Spectrum of Musical Genres

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Administrator
Joined 2004
Paid Member
You're welcome!
I ran peak values on the EDM tracks for you, as those would have the highest bass peaks. See below.
 

Attachments

  • EDM-peaks.png
    EDM-peaks.png
    24.2 KB · Views: 245
Administrator
Joined 2004
Paid Member
Yes, surprising, really. I did see one track that had a low sweep from about 35 to 70Hz, tho it actually rolled off at the bottom. Another track (not included) had a small peak circa 13Hz. Didn't see anything else nearly that low. Basically bass is 10dB above midrage and mids are 10dB above the tweeter range for peaks.

Anyhow, that's the way it's recorded. How you play it back is another matter. :)
 
Administrator
Joined 2004
Paid Member
I went over this paper a few times, it's good work. I tried to compare my findings to theirs, and had to re-run the analysis as close to the AES setting as I could. FFT of 32K bins, 50% overlap, Hamming window. My graphs are similar, tho not identical to those in the AES paper. Same trends, similar low end peaks and roll-offs. But their spectrum does not drop as fast as mine. Mine also show a bit more bass.

The study using longer and different averaging than mine, so that might account for the difference in treble roll off. Certainly their plots are smoother than mine.

Thanks again for posting. :up:
 
Now that this thread has basically run its course (about a year ago at this point), I thought that rather than creating a new thread and copying some of the postings from here to a new thread, that I might add some perspectives from the standpoint of demastering music tracks to the discussion, thus reusing the data and perspectives here in a way that becomes even more useful, especially to those that have taken the time to dial-in their setups (flattening frequency response, extending and smoothing bass response, time-aligning the drivers within loudspeakers and within the entire loudspeaker array) and wish to get significantly more out of their recorded music tracks. These people might already be using parametric/graphical equalizers or "Cello" palette type of equalizers each time they play their favorite recordings to correct for mastering EQ and other issues in the as-is tracks.

If there are objections, I can create another thread and go from there. If there are no objections, I'd can proceed by quoting some posts from last year and lend some perspectives on the basis of demastering approximately 14000 music tracks (about 5000 elapsed hours). The objective is to lend a little more understanding of some of the processes used to remaster older albums to help them sound a much better and more balanced (instead of just making them sound louder and less dynamic).

Pano: it's your call. I really won't be hurt if you want me to start a separate thread and link back. I would guess this will be on the order of perhaps 10-15 postings on my part over time (not counting those that might join the discussion).

Chris
 
Administrator
Joined 2004
Paid Member
Hey Chris. I don't mind if you put that here, but it seems like a topic worthy of its own thread. If you need info from this thread, it's easy to quote it into the new thread. Having two tabs of diyAudio open helps when doing that. :)
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.