i thought it is best to record around -18dbfs

  • Thread starter Thread starter djclueveli
  • Start date Start date
snow lizard said:
At that point, it might be useful to consider that if you're only recording something like an acoustic guitar...
Click to expand...
Something like acoustic guitar is transient heavy, so tracking it with a peak of -6dbfs or so would still have your rms around -20. See how well this works?

If you are recording in 24 bit, you really don't have to worry about the resolution/precision with any sane recording level. Having peaks at -48dbfs will give you the same resolution/precision as the CD that what you are recording will eventually find itself on, so having peaks at -12dbfs on stuff that doesn't have any transients is not going to hurt anything.
 
Farview said:
Something like acoustic guitar is transient heavy, so tracking it with a peak of -6dbfs or so would still have your rms around -20. See how well this works?
Click to expand...

Great.

Now track 48 acoustic guitars individually for a string ensemble. What happens to those levels when you mix it ITB?


sl
 
snow lizard said:
Great.

Now track 48 acoustic guitars individually for a string ensemble. What happens to those levels when you mix it ITB?


sl
Click to expand...
Are they all playing a unison part? It really should just work out, you might have to turn down the master buss a little if you aren't using compression or group busses.
 
Quote of the month.

"This is like quantum physics, there are five people on the planet who actually understand it, and four of them are lying."

Quote of the month.:)

.
 
snow lizard said:
Seems kind of scary. It also illustrates that if you reduce a level in a DAW, you're introducing generation loss of a sort at that point.
Click to expand...

I might also note, this was done in a destructive wave editor (CoolEdit) just so I could see what happens to a higher amplitude wave if it were instead put into a lower amplitude range of bit values. Luckily I don't think we have to worry about this in most DAW mixing applications, but I think it does relate to our initial levels at the A/D.

Are they all playing a unison part? It really should just work out, you might have to turn down the master buss a little if you aren't using compression or group busses.
Click to expand...
I agree with this.:cool:
 
I know, I know... you should use a limiter on each track! Have them all set to -6db reduction. Then limit the whole mix... IT'S CRUNCH TIME!
 
Reggie said:
I might also note, this was done in a destructive wave editor (CoolEdit) just so I could see what happens to a higher amplitude wave if it were instead put into a lower amplitude range of bit values. Luckily I don't think we have to worry about this in most DAW mixing applications, but I think it does relate to our initial levels at the A/D.
Click to expand...

I think it would also relate to the final levels in any DAW. In any case it's a cool experiment and I think it shows that a little comparative testing can probably answer most peoples questions about it way better than quantum physics or whatever.

If you wanted to know all the math formulas involved throughout the A/D/A process, you'd be building converters and software and such, not recording audio.


sl
 
snow lizard said:
If you wanted to know all the math formulas involved throughout the A/D/A process, you'd be building converters and software and such, not recording audio.
Click to expand...
On the other side of that coin, if one understands how this stuff all actually works (or at least has access to a good reliable source that explains it instead of regurgitation of a bunch of internet wikiality), there's no reason to have to perform experiments or do comparative testing, get into 10 pages of forum debate, and then have to deal with it all over again the next time the subject comes up. ;) :)

G.
 
And on the other side of that coin, comparative testing is a critical basis for any type of engineering.

As a machinist, I need to measure parts with a micrometer. The micrometer isn't the standard for accuracy, it's a measuring tool. Its accuracy is adjusted according to a standard reference. Two of the main types are called gage blocks and "standards", machined to very close tolerances to provide the calibration reference.

Audio is similar in some ways. The idea of establishing line level through a sine wave test tone is kind of the same thing. Comparative testing to see where your tolerances line up. (eg. calibrating the VU meter or discovering where your converters are calibrated to). Without the process you're guessing.

Admittedly, gain staging is not really all that complicated and shouldn't need 10 pages of regurgitation, but there's a lot of confusion surrounding the topic (for people that don't understand how it actually works and haven't done the comparative testing - eg. defending the false "wikiality" or whatever of "go for the hottest level - use up all the bits") and some people just like to nitpick. :)

Anyway, people come here to try & learn, so the next thread on this subject is probably due within a month. :)


sl
 
So each track should be recorded at -16dbfs average? How do I figure out what meter I'm using for each component? I don't think I'm in full scale meter. My BBE is rated at -4db, so what do I do with that? Can someone point me to a tutorial on this stuff? Thanks...
 
I think that most of the information in this thread while good is overemphasising the need for a particular level. The basic concept of getting a proper level is to help ensure that transients are not distorted by going beyond the limits of the dynamic range of either digital or analog domains and to keep the inherent noise of the gear and medium that you're recording to at a minimum.

By keeping your peaks in the digital domain at around -6 dBFS it helps to ensure that you don't get overs. In some cases it may also help with not having to perform other unnecessary level adjustments though it's doubtful when doing more than a stereo recording since you aren't going to know what the final level should be until you start mixing.

For analog you need to determine the dynamic range of the weakest link in the chain either through calibration and testing, or checking manufacter specs (though they lie on occasion). The house level that you use should then be based around what your analog gear can reasonably maintain.

The crest value (RMS) level of each track is entirely dependent on the track itself. Generally you want to try to keep the RMS level of the track near the nominal level your analog gear so that you get the best signal to noise ratio. If it's off by a bit but still within what your analog gear can maintain in it's "clean range" life is good. Sometimes you may even want to "push" it a bit for effect. Don't overly sweat the actual level, listen and determine from there. Being a slave to meters isn't what great engineering is about, getting a sound that's appropriate for the song and a great performance is your main concern.
 
this is the thread that never ends...it just goes on and on my friend...
 
is that all they charge around here for bringing that up?
 
Ironklad Audio said:
is that all they charge around here for bringing that up?
Click to expand...
It's all I'm charging this week because next week I'll be at Disney World. "It's a Small World" will cleanse anything else you could mention from my mind. Thus the discount.
 
Farview said:
It's all I'm charging this week because next week I'll be at Disney World. "It's a Small World" will cleanse anything else you could mention from my mind. Thus the discount.
Click to expand...

Oh crap, now I'm gonna have nightmares about little Chucky's from Disney World. You owe me $75 for mental anguish Jay. It would be more but this thread has taken it's toll on the number of brain cells that I have left.
 
masteringhouse said:
Oh crap, now I'm gonna have nightmares about little Chucky's from Disney World. You owe me $75 for mental anguish Jay. It would be more but this thread has taken it's toll on the number of brain cells that I have left.
Click to expand...
Now I'll have to sue Ironklad Audio for $150 for putting me into this predicament.
 
You must log in or register to reply here.
Back
Top