bit perfect rips. But it's nice that it's there. Once naim had no remote control,  no headphone socket etc. the world didn't fall apart once those innovations were embrassed by naim. 

Obviously means nowt to you lads but it is one of the main reasons ( with Spotify) that I steer clear from having for my main streaming source a naim product. 

Obviously not not much has changed in the 2 years since my first post. 

SJB

...it's worth pointing out that this is passively done at the renderer and that the rips are saved as bit-perfect–they are only tagged with replay gain, not re-processed.

Exactly. It's just another tag. You can do it at rip time with dbpoweramp if you want.

But when the renderer applies the replay gain you have lost your bit-perfectness and lost some of the sound quality that was there before applying it.

the renderer also works in the digital domain and any volume modification in the digital domain on the same bit-depth will compromise the sound quality.

the only way to modify volume without reducing sound quality is by increasing bit-depth and do the modifications in that extra space. So the renderer needs to increase internally the bit-depth of 16-bit to say 24-bit or 24-bit music to say 32-bit. For each extra bit there is 6dB volume adjustment available before compromising sound quality.

volume adjustments are best done in the analogue domain.

cheers

Aleg

If you consider the DAC to be a two stage process (simplification I know) of turning digital into analogue and then applying gain so that output is at line level, then the gain is simply controlled in the second part.

And do DACs work like that (if they are aware of RG at all)?

BTW the OP talked about RG on his NAS.

cheers

Aleg