I really need to read and understand this thread. You guys are awesome.

Hi there Sonico This is a good question and I should have made myself a little clearer on this. OK very transient sounds such as kicks and snares and drums overall are going to behave differently to most other sounds.

Most sounds have an rms value that is present long enough for a VU meter to read well but very transient sounds don't. The VU's are slow to respond to very transient sounds. 

What I do with them is basically use your peak meters to simply keep an eye on those sounds. The VU's are OK as well but just go for them to reach say -5 dB or so. They will move but just not as high on the scale as most other sounds do. (this does depend a bit on the drum sounds involved. Very thin, fast sounds will not move the VU's as much as say fatter and bigger kicks and snares)

Send all your drums to the drum buss and when you put a VU on that you will see that when the whole kit is present the VU's will start responding more normally especially with toms present. Watch your peak meters and maintain levels so that the peaks still do not hit close to 0dB FS but observe where the average reading of the VU's are on the drum buss. You will see it is somewhere around -5dB or so. When engineers are using VU's to monitor drums they are often looking around -5 dB or so.

By the time you are VU metering the masterbuss the VU's will be responding normally and you will still be hearing your drums clearly too.

Now here is another head scratcher for me...if I am, say, using ambient sounds such as ventilator fans or crickets and the like would they be classified as transients?

And any suggestions as to how to play those up a bit?

Sound effects are either usually pretty slow rms wise unless they contain a loud transient within the effect. 

If they are smooth rms wise, I would still have them sitting at the ref level and set the desired amount of level of the effect in your final mix with the channel fader.

If there is a loud transient within the effect them I would be watching the peak meters to make sure the transient does not clip and let the rms level sit where it happens to be at. But then I might be looking at the transient and perhaps limiting the maximum level from the transient and get the smooth part of the effect back up to the rms ref level. eg If a transient reaches -2 dB but the rms level is around say -18db (assuming a -14 dB ref level) then I would limit the transient to -6db and add 4 db of gain to the whole effect bringing the overall rms level back up to my -14 dB ref level. 

Before doing that a transient that reaches -2dB and has an overall rms level of -18 dB means the transient is 16 dB above the smooth part of the effect and that is going to be hard to control later anyway. If you are working at K-14 then a transient can be 14 db above the average level before it is going to cause any grief and that is a lot. 

Even if a sound effect does not contain a loud transient but has mainly a smooth overall rms type level but varies in level I am only interested in getting the loudest part to reach the ref level and that is it. All other softer levels will fall below that and hence maintain their dynamic relationship to the loudest part.

If a sound effect does consist of a one shot type sound there will usually be plenty of time for the rms level to develop so all I am aiming for in that case is for the loudest part of that one shot sound to just reach the ref level or 0dB VU.

VU meters?  You guys (esp Matt and Jeff) are awesome at explaining what I can't. 

Hopefully, the Sonar (Vertical/Horizontal) track meters are acceptable in this regard ... as I believe I subconsciously observe their mean and peak levels?

Of course the ears and precise headphones may help ... and/or ARC monitoring.  ... depending on preferences.

Personally, I stay off the trims and busses ... and prefer the red clip-gain envelopes or volume envelopes ... depending on the clip solidity/performance ... on the track level only.  But each of us works differently ... to envelope, tweek up, and reduce our levels.

I find that adjusting buss levels generally results in destruction for me ... due to the delicate vibes of the tracks being so relative .. one another.

IOWs ... Most of my tracks seem to become 'irreducible complexities' ... similar to cascading enzymes or DNA codons ... I suppose.  One deleterious mutation and the whole building collapses.
 
Fortunately, song design seems intuitively forgiveable with many awkward levels tweakable in many ways ... to be exqusitely polished in various loving manners.
Unfortunately, polish takes me days-weeks ... via manual compression of such 'interdependent' and (ultimately) irreducible tracks.  And going in circles ... over and over again ... is par for my neurosis.

Out of interest I actually don't mess with buss levels much either. I just find I tend to tweak the channel faders to get the desired mix on a buss and most of the time the buss VU's are also just hitting 0 dB VU mark or a bit under which is about right. 

Even if I have got the perfect mix on a buss and the VU's might be hitting say +3 dB VU, then I might just put all those tracks into a group temporarily and pull them all down a tad to get the buss meters sitting back to 0dB VU again.

But if you are sending say three or four buses to the masterbuss, in order to get the masterbuss just hitting 0dB VU then you do have to get the individual buses just falling short so that when they sum you end up with the right level on your masterbuss. So I tend to still leave the buss faders at unity but the VU's in reality are falling a little shy of 0dB VU in the end which is not a bad thing anyway. It just means that buss has got even more headroom now than it had before which is surely not a bad thing. I rarely alter the masterbuss fader, it just sits at unity. If a masterbuss starts pushing a VU hard eg up to +3dB or more then I just put all the busses into a temp group and just lower them down a little to get the masterbuss back to normal.

I find when you track using VU's you have got this equal loudness thing happening on all your tracks and you end up with the channel faders also being in nice positions and about the same or close (depending on the mix of course) and you are never having to add or subtract huge amounts of gain on tracks anywhere in order to get the desired mix. 

As I have mentioned plugins can screw things around though so you need to be careful with your input and output settings. But with care you can also set all your plugs to not be clipping anywhere and still send out the right amount of rms level afterwards. I think most plugs prefer not to clip anywhere unless of course there are some that do need to be driven harder to get the desired effect. But that is OK too because all you have to do is lower the output level to get your ref level coming back out and driving the VU meter nicely.

The VU plugin concept is also good because you can put them anywhere in the signal chain which can be very handy at times. They don't use any CPU resources either.

Thanks Jeff!!

Good to know the way you approach this, that's what I've been doing too.

I have a Presonus Firestudio Project (that I love by the way) and their preamps reference level for 0 dBFS is +10dBu wich I understand gives me 10 dB's of headroom.


What I do is track close to 0dB VU (-10dBFS) and since I started doing this, my mixes sound warmer and more open. 

Could it be that the preamps work closer to their sweetspot when not pushed close to 0dBFS (+10dBu)?? 

Jeff Evans


The only thing I find a bit weird now about the BlueCat meter is there is no actual serious marking at 0dB VU but rather a change in colour only. (the nearest marking is at either -2dB and +2 dB) I know it might be nit picking but I find it harder to read in some ways compared to the Klang meters which have a definite 0dB VU mark and it is nice and clear and obvious. 

But the BlueCat as Bristol says does a lot more as well and it is a nice meter too. I am little biased now toward the Klang meter because it visually looks like a real VU and can be set to behave like one too. 

uuummmm...now...just how does someone set this up to act like a VU meter?   

Hi there trimph1. The BlueCat meter does a great job of being a VU meter. It is very accurate and even the ballistics of the meter can be similar to a normal VU meter. But as you can see the display is more the horizontal bar graph type setup but it is perfectly workable and will do all the things I mentioned a VU meter should do.

It is just down to what type of display one prefers. The Klang meters of course look like a normal VU meter and I also like that type of display. I use both at times actually. Also the BlueCat meter offers a lot more in terms of what it can do hence the extra expense.