|
|
Post by Doug on Oct 23, 2014 19:46:31 GMT -5
I know a music thread OMG  How much compression for vocals. My recording program comes with some presets for that, the obvious one is reduce loud parts. Tried that and it cuts stuff in half, feels like it's stealing the dynamics. So I changed the settings so that it's at 75% and created a new preset and that's what I'm using if the differences feel too much and some times not at all. I heard use just enough compression don't use too much or it makes the vocals flat. OK all you recording guys, how much is just enough? I know it varies but y'all hopefully should be able to tell me if I"m way off. Should I adjust to like 95% of the raw or is the 50% right? |
|
|
|
Post by Lonnie on Oct 24, 2014 0:19:00 GMT -5
I don't think there's any one way to do it. It depends on the dynamics of the original vocal. If your recorded level is pretty consistent, not much compression is required. If you're all over the map dynamically, compression alone won't fix it. In the old days when we had real boards, we'd ride the vocal track and bring things up and down as necessary. I'll still write a vocal volume map in pro tools, and then apply compression as necessary to smooth things a bit without losing the dynamics. It's not "one size fits all. "
|
|
|
|
Post by mnhermit on Oct 24, 2014 6:53:16 GMT -5
I guess that answers my question which was 'why compress the vocal track?'  |
|
|
|
Post by Doug on Oct 24, 2014 7:07:15 GMT -5
But how much is a little? I just want to trim the peaks a little I don't want them to go away.
|
|
|
|
Post by aquaduct on Oct 24, 2014 7:21:02 GMT -5
I don't use compression when recording at all. I opt for dynamics over volume and with my wife's vocals there's never a need for it. Set the levels right and there's no problem.
Occasionally I'll use it live when there are multiple singers, and then only as a limiter to catch sudden peaks since you can never be quite sure what's coming.
Compression's a tool like a sledge hammer. If you're breaking up concrete it's great. But using it for carpentry will usually just destroy the piece you're working on.
|
|
|
|
Post by aquaduct on Oct 24, 2014 7:22:34 GMT -5
But how much is a little? I just want to trim the peaks a little I don't want them to go away. Why do you want to trim the peaks? |
|
|
|
Post by Marshall on Oct 24, 2014 8:14:00 GMT -5
The only finagling I do these days are on videos. I'll strip out the audio to a wav file and bring it up in Audacity. I will do a visual look at the tracks and reduce volume on the big peaks. When I listen back it seems to even the audio out nicely without really changing the dynamics. It's almost un-noticeable. I'd say it even sounds more natural.
Then I'll do a little EQ to add some fullness that video mics (iPhone) miss out on.
|
|
|
|
Post by dradtke on Oct 24, 2014 8:18:48 GMT -5
How is compression different from normalization?
|
|
|
|
Post by PaulKay on Oct 24, 2014 8:28:09 GMT -5
I pretty much stick to 14.7 pounds per square inch.
|
|
|
|
Post by Doug on Oct 24, 2014 8:31:09 GMT -5
The only finagling I do these days are on videos. I'll strip out the audio to a wav file and bring it up in Audacity. I will do a visual look at the tracks and reduce volume on the big peaks. When I listen back it seems to even the audio out nicely without really changing the dynamics. It's almost un-noticeable. I'd say it even sounds more natural. Then I'll do a little EQ to add some fullness that video mics (iPhone) miss out on. Marshall, you use volume to reduce the volume on the big peaks rather than compression? |
|
|
|
Post by Marshall on Oct 24, 2014 8:49:49 GMT -5
That's what I said. That way it just takes the big noises and brings them down more in line with everything else. it doesn't affect the whole file.
I'm not saying it's the right thing to do. But it does improve the listenability of a track without changing (coloring?) anything in the track. I find compression protocols change everything; especially if you don't know your way around the block. If you're an experienced hand, like Lonnie, I'm sure he knows how to get done what he wants without injuring the rest. But for a simple beginner trip through a file, some selective volume drops will help even out the audio and not damage the overall sound of the performance.
PS - I literally highlight (select) the peaks in the graphic file one at a time; go into Effect; Amplify and choose a negative number. Then I look at the reconfigured peaks to make sure it's safely reduced (undo is always available). Then I listen to the segment of the song with the lowered peak to make sure it's OK. Then I move to the next one.
I'd say I only do this on 3 to 10 places in a track. If the whole track is peaking near the limits, I'll probably reduce the whole track a little before finagling. I always "listen" to the edited portion before moving on to the next spot. Undo can be your friend.
|
|
|
|
Post by aquaduct on Oct 24, 2014 9:21:06 GMT -5
How is compression different from normalization? Normalization applies a constant amount of gain to increase the entire piece to a target peak or loudness level. Relative dynamics and noise floor are unchanged. Compression applies gain reduction to parts over a specific threshold effectively allowing softer passages to be louder. It essentially squashes the signal of the piece/track/whatever. Applied excessively it kills the musical dynamics of the music. The so-called "loudness wars" in modern music are the result of increasing application of compression in order to hopefully produce a track that's loud enough to stand out from competitors. It's also why much modern commercial music sucks. And compression can be accomplished manually with volume fader adjustments. Same thing. It's particularly easy with automated mixing. |
|
|
|
Post by Marshall on Oct 24, 2014 9:32:53 GMT -5
I love watching a great professional singer work a microphone in live performance. It's the ultimate form of "compression" by changing the distance to the mic for different singing dynamic volumes.
|
|
|
|
Post by RickW on Oct 24, 2014 9:56:32 GMT -5
Doug, there's no easy answer. You really have to listen to what you've done. Listen after recording, listen on the home stereo, listen in the car. If you are hearing something you don't like, you fix it. So, if there is a part where you're peaky, drowning something out, either a note or phrase, you might want to compress it. If an overall track has peaky points in spots, either vocal, guitar, etc., then try compression.
The debate over compression in the media is that producers try to squash everything down, then boost the overall gain, so you get almost no dynamics in it. You don't have to do that, you don't have to not do that. It all depends on what you like to hear. You can crush the lift out of a piece of music.
|
|
|
|
Post by Marshall on Oct 24, 2014 10:31:35 GMT -5
As Peter says, most (all) recorded music you hear over your car radio and home stereo has been compressed. Think; Barbara Streisand singing "People." The vocal dynamic range is from whisper to belting it out (as only Streisand can do). . . . , yet it all sounds the same intensity on your stereo.
|
|
|
|
Post by billhammond on Oct 24, 2014 10:33:32 GMT -5
As Peter says, most (all) recorded music you hear over your car radio and home stereo has been compressed. Think; Barbara Streisand singing "People." The vocal dynamic range is from whisper to belting it out (as only Streisand can do). . . . , yet it all sounds the same intensity on your stereo. Geez, here all these years I thought that song's title was "Peephole." |
|
|
|
Post by millring on Oct 24, 2014 10:39:37 GMT -5
they're the hungriest people in the world.
|
|
|
|
Post by Marshall on Oct 24, 2014 10:54:46 GMT -5
Geez, here all these years I thought that song's title was "Peephole." did you make that up? Or did you read it somewhere? It is fun how sometimes a singer changes sounds being sung so they can come out more listenable. "Peep-hole" would be a great example. |
|
