To Stretch or Not To Stretch - Time-Stretching vs Slice Editing
Many years ago, before programs like Pro Tools added Elastic Audio or REAPER added Stretch Markers, people like me had to edit drums the “slice editing” way. Which meant cutting the audio (splitting) on the transient points and then quantizing all of those little pieces of audio to the grid.
Pro Tools and REAPER made this process fairly easy by providing transient detection tools. Beat Detective for Pro Tools and Dynamic Splitting in REAPER. I’m sure all DAWs have something similar but there were many engineers that still preferred to cut up everything by hand. Manually. Ugh. Very time-consuming.
Programs like Fruity Loops (FL Studio now) and Live did things a bit differently. They time-stretched everything to fit the tempo of the track you were working with. Incredibly fast but also had many audio artifacts.
Many engineers didn’t like the way they sounded so DAWs created time-stretching options that used transient points to stretch from. Making it sound much better but it still wasn’t and still isn’t perfect.
At this point, the best sounding option is still to cut or use slice-editing to quantize our audio. It sounds near perfect because the only thing changing are the fades from one slice to the next. And you can tweak those to taste. But there is one problem with this option.
As soon as you start slice editing or using tools like beat detective, it makes it difficult to just do it in spots. The sections that are not quantized feel jerky jumping from the fixed sections to the unfixed ones. So you wind up fixing more than you had planned. Eventually, you just fix everything as you venture down the rabbit hole.
For example, if you’re editing drums and the drummer plays a ghost note on an upbeat, you might NOT want to fix it. He played it well (swing feel) and you think you only need to fix the downbeats.
That gets tricky though. If you don’t edit the ghost hit too, where does it get placed? Do you trim it out from the beat before or the beat after? Because it will be different and neither option will be what the drummer actually played. So you lose his feel if you edit this in any way.
Now, if you use time-stretching with transient markers that you can drag around, you CAN preserve that ghost hit by just defining the beat before and the beat after and let the ghost note sit or stretch between the two defined hits. So it will preserve the drummers performance more.
The best part about this method is that you don’t have to quantize everything. You can just quantize the top of each bar. Or the kick and snares on the downbeats but leave everything in between (hi hats) alone. It will stretch between the two defined parts and sound great. No more quantizing everything.
The downside to this method is that many people hear the time-stretching algorithm working. It is audible in almost every situation. No matter which algorithm you choose. Especially if you hit solo.
What you need to do is decide if it’s worth it. What I did many years ago is edit a drum track both ways and keep both playlists in the project. When it came time to mix it, I mixed both versions.
I then listened to both mixes for the next few days listening to see if I could hear an audible difference in the end product. Because really. Who cares about what things sound like in solo?
For me, I couldn’t hear a difference. In fact, I found myself choosing the stretched version more often than the cut up one most times. So after that point, I stopped concerning myself with the difference. I enjoyed the benefit of only editing the drum hits that I wanted to edit and ignored the artifacts. I decided that for me, they didn’t matter.
Of course, you have to decide for yourself. Try it both ways and see what works best for you.
You can learn more by watching my videos at www.groove3.com
I hope this message finds you well. Kenny Gioia