Print

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

[richardap]

I have had a look around and seen this mentioned before. I noticed some changes in the changelog for V16 but wasn't sure if this was possible.  Is this possible or is there a standard workaround?
Cheers

[justsomeguy]

Ya it's been brought up several times in the past years but is either very very low priority or may just never happen, not sure. Couple ways around it though.

1) You could sync to your device and then use foobar2k to apply the leveling to the files on that device.
2) You could sync to a folder on the computer then run foobar on it then copy those to the device.
3) Something else I've done. Create a playlist in MC and add your tracks you want to sync. Export that playlist in M3U format. Import that list into Mediamokey using the ImportM3U script then use the sync from that. That program will volume level on sync and it's syncing to devices is extremely configurable. It's sounds like more work but is very flexible and currently the way I do it for syncing to my flash drive for use in my car.

A couple years back I tried to just use mediamonkey exclusively for my music collection but ended up coming back to MC. While the syncing works more like I want it in another program, everything else MC does is just so much better that I couldn't stay away.

[richardap]

Thanks for that.
I think i will give option 2 a try. I already use MediaMonkey but am finding it slightly unstable, but you're right, the sync options are great.

[rick.ca]

Ya it's been brought up several times in the past years but is either very very low priority or may just never happen, not sure.

I don't know why that would be. Surely the need to sync to devices that don't do volume levelling isn't that uncommon. Ah, here's something...

We agree this would be nice.

The conversion system is slated for numerous improvements, like image conversion for library server / WebPlay, better streaming video conversion, better management of cached low-bitrate copies, and enabling of DSP Studio during manual conversion.

We'll add DSP during portable sync to the list.  As a disclaimer, it's going to be a while before we're through the list.

Not sure where we are in that list, but we must be getting closer.

[cncb]

I wouldn't bet on it, hold your breath, etc.:  http://yabb.jriver.com/interact/index.php?topic=63384.0

[rick.ca]

It has been a long time, but I don't see now a non-response to a question already answered changes anything. And patiently waiting is so much easier than installing a carputer, and cheaper than buying a new car (which may still not have a player with volume levelling).

[JimH]

I wouldn't bet on it, hold your breath, etc.:  http://yabb.jriver.com/interact/index.php?topic=63384.0

Send a blank check.  I'd like to hire three more people.

[justsomeguy]

I remember that thread you quoted Rick.

We agree this would be nice.

The conversion system is slated for numerous improvements, like image conversion for library server / WebPlay, better streaming video conversion, better management of cached low-bitrate copies, and enabling of DSP Studio during manual conversion.

We'll add DSP during portable sync to the list.  As a disclaimer, it's going to be a while before we're through the list.

It sounds good and would be a welcome addition. I'm just concerned that just adding DSP during sync may not really accomplish what some have asked for. If I understand right, just adding dsp effects would require that the audio, say mp3, would have to be decoded leveled then re-encoded. Definitely not optimal. What would be best is to implement something similar to the way mp3gain or foobar can directly modify the mp3 without re-encoding, avoiding quality loss.

[BullishDad]

I've asked for this too.  Many devices don't use replay gain info, so it would be nice to output volume levelled MP3 files to a USB flash drive or CD for use in a car, etc.

I use the program MP3 Gain to accomplish this, but I'd rather let MC take care of it.  Don't need the DSP effects, just the volume levelling.

[rick.ca]

What would be best is to implement something similar to the way mp3gain or foobar can directly modify the mp3 without re-encoding, avoiding quality loss.

I found this difficult to grasp at first. That is, how can setting replay gain with mp3gain help when the problem is the player is not replay gain aware? If I understand correctly, the explanation is it's adjusting a gain field of every block of audio data, not just writing a Replay gain tag. I have no technical understanding how audio data is encoded, so I can only accept that on faith. I'm not familiar with the implementation, but obviously this would be lossless as long as there's a way to reverse the adjustment.

I understand this would be an effective way to adjust the replay gain for files to be played in something not replay gain aware. But it seems there are issues with mp3gain writing it's global replay gain adjustment as an APE tag, rather than changing an existing ID3 tag. I'm sure such issues have solutions, but it suggests all this might be a little more complicated than it may seem. So maybe it's just as well this was added to the end of the list. Gives you time to talk Matt into implementing some kind of mp3gain-like solution.

Personally, I don't see much reason to care. In most cases, I would be re-encoding from FLAC (or high bitrate MP3) to MP3 anyway. And where that's not the case, I'm not likely to care about some marginal loss in quality in an already crappy file for which I'm not likely to hear the difference in the car anyway. Also, I assume a mp3gain-like adjustment has no encoding speed advantage, since the whole file has to be written.

[justsomeguy]

Here an explanation of what mp3gain is doing. http://wiki.hydrogenaudio.org/index.php?title=MP3Gain

It's not setting the a replay gain tag but setting a gain multiplier per frame of data in the mp3. This is completely reversible as long as the multiplier used is recorded some how. That is what is stored in the APE tag. The tag has nothing to do with the volume of the mp3 and is only there to reverse the process. It is a very fast thing as well. I just used foobar to apply those changes to 15 mp3s and it literally took no more than 3 seconds.

If your files are lossless (flac), then ya decoding leveling the encoding isn't an issue.

[richardap]

Looking at the setting for burning a CD, is Volume leveling applied here?

[justsomeguy]

Yes, if burning an audio cd the volume leveling in the dsp can be applied. Burning a data cd with mp3's on it, won't have this applied.

[rick.ca]

Here an explanation of what mp3gain is doing.

Thanks. That one's more concise than what I found.

The tag has nothing to do with the volume of the mp3 and is only there to reverse the process.

What happens when these files are played in a player that is able to do volume levelling based on the Replay Gain tag. Is the player not going to read the replay gain originally recorded in the ID3 tag and apply an adjustment that has already been made by mp3gain?

[BryanC]

Thanks. That one's more concise than what I found.

What happens when these files are played in a player that is able to do volume levelling based on the Replay Gain tag. Is the player not going to read the replay gain originally recorded in the ID3 tag and apply an adjustment that has already been made by mp3gain?

MP3gain should set the replaygain values to zero.

[rick.ca]

MP3gain should set the replaygain values to zero.

That's what I would have assumed. But in trying to learn about mp3gain, I found myself wading through a lot of discussion that seemed to have something to do with this. It's quite possible I wasn't reading carefully enough, or it was an old issue since resolved.

I still wonder about the fact the adjustment precision is limited to 1.5 dB. If the Replay Gain is set to zero, isn't some accuracy of the original analysis lost?

[Alex B]

Both foobar2000 and MP3Gain correct the Replay Gain tag values automatically when they are used to adjust the physical volume level in MP3 files. For example:
- the original Replay Gain value: -5.0 dB
- the physical volume level correction by adjusting the scale factor in each MP3 frame: -4.5 dB (only 1.5 dB steps are possible)
- the new Replay Gain value -0.5 dB

However, MP3Gain reads and writes only the APE tag format and many players do not support APE tags in MP3 files. It is possible to disable tag writing in MP3Gain, adjust the file volume levels and after that create new correct Replay Gain tags in ID3v2 format by analyzing the changed files in MC. foobar2000 writes ID3v2 by default.

[rick.ca]

However, MP3Gain reads and writes only the APE tag format and many players do not support APE tags in MP3 files. It is possible to disable tag writing in MP3Gain, adjust the file volume levels and after that create new correct Replay Gain tags in ID3v2 format by analyzing the changed files in MC. foobar2000 writes ID3v2 by default.

Thanks. I believe that's a concise description of the issue I was reading about, but was unsure I understood. So there is a potential for confusion if users of such a feature don't fully understand the implications. But in an MC implementation, would the solution not be to simply change the ID3 tag if one exists?

[Alex B]

Thanks. I believe that's a concise description of the issue I was reading about, but was unsure I understood. So there is a potential for confusion if users of such a feature don't fully understand the implications.

 I agree.

Perhaps a description of my workflow will lessen the potential for confusion:

- I have a single "virtual handheld" in MC16 that syncs files to a disk location. I use it manually in the "drop files" mode. Usually I create playlists and sync them (I don't want to sync each and every file). Naturally I also have the conversion cache feature disabled in MC's File Location options to avoid additional copies of the files (this "virtual handheld" is my cache). The files in other formats are converted to MP3 and if the format is already MP3 the files are just be copied (this is one of the possible conversion modes).

- I fix the file volume levels with MP3Gain. This works fine when I simply drag and drop the Handheld folder from Windows Explorer to MP3Gain. When MP3Gain runs it skips the files that are already adjusted.

- I sync various devices and burn MP3 CDs from this "virtual handheld" with MC15. The location is not configured as a Handheld in MC15. I have simply set its Auto-Import to watch only the "virtual handheld" location. I use MC15 because syncing files from one handheld to other handhelds is not really usable in a single MC instance. Also using separate libraries would not help because the handhelds are not library specific. Alternatively I could use a portable MC16 instance because it would be completely separate from the standard MC16 installation.

- I could reanalyze the volume leveled files with MC15, but so far that has been unnecessary because I don't have portable devices or MP3 CD players that can read Replay Gain tags.

EDIT

But in an MC implementation, would the solution not be to simply change the ID3 tag if one exists?

An ideal MC implementation would apply DSP gain correction when a file is converted to MP3 (or to any other format) and adjust the scale factor values inside the MP3 frames in the resulting files when the source file is already in MP3 format and it is not re-encoded. Fixing the Replay Gain tags accordingly would naturally be important (and perhaps the easiest part to implement).

[rick.ca]

Perhaps a description of my workflow will lessen the potential for confusion:

You're kidding, of course.

But an excellent case for adding mp3gain and a more versatile syncing functionality. You shouldn't have to use a third party tool and two versions of the program for something so basic.

I could reanalyze the volume leveled files with MC15, but so far that has been unnecessary because I don't have portable devices or MP3 CD players that can read Replay Gain tags.

I realize it doesn't matter in this common scenario, but am I correct in assuming it wouldn't hurt if an integrated volume levelling function simply changed the tag? It seems to me it then wouldn't matter how these files were used. It could even be applied to original files, and it wouldn't matter if some files where adjusted and others were not (even files from the same album). MC (or any other replay gain aware player) would still be able to level volume properly. I realize this is not something users would likely want to do, but it does suggest the feature can be implemented in a way that doesn't create problems or confusion.

[justsomeguy]

Alex's description of his workflow for accomplishing this is pretty representative of the type and number of steps needed to accomplish this task no matter how you go about it. I have a different method I use but the time and number of steps are similar I'm sure. It really annoyed me having to jump through these extra hoops to do something I was surprised a program with so many other features couldn't do on its own. I eventually got over it and just accepted the way it was but it would be very nice to have it changed.