Milovana.com

Sometimes, you need a segment of the music for BMG to show the correct BPM.

Other times, you need to test the entire song.

A completed song: https://mega.nz/file/fz4EQTiB # zhg5aeZL0R7pln7OwIDtYgv7RazHrrOULthadXA9_I8

Some screenshots of what I did: https://imgur.com/a/JX1Lc28

The Beatmeter Generator template:
With "in This Moment - As Above So Below," I needed to BPM detect the entire song, and then round up. I needed multiple rounds of detect, listen, move, listen, to find the correct BPM. I found this one quickly by knowing that it would be 4/4, so I could time the signature track in BMG to start at 1 and repeat at 5.

Usually the important thing is for you to choose the correct spot to start your BPM track for detection. So, when rounding BMG's detect BPM up or down, you also must usually move the BPM track a bit to match the song to the resulting signature (usually 4/4) track.

Then, it is important where you start the (usually 4/4, 1, 2, 3, 4 1, 2, 3,4) time signature. That's why I use default metronome in BMG for my BPM tracks, and custom beat for the time signature track. I listen for the 1, 2, 3, 4, and find from that what is wrong. Sometimes it takes much time to move those two tracks to find what is wrong. My first read of "As Above So Below" was 1.5 times slower than its actual BPM, but it only took about 5 minutes to find the real BPM, because I've done it before.

When the song (or a part of a song) is not 4/4 (1, 2, 3, 4, repeat), that will be harder to notice, but will be found when you have correct BPM and signature for other parts or other songs. It will require your practice to know that either you are wrong, the software is wrong, or the musician was playing clever.

Usually, musicians play in 4/4, and when they don't, they use the same signature that isn't 4/4.

There are times when BMG doesn't get the BPM quite right, and it can be tricky if it's a song where it changes, but in those cases I've had luck with pulling a song into VirtualDJ and have it analyze the BPM. Usually that's bang on, unless there's something weird with the song. :)

Yeah, a near-match on BPM will fit where you tested but will drift off beat as you play it ahead.

I've seen some songs that do have fractions in their BPM, like 131.9 might be the real BPM on a song that has a published BPM of 132.

After the recent complaint thread about male vocals and dad rock in Cock Hero, I beatmetered this playlist:
Puscifer:
Rev 22_20 (4_20 Mix)
vagina_mine
queen_b
balls to the wall (silent servant el guapo mix)
balls to the wall (pillow fight mix)
breathe (drumcell rework)
Rev 22_20 (4_20 Mix)
Momma Sed (Tandemonium Mix)

Here is the BMG, and rendered music and bpm tracks: https://mega.nz/file/vroA2JoT # PA594RTt_32RTEayWbaF--ET4n44gtxlKkYBc436a18

The best way to transition between songs is to sequence them in order first, then find their BPM and signature tracks in order. Change the position of the next song to match the previous song, so that the next song starts its first beat where the previous song would start a first beat.



I also changed the torrent link in OP to mega.

If you use DJ-ing software, you can also match BPMs without noticeable pitch change, or even force pitch changes to other keys for harmonic mixing and such, so that's also an option. :) It's what I do with VirtualDJ.

VirtualDJ defaults to 4/4 measure and forces a BPM that will fit. That is helpful for when what you want is a 4/4 timing to mix against, which may make VirtualDJ a good tool for Cock Hero.

It reads "Puscifer Rev 22:20 420 mix" as 154.67 BPM, which is a fitting 4/4, but if you search you'll see that ChordU publishes 115.95 BPM, which is the reference I used to find that the song better fits 116BPM and /6 measure.

What seems most important to me is that the bar length (time between repetitions) is correct. That's why I settled (by ear) on /6 for most of Puscifer. I still get some fraction (decimal) BPM that I just accept because they may be explained by the musicians or engineers missing their timing.

Ah ok, I see what you mean. I'm not sure if I'd ever use anything other than 4/4 for a CH. Maybe a 3/4? CH Waltz, anyone? :D Or do one to Dave Brubeck's Take 5 in 5/4?

So long as we (the beatmeter author) can count beats to find a rhythm that works for metering. I know that I have ignored odd signatures in some of my meters.

Experimental musicians often publish this detail.

Edit: A more common example, to show you what I mean by knowing I've ignored it before:

6. Britney Spears - Womanizer

Another typical 4/4 tune right? Wrong. This lead cut from Spears' Circus album is actually far more interesting than that.

Although you could play it at 4/4 with a triplet feel, the track was written in 12/8 time at a speedy 139bpm.

Quite a number of other pop hits pull off this kind of musical sleight-of-hand. including James Brown's Mans World and Alicia Keys' Falling, although these are both far slower than Britney's breakneck banger.

https://www.musicradar.com/news/tech/10 ... ime-630992

Sometimes I can hear in my BMG metronome tracks that I've only found the correct (or good enough) count of beats-per-bar, and I'll leave it at that.

Also, I looked into the thing about VirtualDJ and found that it's technically able to do only 4/4 on the BPM analysis.

https://www.virtualdj.com/forums/218423 ... ature.html
https://www.virtualdj.com/forums/94752/ ... tures.html

So using it for that is basically doing what the Music Radar article was talking about with Womanizer, which only works because both 12/8 and 4/4 are even.

When I can't synchronize two tracks by ear in Resolve, I use BMG and web search reference for song BPM to get it done. In this case, the cover song is the same BPM and measure as the original.

Fairlight view of the audio waveform can show how the audio tracks compare.



Mute the BMG tracks, and move the audio until it matches.

BMG tracks will nearly match, if they are done correctly, but they will always be slightly off from each other.

Pan one track to left ear, and the other to right ear. Listen for a drum hit, then align them.