Print

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

[jack wallstreet]

I sometimes get duplicate songs from different records but same artist where the song version is identical and has he same exact time, soundsthe same and same intensity, BUT the audio analyzer gives wildly different BPM - e.g. 65 on one song and 100 on the other.  This doesnt' happen often (most duplicate songs have bpm either identical or within a few beats.  Anyone else see this or have a suggestion.

[Alex B]

BPM works by sending a little snippet of the song to JimH's office.  He taps his foot as he listens, and then sends the correct BPM back to your computer.

During off hours, a random number generator takes Jim's place.

Hopefully this explains any inaccurate results you may see.

http://yabb.jriver.com/interact/index.php?topic=33915.0
http://yabb.jriver.com/interact/index.php?topic=37249.0

[jack wallstreet]

Aha!!!  Thank you.  I will run the audio analyze only during normal work hours. 

[Alex B]

Seriously speaking, are the files really identical or are they slightly different? If they are not exactly the same MC may detect the beat differently.

When you run the analyzer on the same file several times MC should always produce the same BMP value. If that does not happen then it may have a bug.

In any case you can always edit the BPM values if your opinion differs.

[jack wallstreet]

As far as I can tell, these files are substantially identical, but NOT to the extent of being identical bit by bit copies.  I cannot tell the difference by listening.  They tend to be the identical song/version/artist put on different records (e.g., a "best of" and the one from the original record/cd.  These songs can sometimes be slightly different in length (e.g. a second or so) due to quiet time at beginning or end or can be digitally enhanced.  I noticed the issue because I look for duplicates to eliminate them and was not expecting exact duplicates (as far as I can tell) when the BPMs were so radically different.  A forced recalc on the songs yields the same results (within a BPM or two).  I have just done a purge of duplicates and found maybe 3 (out of 75) dups that did this.

[scthom]

I've sometimes had the analyzer tell me the BPM was 160 when it was actually 80 (or vice versa), but it's usually pretty good.  I can only think of one time where it came up with a BPM that wasn't either double or half of what I thought the actual BPM was.  In my experience, re-analyzing doesn't help (I typically get the same result), so I just change the info by hand.  When I have seen the analyzer do that, it's because the song has a very strong back beat or a very weak main rhythm.

[jack wallstreet]

Thanks,  The numbers may have been double or 1/2.  I am not sure, but I do remember them being far off.  You're explanation sounds good to me.  thanks.