Print

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

[FenceFurniture]

I'm just starting to use Analyze Audio, and thought it would be interesting to compare two versions of the same album. One is 24bit 96000Hz which must be a vinyl rip - it is the muddiest sound I have ever heard (which is why I selected this album for the comparison). It is the top one in the image capture.

The other version of this album is a standard 16-44 CD, and it is the bottom one.

Whilst I don't really understand the differences in the figures, I can see that there are vast differences in volume, dynamic range and peak levels, and perhaps a kind soul might roughly explain what they mean.

However, my main question is:
How can there be such a wild variation in the BPM? Tracks 1 and 4 show the biggest variations: 127 and 84 for T1, and 75 and 100 for T4. I may not understand tech stuff too well, but Beats Per Minute is pretty fundamental  (I have listened to them both and they are the same song and tempo    )

Is there perhaps a clue in how the software determines the BPM or is that just taken from the tags? What I'm driving at there is could the awful muddiness of the top one somehow interfere with the software's interpretation of the BPM?

[DrKNo]

BPM detection is not an exact science. You might notice that 84 is roughly two thirds of 127 and 75 is three fourths of a hundred. That indicates that the algorithm has the tempo right in principle, but is missing beats during detection. That is an error that is quite common in BPM analysis, I see that happening occasionally in all the professional software I use as well. If the signal from one source is muddled, then beat detection becomes more difficult, explaining the possible error. The errors also become more pronounced if music doesn't use the prominent 4/4 timing of most western music.

[FenceFurniture]

Ok, thanks - kinda as I suspected. It'll be interesting to see how it goes because I have a lot of Jazz and Prog Rock that is not just "not 4/4" it's often a few different time signatures within the same song. Even Sting uses a lot of 7/4. Mind you BPM is not something I pay much attention to....until I see results like those!