Mp3 Converter Comparison

May 8th, 2011

This video isn’t so much a tutorial as it is a comparison between two current, popular mp3 converters. Essentially I just did a simple phase comparison between the LAME and iTunes Mp3 converters. The only thing that can be proved here is that there is a difference. I’m not saying that one converter is better than the other. I’m just proving that there is a difference.

I thought it was interesting so I’m sharing the knowledge with you. The fact that they add different amounts of level to the final exported file was very odd I thought. LAME ends up peaking at 0dBfs so I’m assuming it has an inbuilt limiter or compressor in it, either that or it was just a coincidence that I set my limiter to -0.3dB on the mastered .wav and LAME added exactly 0.3dB to the exported mp3.

Anyway, make your own decision as to which one sounds better. Just be aware that they are definitely different. If you can’t tell the difference, then I wouldn’t worry too much. I’m sure the difference is quite minimal, I can’t hear it. I’m convinced any differences I’m hearing are influenced by the knowledge that there is a difference between them.

What do you think? I’m interested to hear other peoples thoughts on this subject.

Cheers,
Bill.

http://soundcloud.com/Urple_Eeple/ Urple Eeple

Hey Bill! I think the speed of your LAME encoding has to do with the converter program you’re using. I use dBpoweramp music converter and it’s just as fast as iTunes when I select LAME for my encoding.

Also, I’m not sure that mp3 converters all add a “little bit extra” volume. This may be the case with iTunes, but from your results it seems to indicate to me that what your LAME converter did was normalize the track, by setting the highest peaks at 0 DB. This may be a standard feature of LAME, although I’m not sure.

On a different note, have you tried any other converters? My music converter also offers Helix, and Fraunhofer IIS. Can you speak from experience on using these types of mp3 converters as well?

http://www.facebook.com/mrbillstunes Bill Day

Hey Urple Eeple,

Every mastering engineer I’ve talked to has said the same thing about mp3 players adding extra volume (including Bob Katz), they all say to limit tracks at -0.3 for that reason. It does it with the LAME encoder and the iTunes encoder. I’m not 100% on other encoders, I haven’t used a lot of them.

Regarding normalisation, I definitely didn’t do that. The extra volume is added from the conversion process. As for using different encoders, I haven’t tried too many, no. I usually just use the iTunes or LAME ones. Personally, I try not to encode to mp3 if I can help it, but until the internet gets a little faster I’m kind of forced to sometimes.

As for comparing the WAV to the mp3, that’s actually a much better idea, and a few people have brought it up to me since I did this video, and I cant believe I didn’t think of it. All you would have to do is line the wav up with both mp3 converter files, put it out of phase, render it, then compare the noise of both files. Then it’d be the true noise difference I guess. I think it would be quite similar in both files though. The point of the video was purely just to show people that mp3 converters ARE different, they’re not all doing the exact same thing.

Cheers,
Bill.

http://soundcloud.com/Urple_Eeple/ Urple Eeple

Thanks for responding Bill. I definitely agree with you that diff audio codecs create diff results.

As for everything else we talked about, I think I wasn’t as clear in my comment as I could have been and hence some misunderstanding I noticed between what I was intending to say. I wasn’t trying to say that mp3 converters don’t add extra volume. What I was trying to say was that it seemed to me as though the iTunes converter added a set (percentage or db) amount of volume, whereas it seemed to me that the LAME encoder somehow may have normalization built in.

It just seemed strange to me that the highest peak was right at 0 DB for the LAME converted file, and seemed to imply to me that perhaps the LAME encoder automatically normalizes files as it encodes them to mp3.

As for checking the quality of different codecs, I just ran across Sonnox Pro-Codec plugin. Have you checked it out at all? It lets you A/B diff audio codecs on a set in real time. I haven’t tried it out personally to be able to say if its worthwhile or not, but I’ll shoot you an email if I do and let ya know if I found it useful.

http://soundcloud.com/Urple_Eeple/ Urple Eeple

Oh! One other thing I found from a Bob Katz video. You can take the WAV you’re comparing, and subtract its waveform from the mp3, leaving behind essentially the distortion that the mp3 itself creates, and then you can a/b the two separate subtracted pieces. This might give a more accurate comparison of which mp3 converter creates less artifacts.

Username

Password

Remember me

Forgot Password?
Join Us

Mp3 Converter Comparison

Password Reset