Hi guys, long time semi-lurker here. This is the first time I actually have a reason to start a thread.

As stated in the title, I started wondering, which process is more destructive: change of pitch or tempo? Is there any smart way to measure it?

For example, theoretically, if I have two samples and one is off by 30 BPM from my target and the other one is off by 6 semitones, but has correct tempo, which one should I use? Is there a general rule of thumb? Should I prioritize one thing over the other? What's your opinion on this?

I hope you will be able to help and dispel my doubts

And of course, I know there are different algorithms, etc. - please assume that we put two good ones against each other.

 

Changing the pitch for a song may affect some of the instruments, because not all instruments have the same range. Even a couple od semitonews can make things like violins and brass / woodwinds, lose some of the higher or lower notes.

 

Neither are more "destructive" or harmful.
Non-destructive editing is an edit that can be changed back.

Only rule of thumb: speeding up tempo is more transparent than slowing down, and tuning up is more transparent than tuning down.
@WillyA Talking about samples here.

 

I would say it mostly depends on the source material, what that source material is to be precise. To me, majority of vocals start to sound bad even after shifting them only by 1 semitone up or down. Percussion on the other hand usually sounds pretty decent no mater how much I pitch shift it. Some sampled instruments like brass and strings also can sound pretty good even if they're pitch shifted by like an octave.

Try out them both and see which one you like more and what works best for your project. Experiment with this, sometimes you might stumble on some interesting results.

 

Yeah, I shouldn't use the word "destructive", that was unfortunate. "More transparent" is definitely what I had in mind. Thanks for your input!

 

Initially, I'd say use the one with correct tempo, but only if you are pitching up that 6 semitones.
And if the other sample is 30 BPM too slow, then speed it up. But not down.

Look into what 'Formant' is.

Hard to imagine. Again, the lack of formant change in the instrument will be the same as with the vocal, but we are just more sensitive to how natural or not a vocal sounds.

 

Another important parameter when changing tempo is that accelerating it is completely different than slowing it,
2 Examples
Speed a snare drum sound, or a drum track in general, and it just seem to be more punchy. Only the cymbals may be noticeably too short, because their decay is fixed.
Speed a voice, or a sax, and suddenly, any vibrato/modulation seems weird and too fast.

Slow a drum, and you immediately feel the problem in the attack.
Slow a voice or a sax, if it's not much, it can still be good.

So it depends of the source.

 

As already mentioned above, the source matters a lot; what works for one source may not work well for another.
That makes your search for a universal 'which is better' approach very hard to achieve.

Maybe better to be familiar with different techniques for different scenarios.
For example... Oly mentioned a specific problem with slowing down drums and hearing the attack get washed out.
As pitch-shift/time-stretch packages get smarter they get better at coping with issues like that.
So, if your time-stretch facility is offering you options like 'percussion' then it's a fair bet they are trying to identify transients
so that they can minimise the effects on those transients.
(stretch the transients less than stretching other areas when slowing down, etc)

One DAW which claims to do this explicitly is MuTools M8
See how their time stretch works at http://www.mutools.com/info/M8/docs/mulab/timestretching.html (worth a quick read)
Extract: The Head and Tail sections are not stretched and allow to neatly preserve the transients.
Only the body section is effectively stretched.
So, if you're slowing a drum track down and ruining the attack, maybe see if MuTools M8 can do it better.
Sometimes it's worth temporarily using a different DAW just to get a specialised task done better (maybe)

 

Yeah, this is definitely a complex problem. I should have seen this coming. I thought that maybe more experienced people will have some easy answers, but it seems that the usual "it depends" applies to this one as well. I still get some useful info from your comments, though, so thank you all for that!

What actually got me thinking about this whole thing was cataloging the Black Octopus Sound vocal samples I downloaded from the sister site. Basically, I started wondering, which problem will be easier to fix and how should I catalogue these vocal samples (by BPM or by key). Trivial problem, I know, but in my defense, I was composing only instrumental songs or using libraries like Audio Imperia Hangar IV so far, so I don't have much experience with fixing samples manually or working with "real" vocals

 

Good read, thanks! I probably wouldn't have thought about this myself.

Fortunately, I think I will be able to achieve the same effect with Reaper stretch markers, so no need to change DAWs

 

Tempo (time-stretch): Changes the sound to play at faster or slower speed than original, without affecting the sound pitch. Pitch (key): Changes the sound pitch or key, without affecting the sound tempo or speed. Playback Rate: Changes both the sound tempo and pitch, as if an LP disc was played at wrong RPM rate.