Looking for a bit of advice / insight into how other prepare mixes they get from clients?
 
I get audio files sent to me and some are appalling and some are good. But as we all know, as a service we offer we have to deal with whatever hits us. I would like to see what steps other engineers take to prepare the types of audio files after importing them into Sonar?
 
Some Questions

• Sonar has a Trim Knob at the top of every channel, is this something any of you use to balance out the audio before you start mixing?
• there are times I notice my plugins are clipping (or the light indicator on the clipping flashes yellow) How do you deal with this?
• Do you bring down your volume faders for each channel before you start mixing?
• What level to you start mixing at on the Master Bus?
• Do you use normalisation at all during this process?
• What exactly is gain staging? I was under the impression, gain staging is when you use the Trim Knob at the top of each channel to balance out the audio so it all peaks at the same level before moving onto mixing. Is this correct?

I am looking for the best workflow for importing and setting up a mix to ensure I am not overdriving my channels, plugins and Master Bus.
 
Please share your process and ideas with me, it would be greatly appreciated.
 
I have had a few songs submitted to me the past few days and it made me think. I want to deliver the best service as a mix engineer that I can but I am unsure as to how to deal with this scenario?
 
I usually record my own stuff, so I am in more control with the audio. When someone else submits things to me, I have to take what I can get. After all, turning away a mix is a way of turning away money. Sometimes I am forced to tell a client to re-record something, that is understandable, but I will like to get a better idea of how other take this into account and prepare their mixes?

Without answering your questions directly, I will give you my workflow and I think it will address them.
 
I start out with project template which is set up for mixing.  It has a number of busses that I expect to use already set up with the appropriate PC modules and FX in them.
 
I import the audio and list to each track solo'd so I can identify what it is and rename it something that makes sense.  I also route the tracks to the appropriate busses and create aux tracks at this point (e.g. to route a top and bottom snare to one fader).  I use the gain/trim knob at the top to bring each track's highest level down to -16.  I usually do this by using the max value at the bottom of the fader strip to determine how much I need to bring stuff down.  I leave the mix bus and master bus at unity.  My master bus meter is usually peaking at -6 at this point.
 
I then spend 20 minutes or less setting panning and rough levels.  I will usually apply light compression and a little bit of EQ on the mix bus at this point.  After that I will edit the vocals pitch and amplitude using Melodyne.
 
I then set the master bus to mono and start using compression and EQ to unmask each instrument and get it to sit in the mix.
 
After I have satisfactorily eq'd and compressed the tracks, I go back into stereo and begin apply effects like reverb and delay.  Then I work on automation to create tension and feature different parts in the arrangement.
 
Gain staging is probably better explained by some of the real experts here, but as I understand it is managing the gain/levels throughout the entire signal chain so that the correct amount of signal hits each part of the signal chain to deliver the appropriate effect.
 
Hope this helps.

Dave,
 
That is great. Nice feedback. Gives me an idea that I am on the right track with what I do. So, you trim the tracks that much? I always feel like I am degrading the audio by bringing down the levels so much with trim and then using compression, etc... to slowly bring levels mildly back up before I use a mastering plugin at the end.
 
Just want to get a better idea of how everyone is doing things, especially some of the experts.
 
Thanks for your speedy feedback.

I normalize every track to -1 to start with and I cant think of a time where Ive touched the gain/trim adjustment.
 
some tracks are going to be audibly louder than others at -1(as a starting point) just depending on whether or not the source material has some pretty big transient peaks or if its a pretty solid wave.  I dont worry about that too much.  that will be dealt with using the faders and compression/limiting over the life of the mix.
 
next thing is I always start with a channel strip with eq and compression, and console emulation on every track.
 
I'll use a modest 2:1 or less compression with a slow attack(20-30 ms) and modest release on every track and pull the threshold til I see 2-3 db of gain reduction.  Then I use makeup gain in the plug in to bring the volume back to where it was.  I try to avoid changing the perceived volume of the track with the plugin.  So in other words....  when I cycle the plugin on and off...you notice no volume difference.  Just the effect.  this is on the individual tracks.  This holds true with every track.  If I go crazy with 10 plugins on a single track...  it doesnt get consecutively louder with each plugin.  I dial the input or output volume or makeup gain or whatever in order to retain the source volume.  I can bypass the FX bin and, killing all plugins and the audible level will stay the same to your ear but the effects will go away.
 
If any plugins are peaking I either dial down the input or output in the plugin.  Again I dont mess with track gain.  I just do it within the plugin.
 
I will add some drive from the console emulation too, but pull back the volume within the plugin to make sure it only adds drive and not volume.
 
I buss everything to various busses and use the same style of compression described above and often a master buss eq for each bus just in case some overall shaping to a group of instruments needs to happen.
 
Then I mix and pan and automate and play with individual channel EQ all that fun stuff.
 
Once the balance of the mix is where I want it I check the master buss meter.  I want to be around -4 pre master.
 
if the mix is too loud I use the individual track faders to adjust, and by selecting all tracks and holding ctrl, I drag down every fader the exact same amount all at the same time to lower overall volume and maintain balance.
 
if its too quiet I do just the opposite and push everything up simultaneously til I see -4 on the fader.
 
At that point Id walk around the room and listen to the mix better.  Id do a mono test and a few other last minute checks.
 
If I like what I hear, I bounce it out and bring it into a new session to master.
 
 
 
 

greg_moreira
I normalize every track to -1 to start with and I can't think of a time where I've touched the gain/trim adjustment.

 
I have read before that, normalizing your audio is a no go? Just want to see what your response to that is? And maybe other people's responses to that?
 
Some great tips in your post though. Thank you for your time.

I can count on one hand the number of times I've normalized an audio file. And I've never raised any track to -1 dB on purpose. That almost guarantees a bad-sounding master, because 1 dB is not enough headroom for the master limiter to work with. And if every track is at -1 dB, when summed they will NOT total -1 dB on the master bus, but will almost certainly be in the red. (-1 + -1 != -1)
 
On hardware consoles, the Trim control is for reducing the levels of incoming signals, usually to avoid overdriving the board's preamp. Its role has evolved somewhat in the digital era. Now, SONAR calls it "Gain" rather than "Trim", reflecting the fact that it's not just a passive input pad anymore, but rather an early-stage volume control.
 
And yes, if you're dealing with pre-recorded tracks, the gain control should be used to get all the tracks into the same ballpark. There should be a reasonable balance - everything clearly audible - between tracks with all volume faders at unity. That's a good starting point for your mix. (It's what we used to call a "rough", although that term nowadays is just as likely to mean "first attempt at a mix".)
 
Once you've achieved that initial rough balance, look at the master bus peak levels. If it's hitting anywhere near the red, then your tracks are too hot and now is the time to get them under control using the Trim/Gain slider. If you find that one track needs an excessive amount of Trim/Gain adjustment, then that's when you reach for the Normalize feature.
 
 

maybe Im explaining myself wrong.  I normalize everything to -1 before any mixing of any kind is done at any point.
so...to start with....  the raw tracks individually peak at -1.  of course if just left that way its going to smash the master and there wont be headroom.
 
I fix that in the mix, and I dont have any issue getting 4-6 DB of headroom in the master bus.
 
like for example say I get tracks from someone and the bass guitar recorded signal was trimmed nice and hot while recording, practically sitting at -1 as is.
 
and the vocal track signal was trimmed a little low, so 90 percent of the vocal is pretty far down on the meter when the fader is at unity.  I could either use the gain adjustment to bring down the bass to vocal level, or bring up vocal level up comparable to the bass with gain...   or I could just normalize them both to the same number.  I hear no different in the end result, and I dont believe there is in reality.  I dont think that 'normalizing' is doing anything at all any differently than what the gain adjustment within the DAW would do.  You can test this by bringing in a track and duplicating it.
 
Set one with your gain/trim to whatever level you want.  "Normalize" the duplicate to the same level and they'll still phase cancel just fine.  You can normalize it in steps if you want.  Bring it down, then up, then down, then normalize to the same level, and it'll still phase cancel as if nothing has changed.  This tells me that normalizing isnt adding any artifacts of any kind that the trim/gain adjustment doesnt also add.
 
So...  to me it is a much faster way to get all tracks at a similar peak level to start with than individually 'gaining' each channel.  Just CTRL A, and normalize.  
 
You could debate whether starting each track at -1 is a bad decision.  I wont argue with you on that.  I just happen to like having hot signals on all my inputs/tracks as a starting point
 
If I am missing something with my understanding of normalize vs gain though, I'm all ears.  
 
 
 
 
 
 
 

DJ Darkside

greg_moreira
I normalize every track to -1 to start with and I can't think of a time where I've touched the gain/trim adjustment.

 
I have read before that, normalizing your audio is a no go? Just want to see what your response to that is? And maybe other people's responses to that?
 
Some great tips in your post though. Thank you for your time.

I've heard that before as well.  Im just not sure why.  I dont know if there is something going on that I dont know about(which is entirely possible), or if this is just a myth that folks perpetuate that normalizing does something "bad".
 
As I was explaining in my last post....  you can take two duplicating files and adjust one with the input gain in the daw....  and normalize the other to the same level and I hear no difference at all.  Whether you put them both to -6 or you put them both to -1 or wherever you put them,...  So to my ears, and with a phase test, both tracks "seemingly" respond identically.
 
So I normalize simply because its faster.  

One last thing.  If we are talking perceived volume....  then obviously my method doesnt create balance.
 
I tried to explain in my first post, a solid wave with minimal transient peaks normalized to -1(or using gain to put it to -1) is going to be a lot louder than a very dynamic instrument/voice with big transients also set to a peak of -1.
 
they wont be level matched. They will 'peak' at the same point.....but obviously one is louder than the other overall.  And RMS check would show this too. 
 
If you're trying to level match all the tracks as a starting point...  then you need to play with gain.  Normalizing cant do that if balance from the start is what you want, so sorry if that caused confusion.
 
 

ok guys some interesting advice here, i am a little confused, i understand there is "normalize" found in the process menu under "apply effect" and there to is the "gain" function, but there is also a "gain" knob at the top of the tracks strip and there is a "gain" automation function by pressing t in the highlighted clip then choosing clip automation-gain.. and moving the gain automation with ctrl up or down with mouse...so which "gain" do we use? also for reference i am looking at a groove 3 vid with Kenny Gioia, mixing in reel time..he is using reaper, he normalize the track in question then fine adjusts with gain to get a level in a rough mix..so this is a bit complex..it seems there is no right or wrong way???
Peter...learning as i go along....