Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Vocalpoint]

Finally found some time to test this out and while it seems that each "zone" has it's out set of audio controls (under Playback Options when I right click on WDTVLIVE under Playing Now)...I do not believe that any of these controls are actually doing anything (Output mode, track change (crossfades) and volume leveling)....

Comments?

VP

[mbagge]

When you are using DLNA for transport, MC works at the file level and no dsp function is performed. The files are simply streamed over the LAN/WLAN.
It would be a nice feature though if the DLNA protocol had commands for remotely controlling the DLNA renderer - your WDTVLIVE box.

[Vocalpoint]

When you are using DLNA for transport, MC works at the file level and no dsp function is performed. The files are simply streamed over the LAN/WLAN.
It would be a nice feature though if the DLNA protocol had commands for remotely controlling the DLNA renderer - your WDTVLIVE box.

So - conversely if I go to Tools->Options-Media Network->Add or Configure DLNA Servers and I see the Generic Server entry there...do these settings apply to how audio would be processed when "looking" for DLNA (Media) servers from the WDTV Live itself?

Because I do see a Volume Leveling checkbox under the Audio->Advanced area here. Maybe I going anout this backwards - maybe I should be streaming from media servers via the WDTV Live?

VP

[csimon]

There is only one audio-processing option for DLNA and that is to do volume levelling.  As you've correctly found out, this option is available in the DLNA settings menu, but note that you have to do conversion in order to allow volume levelling as otherwise the file is sent verbatim to the renderer.  MC has to do conversion in order to process the volume.  I think the options you get when right-clicking on the Zone are a mistake.

The DLNA settings are relevant for both pulling the media from the server and pushing it to the renderer, and for when you're using another device as a controller instead of MC itself.

[Vocalpoint]

There is only one audio-processing option for DLNA and that is to do volume levelling.  As you've correctly found out, this option is available in the DLNA settings menu, but note that you have to do conversion in order to allow volume levelling as otherwise the file is sent verbatim to the renderer.  MC has to do conversion in order to process the volume.  I think the options you get when right-clicking on the Zone are a mistake.

The DLNA settings are relevant for both pulling the media from the server and pushing it to the renderer, and for when you're using another device as a controller instead of MC itself.

csimon,

So if I am accessing a MC client via the WDTV Live menu and all that client has for music files is FLAC - are you saying I have to force MC to convert everything to say - MP3 before leveling will have any effect? Right now - I am sending everything FLAC based - right thru with no conversion.

But since the whole point of the WDTV is to if fact play high quality FLAC to my big media center with the big TV and big surround receiver - having to convert it down to some lossy format just to level the volume doesn't make a whole lotta "audiophile" sense to me. Will MC convert to say - uncompressed wav and still process the volume?

As it is - this is really only a problem for playlists where there is potential for wide volume swings but I would still like to investigate any possible way for MC to process (and then vol level) lossless files and keep them lossless for highest quality sound.

Cheers!

VP

[csimon]

Will MC convert to say - uncompressed wav and still process the volume?

Yes, that's one of the conversion options, you don't have to convert to MP3.

[Alex B]

You are using the playback engine in the player device, not MC's playback engine. The MC DLNA server just provides the file URLs. In an ideal world that playback device would read the replay gain tags and support volume leveling, but that isn't usually the case.

However, if your device supports uncompressed you can set the MC DLNA server to always convert to uncompressed and apply volume leveling. MC would then just decode the lossless files and adjust the volume. No lossy compression would happen.

EDIT: as scimon said. I was checking the DLNA options while scimon posted...

[Vocalpoint]

You are using the playback engine in the player device, not MC's playback engine. The MC DLNA server just provides the file URLs. In an ideal world that playback device would read the replay gain tags and support volume leveling, but that isn't usually the case.

However, if your device supports uncompressed you can set the MC DLNA server to always convert to uncompressed and apply volume leveling. MC would then just decode the lossless files and adjust the volume. No lossy compression would happen.

EDIT: as scimon said. I was checking the DLNA options while scimon posted...

Well - the WDTV Live supports uncompressed wav (WAV/PCM/LPCM) so I may be on my way here:)

Q: When you say "MC would then just decode the lossless files and adjust the volume" - what exactly is MC reading to "adjust" the volume - my meticulous Replay Gain tags already present within each FLAC file or some other method?

I am now excited to give this a shot later this evening and see what happens. Thanks guys!

VP

[csimon]

Yes, it's the Replay Gain tags.  If you rip CDs via MC then you can set it to analyse the file as ripping and automatically populate the Replay Gain tags, but I guess you would need all your files analysed by the same source as otherwise the reference level would be different?  I don't know if Replay Gain has a standard absolute reference point that is common to everything.

[Alex B]

... what exactly is MC reading to "adjust" the volume - my meticulous Replay Gain tags already present within each FLAC file or some other method?

... I guess you would need all your files analysed by the same source as otherwise the reference level would be different?  I don't know if Replay Gain has a standard absolute reference point that is common to everything.

By default all common Replay Gain tag analyzers use the same reference volume level when they write file tags that are in the de facto standard format. If there are implementations that use a different default reference I am not aware of them.

The displayed Replay Gain field value inside MC is 6 dB lower than the value in the FLAC file tag for historical reasons. Ten years ago JRiver Media Jukebox was the first player with a built-in RG analyzer and it was programmed to use the original, 83 dB SPL reference. Since then each MJ and MC version has maintained compatibility with that reference.

The other programs use the newer, 89 dB SPL reference standard. When MC reads or writes RG tags that are in the de facto standard format it adjusts the value by 6 dB to make it compatible with other implementations.

[mbagge]

...the whole point of the WDTV is to if fact play high quality FLAC to my big media center...

I am doing this for the same reason and I am fine using the remote for the receiver to adjust sound levels.
It is indeed confusing when you can do the same thing at different points in the music playing chain. If you were to control sound level in MC only, your power-on sound level on the receiver would be at 0Db - or full throttle. A rather risky setting I would think.

When choosing DLNA for transporting audio streams, in my oppinion, is just a little smarter than mounting CD's in the CD-player. Controlling the receiver is still a matter for the receiver itself.
If you want MC to control the sound in every acpect, you should connect your receiver by analog connections directly from your soundcard or some kind of high quality DAC.

[Vocalpoint]

Just an update - that this works beauty. I connected to my MC library from the WDTV Live (with MC serving my lossless FLACs using the following DLNA settings:

Audio: Conversion - Always Convert
Audio: Encoder: Uncompressed
Audio: Advanced: Volume Leveling - On
Audio: Advanced: Sample Rate - Same as Source


Then I just fired up a playlist that has all sorts of volume nonsense ranging from a very low in volume 1981 track to the latest Florence and the Machine which is so compressed and loud to almost be listenable...and the playback on the big rig via the WDTV required NO riding the control at all.

This is huge a break through for me since this volume wackiness has been driving me crazy for years...

Cheers!

VP

[srwooten]

Just an update - that this works beauty. I connected to my MC library from the WDTV Live (with MC serving my lossless FLACs using the following DLNA settings:

Audio: Conversion - Always Convert
Audio: Encoder: Uncompressed
Audio: Advanced: Volume Leveling - On
Audio: Advanced: Sample Rate - Same as Source


Then I just fired up a playlist that has all sorts of volume nonsense ranging from a very low in volume 1981 track to the latest Florence and the Machine which is so compressed and loud to almost be listenable...and the playback on the big rig via the WDTV required NO riding the control at all.

This is huge a break through for me since this volume wackiness has been driving me crazy for years...

Cheers!

VP

When I do this it changes my 24 bit music to 16 bit. Is there a way around this without upsampling everthing to 16 bit up to 24 bit?

[bob]

Try using the uncompressed 24 bit wave setting (if you've got MC17).

[srwooten]

Try using the uncompressed 24 bit wave setting (if you've got MC17).

That is what I'm doing, but doesn't that upsample all the 16 bit stuff to 24 bit?

[Alex B]

Upsampling is an incorrect term for this conversion. Zero-padding would be correct.

When a 16-bit signal is directly converted to 24-bit, the original 16-bit range is preserved as it is. It includes the volume level range from 0 to -96 dBFS. The eight additional bits are added "below" these 16 bits and they contain nothing.

In theory it is possible that when Replay Gain is applied and it reduces the volume level, the eight lower bits in the 24-bit output may actually preserve some very quiet signal that would be lost when outputted in the 16-bit mode. The additional bits include the volume level range from -96 to -144 dBFS. In any case such a volume level range is utterly inaudible in a real life listening situation.

[srwooten]

Upsampling is an incorrect term for this conversion. Zero-padding would be correct.

When a 16-bit signal is directly converted to 24-bit, the original 16-bit range is preserved as it is. It includes the volume level range from 0 to -96 dBFS. The eight additional bits are added "below" these 16 bits and they contain nothing.

In theory it is possible that when Replay Gain is applied and it reduces the volume level, the eight lower bits in the 24-bit output may actually preserve some very quiet signal that would be lost when outputted in the 16-bit mode. The additional bits include the volume level range from -96 to -144 dBFS. In any case such a volume level range is utterly inaudible in a real life listening situation.

Thanks Alex, for your detailed reply.