Hello. I was wondering what good tools were available for testing and comparing mp3 files of the same track but encoded differently. Are there any tools that indicate which version is of higher quality even if they are both perceptually indistinguishable (well, at least to me)? If so, how can you determine the higher quality file? Thanks.

EncSpot and Mr QuestionMan

You can use ABC-HR which you can download at www.rarewares.org. These programs will perform ABX tests on the files and they will determine whether you can hear a difference between those files and the original audio file. There is no program that will tell you what MP3 is better, the only programs that will do something close to that will test if you can hear a difference between an MP3 and an original audio file.

Edit: or you could use the programs suggested above to estimate the quality of the file based on the encoder and bit rate. Forgot about those.

I've seen discussions of mp3 quality that referred to a (I may be using the wrong word here) waveform or graphical representation of the sound file (similar to the visualization you'd see in winamp) that showed one encoded version of the same source file with all of the peaks chopped off in a straight line, while another had taller, jagged peaks, that indicated one file was of higher quality than the other. What kind of test is that, and is it not what I'm looking for? Thanks.

Are you referring to discussions about mastering quality, limiting, clipping, etc? If so, the waveforms are in screenshots usually from audio editors like Audition or Audacity. So it's not really a test. It's usually just people proving a point or comparing different masters of specific recordings, and it's not at all specific to MP3 or any other format.

The visual representation they were refering to, was a spectrogram, which by the way is pretty useless when it comes to judge encoders quality. Spectrograms consist of time axis, frequency axis, and colour notation of level. When on spectrogram you see the peaks chopped off in a straight line, it means that encoder used a lowpass filter to filter out all the inaudible high frequencies which would be impossible to encode properly at given bitrate - this is absolutely a good thing. They probably think that the one with taller, jagged peaks is of higher quality, which is wrong. Encoding is a really much more complicated process than just lowpassing which you can see on spectrograms, and except your ears, there's no tool that will tell you which is better.

Thank you all for your thoughtful and illuminating replies. I consider my question thoroughly answered.