N-track recording level ? Pt. II

[cellardweller]

Before mixdown, where should levels be for optimizing volume levels, without clipping or distortion?

I've always suffered from the kneejerk reaction of cutting volume when seeing yellow and/or red on the level indicator, but after mixdown, its so damned low on volume I have had to wonder of late.

So what is the common consensus of usable volume?
Peeking only in the yellow? No Red?

Yea, I know, this is probably just another one of many "Thank You Mr. Obvious" questions that I seem to come up with on a regular basis...

 

[Cazzbar]

We seem to be curious about the same things cellar. I tend to record as high as possible without clipping (minus about 5% for good measure) and the same when mixing down.

From reading this forum the process appears to be to normalise certain tracks (increase volume so the loudest part is at the maximum) and / or use compression (to boost the lower volumes but ensuring the higher volumes don't go off the scale)

Because of this boosting I'm assuming recording a strong signal in the first place is the way to go.

 

[cellardweller]

The good ol' "it can be fixed in the mix" lie, rears its' ugly head yet again.
It seems no amount of mastering, compressing, normalizing, can make up for a poor initial recording...

I'm still trying to figure out the intricacies of compressors. Some of the compressors I've had bad luck with it jacking up the equalization (oh yeah, I'm still mastering the intricacies of EQ's also).

Does the N-track compressor have a normalizer on it as well?

Do you use any other .wav editors?

 

[Cazzbar]

Yeah, a good initial recording is important I feel. My first few mixes only had slight track volume changes, no compression or anything and sounded fine. Now I have all the knobs at my disposal I feel I'm tweaking all the time... Hmmm

ntrack has a normalise option off the menu by default I think, not part of any effect.

I'm not using any wav editing software (yet). morindae says he uses that to 'warm' vocals... something else I need to learn... or is that just more knobs to twiddle with

 

[cellardweller]

Yes, more knobs it would seem... As if I am not inundated by knobs already...

 

[fritzmusic]

As a general rule of thumb, I try to "peak-out" at -6db (in the yellow) if I'm not running compressor or preamp. This is when I'm recording at 16bit/44.1khz. If you can tame the signal via compresor or preamp, try to get it as close to 0db as you can without one clip. If you're recording at 24bit, you can get away with less volume and bring the levels up with less signal to noise ratio. But as a genral rule of thumb in my opinion, -6db no matter what bitrate you're recording at won't do you wrong.

 

[Kim L. Ground]

What about the 'auto level record' option ?

It seems like that might be useful for making an initial recording. I have had cassettte recorders where this worked just fine (and at least one where it didn't). BUT I can't seem to activate it in N-track for recording the first track -- that selection is greyed out until you already have a track showing on the timeline display.
What good is that ?? Is there something I'm missing here ??
If I have to set levels manually for live material, I tend to fiddle with the knobs too much and get noticable changes in volume level during the recording if I am not careful to leave the knobs alone.

 

[Teej]

Welcome to the world of recording engineers.

Getting a usuable level without clipping is a challenge for everyone, Kim. That's why its being suggested you shoot for -6. The strategy I use is:

- Ask the performer to play/sing as loud as they ever will during the take.
- Adjust the gain at the mixer so the signal never clips (shoot for -6).
- Monitor N's Recording VU Meter. If your system has been set up properly, N should be reading the same level as your mixer. Otherwise, adjust the mixer's output so N reads about -6.

This is digital; you never want a signal to go above zero. Better to lose a bit in terms of of signal-to-noise than clip. Ugh..

If your finished rendering isn't loud enuf, use Normalization to bring the very highest peak up to -1. If the song STILL isn't loud enuf, then you have to decide if you want to mess with the dynamic range of the song. In other words, the peaks are too high. Compression or manually drawing volume evolutions to reduce the peaks would be your next step.

It's a trade-off between volume and tone. Compression DOES effect the tonal qualities of a track.


Hope this helps.


tj

 

[Kim L. Ground]

so about the auto level

You feel it is not any use at all then ? Or is that what you mean by normalization ? And it seems to me that auto level is just a crude form of compression, where you don;t get to control the curve ... But you can't use compression on the incoming signal (not that I think it would be a good idea, except maybe to save you from clipping), evn though I seem to always end up using the 'soft knee' to make live recordings seem 'fuller' and sounding more like a commercial product.

 

[Teej]

You mean compression? Not any use? Sure, its of use; especially live. Drums, guitar, bass... all these things can benefit from compression during recording too. But N can't do that; you'd have to have an outboard unit.

I'm just saying that in a perfect world I'd rather NOT compress as i'm laying tracks. I'd must rather compress after the fact. You have much more control that way. If you compress during recording, you're stuck with it.

And yes, compression changes the tonality of the track. Sometimes good, sometimes bad.

For me, I'd rather record at a safe level (clipping sounds yuk!), do everything i can to reduce noise (background, PC fans, etc. so the signal-to-noise ration is better), and do my compression work after the fact.


Hope this helps.


tj