dBVU vs. dBFS

stupid stuff

ok so first off i dont know what the heck im doing!! i haev a line 6 toneport ux2 or soemthing like that and i've been trying to lay down some tracks with cool edit and when i select my driver it dont work :mad: ...i know its powered through the usb but my cousin had it working just on that with sonar ..but i dont understand why it wont work like that on cool edit ..any suggestions?

wes
PS ....im VERY new to this and am horrable with computers to begin with
 
There is still the potential for ambiguity with RMS amplitude measurements, since some use the RMS value of a full-scale square wave for 0 dBFS (which corresponds with a 0 dBFS peak amplitude measurement), and some use a full-scale sine wave (which corresponds with typical analog RMS measurements).

* In the case of a FS square wave = 0 dBFS, all possible dBFS measurements are negative numbers. A sine wave of larger amplitude than −3 dBFS would be clipping by this convention.
* In the case of a FS sine wave = 0 dBFS, a FS square wave would be at +3 dBFS.

The measured dynamic range of a digital system is the ratio of the full scale signal level to the RMS noise floor. The theoretical dynamic range of a digital system is often derived by the equation

\mathrm{DR} = \mathrm{SNR} = 20 \log_{10}(2^n) \approx 6.02 \cdot n

This comes from a model of quantization noise equivalent to a uniform random fluctuation between two neighboring quantization levels. For instance, 16-bit audio has a quoted dynamic range of 96.33 dB.

To make an equivalent measurement of a system's noise floor, the full-scale square wave convention is used. A signal which fluctuates randomly between two neighboring quantization levels will measure at −96.33 dBFS with this convention. :cool: :p
 
There is still the potential for ambiguity with RMS amplitude measurements, since some use the RMS value of a full-scale square wave for 0 dBFS (which corresponds with a 0 dBFS peak amplitude measurement), and some use a full-scale sine wave (which corresponds with typical analog RMS measurements).

* In the case of a FS square wave = 0 dBFS, all possible dBFS measurements are negative numbers. A sine wave of larger amplitude than −3 dBFS would be clipping by this convention.
* In the case of a FS sine wave = 0 dBFS, a FS square wave would be at +3 dBFS.

The measured dynamic range of a digital system is the ratio of the full scale signal level to the RMS noise floor. The theoretical dynamic range of a digital system is often derived by the equation

DR = SNR = 20 \log_{10}(2^n) \approx 6.02 . n
This comes from a model of quantization noise equivalent to a uniform random fluctuation between two neighboring quantization levels. For instance, 16-bit audio has a quoted dynamic range of 96.33 dB.

To make an equivalent measurement of a system's noise floor, the full-scale square wave convention is used. A signal which fluctuates randomly between two neighboring quantization levels will measure at −96.33 dBFS with this convention. :cool: :p
 
I think the point is that you aren't supposed to be, and don't have to be, anywhere near clipping when recording in 24 bits.
 
Hi guys, I always have trouble configuring my mic volume....

I'm using Adobe audition 1.5 and when I record, the bar is moving..is that the VU meter? Somebody told me that the recording should not go above -18 dB...is that the same -18 in the meter in Adobe audition 1.5??

I don't know why, but when I record my stuff, it seem the meter ALWAYS go above -18 and it goes red 0dB in louder parts... If I lower the volume, it seem my recording is not loud enough...

Now, I just download a podcast from a web site and some mp3 files, it seem their recording peak is around -2, -1dB in the meter of my Adobe Audition. So I hope they are not talking about limiting our vocal to -18dB...
 
The 'bar going up and down is a peak meter, not a VU meter. -18 on that meter is line level. Your signal should average -18, not peak at -18. If you are clipping, you are recording too loud. If the mix is too quiet, then you need to compress or limit it to get the volume up. The recording level and the mix volume are two separate things that are not necessarily related.

Those podcasts were probably compressed then normalized to get the peak level at -2.
 
Farview said:
The 'bar going up and down is a peak meter, not a VU meter. -18 on that meter is line level. Your signal should average -18, not peak at -18. If you are clipping, you are recording too loud. If the mix is too quiet, then you need to compress or limit it to get the volume up. The recording level and the mix volume are two separate things that are not necessarily related.

Those podcasts were probably compressed then normalized to get the peak level at -2.


Okay, if we average at -18dB at that adobe audition bar, then we can peak at around -3dB right? So it never go too red red....Oh that means adobe audition's bar is showing dBFS and not dBVU??

But sometime when we use compression(with output gain more than 0) then yeah, our track's volume will be higher, but I always feel that the compression process may distort my sound track more or less....
 
VictorGalaxy said:
Okay, if we average at -18dB at that adobe audition bar, then we can peak at around -3dB right? So it never go too red red....Oh that means adobe audition's bar is showing dBFS and not dBVU??
dbVU is only in the analog world, dbfs is only in the digital world. You should never go into the red. Your peaks can be at -3, as long as it is only occasionally (or you are recording percussion)

VictorGalaxy said:
But sometime when we use compression(with output gain more than 0) then yeah, our track's volume will be higher, but I always feel that the compression process may distort my sound track more or less....
You have your choice of either pristene dynamics and a quiet recording, or a compressed loud one. You should not be getting distortion from your compressor. You may have the attack and release times set wrong.
 
And to add - Normally, a pristine, dynamic, clean mix can be compressed to snot a lot better than one that's been tracked too hot.

(Personally, I'd use -10 as a ceiling... -3 is cooking the preamp just a whisker more than I'd be comfortable with. Even a really, really great preamp).
 
so...does -18dbfs average level apply only when going from analog outboard gear into a DAW?

On a self-contained multitrack recording unit does the same rule apply?
 
lacmackenzie said:
so...does -18dbfs average level apply only when going from analog outboard gear into a DAW?

On a self-contained multitrack recording unit does the same rule apply?
It would depend on the calibration of the unit. A self contained unit is the same thing as having outboard preamps, an interface, and a recorder. They just put all that stuff in the same box. The same rules apply.

If you are talking about the Roland VS-series stuff, it is really important to not track very hot. The mix buss on those things is only 24 bit, it runs out of headroom very quickly and turning down the master fader doesn't help.
 
anything over 0dvu clips though. so would the best bet to have a light compressing on the instrument while tracking?..if you have that advantage..also i've been recording my mixes extremely low because im always scared of digital distortion!!! but this thread helped me and i just breezed in here, but i've noticed now that my mixes have suffered due to recording at such low volumes.. -9db (on cool edit meters)

good thread though fellaz
 
(1) Anything above -0.0dBFS is clipping -- Not 0dBVU. Although if you're using older converters, I could see a very dynamic 0dBVU signal clipping it here and there.

(2) Mixes that peak at -9dBFS aren't suffering from anything -- Other than "suffering" from a decent amount of headroom. Nothing wrong there.
 
If I know a recording won't clip, for instance heavily distorted guitars, I suppose there's no need for 18dBFS headroom and I can just as well set the recording level to peak at -3dB in AA. Is my assumption right?
 
If I know a recording won't clip, for instance heavily distorted guitars, I suppose there's no need for 18dBFS headroom and I can just as well set the recording level to peak at -3dB in AA. Is my assumption right?
No. The problem with running the levels that hot on something without any dynamic range, like distorted guitars, is that you are running the preamps and everything in the analog chain 15db above where it was designed to work.

As it probably says somewhere else in this thread, the average level of the signal is what matters, not the peak level. You should be running the analog side of things at around line level, which is an average level.

Where it gets confusing is when you don't have any meters in the analog side of your setup. This only leaves you with peak meters in the digital world. The rule of thumb is that line level equals -18dbfs, however this varies from converter/interface manufacturer to manufacturer.

In order to set recording levels on something like distorted guitars, hit a power chord and hold it. The recording levels should be set so the sustain sits at -18dbfs on the peak meters in the computer. As you play, the levels will bounce around above -18dbfs and that's OK. That's what the headroom is for.

With percussive instruments that have no or very little sustain, you can just set the levels so you are far enough below clipping so no 'surprises' happen and it accidentally clips. (there is no penalty for recording too quiet, but there are big penalties for recording too loud)

The reason you can do it like that is because a preamp, or any piece of analog audio gear, is designed to be able to put out a certain amount of signal level at a certain spec (distortion, transient response, etc...) for a certain amount of time. While it can put out a line level signal at spec forever, it will only be able to put out a +15dbVU (15db above line level) for a very short period of time (micro seconds) without distorting or somehow changing the signal. So the headroom can handle short transients, like the initial strike of a snare drum, very well. But if you try to run a steady signal, like a distorted guitar way up in the headroom, the signal gets changed because the equipment wasn't designed to run sustained signals at that level. So it doesn't have the energy to do it cleanly.

There are other reasons not to record so hot, one of which is plugins. With the popularity of hardware emulation plugins, if you run your levels too hot, you will be distorting the crap out of your plugins. This is because the plugins emulate the behavior of the actual hardware. If you ran a +15dbVU signal into the analog piece of gear, it would distort and do all kinds of unpredictable things. So that's what the plugin will do too.


With all that said, once you are in the computer, the signal level doesn't make much difference at all. (other than overdriving certain plugins) The 'record at -18dbfs' thing is all about how the analog side of your recording chain works and reacts to different signal levels. You must remember that your all-in-one computer interface is an analog unit right up to the point that the signal hits the converters. It still has mic preamps that can be overdriven, possibly insert points (and whatever you insert there), and the whole signal path leading up to the converters that can be overdriven and mess up the recording before it even gets in the computer.
 
Hi,

I'm recording onto a Yamaha AW4416, 24 bit @ 44.1 - what should I be aiming to record my highest level (peak) at?

Thank you.
 
Back
Top