Audio Level Advice for Recording and Mixing

onetruth

Member
Apr 21, 2006
100
0
16
Worcester, UK
www.myspace.com
Hi everyone,

I joined this forum ages ago and have picked up some excellent advice over the past few years, but I'm still learning and have a few questions to ask.....hope someone can answer them for me! :) Also, this is my first post....so please be gentle!

I recorded a band a few months ago at a small studio in herefordshire - my first proper paid recording session as an engineer! Managed to record 3 tracks in 2 days......drums, bass, 2xguitars + overdubs (solo's etc) and vox. I then took the project back home and began mixing on my project studio setup. I was constantly refferring to threads on this forum for mixing advice and all seemed to be going well (to my ears) until I came across the thread about gain staging......

I've read the thread on this forum and been on the PSW forum. Also read this article here: http://www.massivemastering.com/blog/index_files/Proper_Audio_Recording_Levels.php which has opened my eyes a LOT!

My problem is that I didn't know much about gain staging during the recording session and so I'm not sure if I recorded the instruments 'too hot' or not. I know for sure I checked that their was no clipping while tracking and so must have left a bit of headroom.....but looking back, I think I was aiming close to 0db. a mistake!? possibly?! but now i'm at the mixing stage, i'm trying to figure out how best to mix ITB, in terms of levels. From reading through the various threads on this subject, the main piece of advice I saw repeated a few times was to mix at an average level of -20dbfs with peaks around -10 and -6dbfs. have i got that right? so in practice, lets say ive got the kick drum on a track with the fader on 0. Do I just drop the fader down so that the hardest hits are peaking around -6 and -10dbfs? Also, is the meter on each track in cubase 5 measured as dbfs? I'm currently at work so not at my recording PC to check!

I also read it may be a good idea to put a trim plugin on the first insert of every track. Is that a good idea? If so, would I then set the trim to drop the level down to an average of -20dbfs? thereby allowing me to leave the fader at 0?

I also read that the master bus should be averaging between -18 and -5dbfs. Is that right?

Any help that you guys could give me would be greatly appreciated. I'm rather new at recording/mixing but have a decent ear....i can't afford to quit my job and go on a course to learn all about recording properly so just trying to learn by reading up on forums and trial and error really. like i said, this is my first proper paid job for recording/mixing (only £100 though.....the band are friends of mine and none of them have jobs so Ive done it more as a favour) so i'd like to do the best I can for them.

Thanks in advance!
Jim
 
I also read that the master bus should be averaging between -18 and -5dbfs. Is that right?

Maybe someone could confirm this, I'm not a professionnal AE, but I think you may be talking about dbRMS, wich I think is more or less your average level. If I'm not mistaken dbfs is the scale used for digital metering, not audio metering.
(Unless that's what you were talking about, in this case, nevermind!)

The thing is you CAN have, say, -3db peaks and still have a reasonably low RMS level, wich is kind of what you want (headroom).
 
RMS is the power of the signal (an average over a short period of time). Less dynamics is a higher RMS. Peak and RMS are related but you can have a clipped signal with a low RMS too.

When someone says "your average level should be around -18dB to -15dB" that means a little lower and a little higher at any given time. It's not an absolute value. This is the ideal level when tracking to a DAW. This level (depending on what the converters are calibrated to) will be roughly 0dB VU to +4dB VU where all analog circuits (including your mic pre on the interface) are designed to perform optimally (lowest noise, lowest distortion, and fastest transient response). Beyond that level you're pushing the circuit past it's design. if you track at 0dbFS (or just below) that's running your preamps at +22! Way out of spec.

It's not the 1980s and converter design has improved tremendously and even the cheapest converters don't have the problems of the early generation AD/DA designs. "using all the bits" is nonsense.
 
Honestly at this stage in the game I wouldn't sweat it too much. As long as you didn't clip anything, the recording will be just fine as long as you use your ears. This is your first gig, do the best you can, recording too hot is the least of your worries first time out. Plenty of mistakes to be had later in your career. And this isn't even a big one, and debated by pros.

There is a big range of metering and such, but personally I don't think any of that is super necessary. I have heard great recordings pushed to the stars and beyond, essentially hitting the internal DAW bit limit, etc. May not be the purest most dynamic thing, but at the end of the day it sounded way better than most recordings out there.

But you do want to leave headroom for mastering. So try this first, just make sure nothing peaks over -6dB on your master meter. And leave the master fader at Unity. As you go this will jump, pull stuff down to keep that peak that low. And keep the master bus clean, or maybe a very slight bus compressor if you prefer.

If your peaks stay at -6dB (usually kick and snare) and you aren't compressing the holy hell out of each track, then I bet your dbRMS will be about -16 to -18.

Then if you want to tread into faux-mastering, then work with that. Of course here your monitors will need to have an external volume control outside the DAW, and it will likely be cranked. So turn them down before listening to mastered material.

Then check clip lights on plugins, and of course listen, listen, listen. Just a few good fundamentals with the rest, and you will be fine.
 
Oh and to add... don't over-research and over-think it as you are doing. Mixing is an art as much as it is a science. Just make it sound good, at the end of the day, that is all that matters. Shoot from the hip if you have to.
 
Headcrusher - spot on! that thread totally confused me!

AudioGeekZine - thanks for all that info. if I may, can I follow up on a few points you made there? Let's say i was to go back into the studio with the same band and record another song later down the line.......and lets say i'm getting the level for the kick drum for example, would i keep the fader in my sequencer for the kick drum track at unity gain and then adjust the level of the mic pre accordingly to get the audio level meter on said track to float around the -18 to -15dbfs area?

Also, I'm a little confused about the calibration of my audio interface/mic pre's etc........you mentioned that the mic pre's on my interface should be calibrated to something like 0db VU to +4db VU. Ive also read that each interface and mic pre's are all calibrated slightly differently. I've checked the spec for the recording interface I use in my project studio but couldn't see where it showed what it is calibrated to. The device I use is the Tascam FW1884...here's a link to the manual: http://zmtest-a.com/content/downloads/products/229/fw1884_manual_e_1_00.pdf
would someone mind taking a minute to check the specs and let me know what my device is calibrated to? I'd be very grateful! How then does knowing this calibrated value help me to run my pre amps at the correct level and not push the circuits past their design? As long as the meter on the track in my DAW floats around the -18 to -15dbfs area, does that prove that I must have the pre-amp at a reasonable level and therefore have plenty of headroom? Or should I be also monitoring another level somewhere? Or measure the level but as a different value? Ie. not dbfs?

Apologies for possibly asking some terribly basic questions! I really do try to read as much information and guidance for recording/mixing on the web but sometimes, no matter how much I read, I can't quite understand. I guess I need certain topics to be explained to me in lamens terms....maybe with a nice analogy! haha :)


aortizjr - many thanks for your advice......maybe i am getting a bit too caught up in all the science rather than just trusting my ears and "feeling" the music/mix?? I'm 99.9% positive that nothing clipped while tracking so fingers crossed, i'm ok there. All the waveforms look good in my sequencer with no obviously huge spikes/clips. brilliant advice about what I should do with the master fader while mixing.....thankyou so much!! i'll definitely make sure there are no peaks above 6db on the master while i'm mixing. i was thinking of running a gentle comp on the master bus.....only compressing 1 or 2db just to add glue to the mix? i've read that can help a lot while mixing?

can i just pick up on the point you made about the kick and snare peaking at -6db and the compression levels of those tracks? when I compress the kick drum track for example, should I be aiming to get for instance, say around 5 or 6db of gain reduction and then use the makeup gain so that the audio level hitting the comp and the level leaving the comp are the same? I noticed on the thread regarding gain staging that a few people mentioned about keeping levels of plugins (eq's, comps etc) the same on the way in and out?

Another question I have is if I did record the audio quite hot, when I run an individual track through an eq for example, will I be unable to push certain frequencies up by a considerable amount? So if I have an eq on the first insert of a track, will the audio level hitting the eq remain high even if I adjust the fader of a track? And so would I then need to lower the input volume of the audio within the eq plug, make my boosts or cuts, and then adjust the output level of the eq so it matches in the input level?

ahhh i'm glad you mentioned faux-mastering......i have a quick question about that! haha. Lets say I mix the track and I'm totally happy with it. Should I then mix it down to a .wav file, start a new project within my sequencer, import the wav file onto a track and then attempt to master it somewhat with plugins etc? Or could I just add some mastering plugins to the master bus in the mix? as for a good suggestion of plugins used in mastering and also the chain....what would you guys suggest? I've read many articles regarding this and was thinking of running something like this:

A good EQ to start
A tube saturation plug to add an analogue feel
a good compressor (compressing 2db max)
a tape saturation plug
a clipper such as GClip to raise the volume a touch?
and then a limiter, such as Voxengo Elephant to get near to commercial CD volume?

does that sound like a good idea?

(really sorry this post is so long by the way! i've just got so many questions and I don't know any recording engineers in my hometown to ask!)
 
A good EQ to start
A tube saturation plug to add an analogue feel
a good compressor (compressing 2db max)
a tape saturation plug
a clipper such as GClip to raise the volume a touch?
and then a limiter, such as Voxengo Elephant to get near to commercial CD volume?

does that sound like a good idea?

no...just use a limiter for now. if you need to ask, you'll probably do more harm than good piling on a ton of processing. elephant will go as loud as you want...anything beyond that should be fixed in the mix.

bouncing/importing into a new project is fine, so is just inserting a limiter to the master bus. doesn't really make a difference apart from CPU usage. I like mixing into a limiter, atleast a little.

and you shouldn't worry about gain staging anymore, if your 2bus is peaking at -6 then that is entirely quiet enough.
 
thanks for the advice darthjujuu!! i'll try just mixing into the limiter and see how I get on.

excellent! so as long as i keep peaks on the 2bus at -6 then i get forget about gain staging and breathe a sigh of relief! haha :)

with regards to my question about input and output levels into eq's, compressors and other plugins on the inserts of individual tracks - was I right in saying I need to make sure they are equal? thats probably the last thing I really want to be clear on before I get too far into the mixing process. if you could advise, i would really appreciate it.

many thanks!
Jim
 
DAW faders should be at Unity when getting levels or showing Pre-Fader levels (possibly an option in your DAW)

As long as the meter on the track in my DAW floats around the -18 to -15dbfs area, does that prove that I must have the pre-amp at a reasonable level and therefore have plenty of headroom?
That's right. Even if the drummer slams on the snare you should be safe but also running your gear in the optimal range.
 
As long as the meter on the track in my DAW floats around the -18 to -15dbfs area, does that prove that I must have the pre-amp at a reasonable level and therefore have plenty of headroom? Or should I be also monitoring another level somewhere? Or measure the level but as a different value? Ie. not dbfs?
Here's a pic that shows how your analog signal measured at +4dBu relates to the digital scale dB full scale.

Most analog gear in designed to work optimally around + 4dBu,... but has a bit of headroom above that.

You can see that if your hitting the digital scale anywhere close to zero, that you'd be running the analog recording chain pretty hot.

xcq6ms

...also depends how your converters are calibrated, but most are int the -15 -20 area.
 
thanks for your replies AudioGeekZine and waltz mastering.......thats cleared things up for me as far as tracking into my DAW goes. top stuff....cheers guys! :)

before I knuckle down later today and start mixing.......should I be running a trim plugin on the first insert of each track to reduce the level? (if I find I did record the source material too hot). And if so, is there a trim plugin built into cubase for example? Or can anyone recommend a freeware trim plugin? I've searched google for "trim vst plugin" but cant find anything.

Cheers
Jim