MP3Gain Usage Guide

[jaidog]

mp3Gain doesn't seem to be helping with Nano

I'm attempting to use mp3Gain with unsuccessful results.
Here's what I did:

- I used iTunes to convert about 50 songs from CDs to mp3s.
- I opened up up mp3Gain.
- From iTunes, I dragged these 50 mp3s into mp3Gain.
- I used mp3Gain to normalize the songs to 89 db.
- I noticed that the timestamps on these mp3 files
got updated in my iTunes window.
- I created a playlist with these 50 songs.
- I connected up my Nano to the PC and Sync'd the
playlist to the Nano.

Now, when I connect the Nano to my stereo and play the
songs, some are definitely louder than others. Not sure
if Soundcheck was on or off in iTunes, but it is definitely
off on the Nano.

What am I doing wrong?


Thanks in advance for any advice.

 

[Germansuplex]

If you have Soundcheck on in iTunes before importing (or ripping) your CDs/mp3's, iTunes is going to apply a "formula" in its own special tag based on the volume of the music, to make them all play at the same perceived volume. If you then run mp3gain, then iTunes is applying the Soundcheck to the original volume, not the gained volume, so you may still get different volumes. Turn Soundcheck off on your iPod/iPhone and see if they sound the same.

If you turned Soundcheck on in iTunes AFTER using mp3gain, the iTunes will apply the soundcheck volume based on the mp3gain values and so, with soundcheck on or off on your iPod, they should still play at the same volume.

If not, then maybe you used album gain and there may be slight volume differences between track since album gain takes songs and applies a relative volume change to them, so that the volume difference between tracks remains.

I can't remember if mp3gain reads the tag info to apply album gain, or if it just assumes all tracks in the same folder/location are part of an album. If you don't have iTunes organizing your library and just have a bunch of random mp3's in a folder and apply album gain, you may end up with varying volume differences.

Lastly, its possible you did everything right but some volumes are just different. Even if two tracks both are set to 89db, a soft piano and flute jazz track will never be as "loud" as a blistering metal track with tons of drum licks and wailing, screaming guitars.

What I do is set all my albums to 89db with mp3gain's album gain feature, then run iVolume to gain the tracks to 95db. If I have Soundcheck on, all of my tracks play quite loudly at 95db. If I have Soundcheck off, they all play back at around 89db (and if I listen to a full album, all the relative volume differences are intact).

I would stick with kornchild's advice of not using both Soundcheck and mp3gain, but iVolume is a paid app and will override iTunes' Soundcheck values with its own algorithm, so whether or not you use mp3gain before or after importing to iTunes with Soundcheck on is irrelevant, since iVolume will look at mp3gain's tag and base its Soundcheck volume on that.

 

[jaidog]

Germansuplex,

Thanks for all the advice. I decided to re-perform my steps using
mp3Gain and now all sounds well. It may have been one of two
problems. The first thing that may have happened during my initial
try is that I hit the button on mp3Gain to calculate all of the adjustments,
but forgot to hit the button that applies these adjustments. The second
culprit may have been that my iPod already had some of the songs on
it which were in my playlist, and these songs it contained were't
mp3Gain'd. So, I cleared all songs off of my Ipod the second time
around, and then sync'd up the playlist that was mp3Gain'd. Now,
all is well.

Thanks again.

 

[Boogie]

Well, I've waited...

I can't put it off any longer, I'll have to ask for help from one of you guys.

I read the guidance notes and threads with the intention of doing this once only, Hmmm. That worked out well!

Here's my problem:

I reset my iPod Touch to ensure that all music files were removed from the player. I also took the opportunity to move the Library from the HDD of my laptop to an external HDD with a little more headroom.

Once the library was loaded on the new drive I ran MP3 Gain.

Then I reassigned the new library location with iTunes, hooked up the Touch and loaded up my files.

Three tracks in to the first timeI play it and I find there's a huge difference between volumes, and I mean huge.

Is there something glaringly obvious that I've done?
Any help would be gratefully received as it's driving me scatty.

Thankyou

 

[Code Monkey]

Boogie, first, since you read the guides I assume you have soundcheck off. So, my next thought is to ask what interface are you using? The mp3GainGui has issues with certain file name characters and won't process those files (it won't even "load" them, it just plain ignores them). Without you having verified, or at least said, this problem is greater than this one rogue file, this is what I think of.

 

[jasn]

A couple of other things...

Did you use Album or Song analysis? If Album then you won;t necessarily have a consistent volume between songs. It shouldn't be terribly different though.

After you ran the analysis did you actually commit the changes? Perhaps you just analyzed the tracks. You can reload some of the tracks and see if the gain changes were committed.

 

[Boogie]

Boogie, first, since you read the guides I assume you have soundcheck off. So, my next thought is to ask what interface are you using? The mp3GainGui has issues with certain file name characters and won't process those files (it won't even "load" them, it just plain ignores them). Without you having verified, or at least said, this problem is greater than this one rogue file, this is what I think of.

Hi Code Monkey,
No, Soundcheck is off and yes, I did use the MP3 Gain GUI.

I wish it was just the odd track but it isn't unfortunately. In an extended listening session in the car I would say every third track was noticebly out.
I'm about to wipe the iPod and start again and I thought I'd better check if there is a better way of doing this or whether I'd omitted something glaringly obvious.

 

[Code Monkey]

In an extended listening session in the car I would say every third track was noticebly out.

You say, "car", was all your listening done in the car? And do you have a hookup through the dock connector or the headphone jack? Many car units that hook directly through the dock connector do NOT read or use replaygain tags, which is the basis of mp3gain and most other solutions.

 

[Boogie]

You say, "car", was all your listening done in the car? And do you have a hookup through the dock connector or the headphone jack? Many car units that hook directly through the dock connector do NOT read or use replaygain tags, which is the basis of mp3gain and most other solutions.

Sorry, I wasn't clear. I first noticed it in the car (connection via a Tranceiver in the base/power socket) but also noticed it when using headphones later.

 

[dcumpian]

I've seen MP3Gain ignore tracks that are being accessed by other applications like AV and WMP.

Dan

 

[Code Monkey]

One last pedantic stab: Do you have a mix of AAC & mp3, or even Apple lossless, and if you do have AAC & mp3, did you use the AACgain.exe renamed to mp3gain.exe?

 

[Boogie]

One last pedantic stab: Do you have a mix of AAC & mp3, or even Apple lossless, and if you do have AAC & mp3, did you use the AACgain.exe renamed to mp3gain.exe?

Good point Code Monkey. Being a simple man, all the files are mp3. There, unfortunately, isn't a mix of media or compessions.

Anyway, I'll give it another go and see where we are afterwards.

 

[Boogie]

A couple of other things...

Did you use Album or Song analysis? If Album then you won;t necessarily have a consistent volume between songs. It shouldn't be terribly different though.

After you ran the analysis did you actually commit the changes? Perhaps you just analyzed the tracks. You can reload some of the tracks and see if the gain changes were committed.

Sorry, Jasn,
I didn't see this earlier.

I used Song and made sure that I didn't just run the analyser.

 

[Boogie]

Hmm, a little less variation now.
One thought crossed my mind. If I've converted a vinyl track to mp3. This was notceably quieter than cd tracks. Normally, vinyl tracks have been much louder than cd tracks. Would converting to mp3 mean that it doesn't respond to MP3 Gain?

 

[Code Monkey]

Hmm, a little less variation now.
One thought crossed my mind. If I've converted a vinyl track to mp3. This was notceably quieter than cd tracks. Normally, vinyl tracks have been much louder than cd tracks. Would converting to mp3 mean that it doesn't respond to MP3 Gain?

No. All mp3gain, iVolume, iGain, etc. do is they run the audio file's wave form through a analysis algorithm that gives an intended (arbitrary) average decibel level for that audio file if played back at full volume (I say arbitrary because the actual decibel level of the file when played back is also determined by impedance of the speakers, gain added/decreased by the amp, etc.). Then it writes a little tag that tells any compliant device to increase or decrease the volume by X decibels so they are all within +/- 1.5 dB of one another.

So long as the analyzing program can decode the audio format and the format container supports replay gain tags (and most do nowadays), it works. The initial source and the initial set volume are irrelevant in as far as the process working.

Now, just for some food for thought, even when working perfectly, replay gain algorithms still are not perfect. The reason is that they operate on a model of what they call psychoacoustics, or "how loud the track sounds" to a human ear. However, because that largely comes down to looking at some interrelated formula of average and modal volume levels, there are quirks based on the type of music. The one I notice the most is that "true" heavy metal with the constant drums and guitar registers as sounding louder than a pop or more normal rock song with the much more pronounced rise and fall of sounds. The result is that my heavy metal tracks often wind up sounding a bit quieter than a calm classical piece - not horribly so, but enough that I will sometimes need to tweak the volume.

Eventually my plan is to go with iVolume. It uses replay gain algorithms, but writes the info to the the SoundCheck field that iTunes uses exclusively. The reason I eventually want to migrate is that it's possible to set up sets of music for different treatments and then have it keep them updated automatically, so the heavy metal can be set to 1.5 dB higher than other genres, which will actually make them all sound the same. But to use iVolume, I've got to complete my entire archive project, and that's probably going to be at least another 3 years at the pace I'm going. An organized music collection is a huge, ongoing project to say the least

 

[Boogie]

I know what you mean CM, I've got something like 2,000 albums and a pile of singles/tapes to archive.

In the meantime I'm going to give iVolume a try. I've pent too much time with mp3 Gain. The differences in volume are dramatic, to the point that one track can be so quiet as to be almost unlistenable while the next is at an ideal volume.

 

[AFMG]

I know it's been a while since anyone checked this, but: Is there a newer alternative to MP3Gain under Windows? In fact, Is there one that handles aac files too?

 

[kornchild2002]

Someone came out with a patched version of MP3Gain that would work with AAC files called AACGain. You download the exe and replace it with the original MP3Gain located wherever you installed it. I should note that the algorithm used by MP3Gain (and AACGain) has been outdated for a while. You can actually get better results with ReplayGain and Soundcheck. You can enable Soundcheck in iTunes, let it scan your library (this may take a while), then open up a program that will scan the files using ReplayGain, and then convert the ReplayGain values to Soundcheck results so that iTunes (and iDevices) can read them.

There is a free but complex way of doing that under Windows using foobar2000. It works with many different file formats including mp3 and AAC. Converting the ReplayGain results to Soundcheck numbers is the tricky part. There is also a paid program for Windows called iVolume. It actually works with iTunes. It will read your iTunes library, selections in iTunes, etc. iVolume uses the ReplayGain algorithm and converts the results to Soundcheck numbers for you. It also happens to be extremely fast taking only a few hours to scan and fix my entire 500GB+ library. iVolume will even adjust Apple Lossless files in your library. The program set me back by $30 for Mac OS X and I think the Windows version costs the same.

Sure, it isn't free but it is a much more automated process that produces better results without requiring you to jump through a bunch of different hoops.. The cost is well worth it if you like to get programs up and running and then walk away letting them do all of the work without you having to constantly point out files you want to adjust (just like you have to do in MP3Gain/AACGain).

 

[AFMG]

As I said in the other thread, seems like spending some money might worth it. Thanks.