Overall approach / Normalize

[gibsonsgharp]

As I am new to this site, I posted this question on the "Techniques" forum but realized that it might be a more appropriate question for the "Mixing Mastering" forum.

My overall approach has been to record each instrument/voice on one track, recording it with desired EQ/FX/Reverb using external mixer getting each track to sound as desired. Then mixing each track using fader,pan, etc. into the "final" mix. Then, if necessary, adding something via the software, usually reverb or some final EQ, to each track or the mix overall if needed to the "final" mix. Results have been "amateurish". The subtleties of each track get lost in the mix. (i.e. that reverb/echo on the vocal is washed out by the rythmn guitar)

From the response on the other forum, I am thinking a better approach would be

1. Get a better computer so I dont have to worry about software EQ/FX/Reverb eating up processing resources and only use external devices if desired sound cant be achieved by software modules
2. Record Dry
3. "Inital" Mix Dry - using faders and pan
3. Add FX and EQ where desired
4. Add reverb to individual tracks as desired in the mix
5. Add reverb to the overall "final" mix if desired
6. Normalize?

I am new to this , (my previous recording experience was on a Tascam 4 track with cassettes) so I have a vague understanding of what "Normalize" is and when to do it. Does most software do this for you?

Thanks for listening and any advice.

 

[bozmillar]

That seems like good advice to me. I wouldn't worry about the last step though. There are effects that you can use to get your sound loud. Normalizing it just dumb old hard coded volume effect.

 

[Chili]

My overall approach has been to record each instrument/voice on one track, recording it with desired EQ/FX/Reverb using external mixer getting each track to sound as desired. .

Here's where I see a problem. You're working towards getting a great sound from each individual track, but that's not what you want to be doing. You want each track to be able to fit in with every other track. Try recording everythiing dry, then during mixdown, you apply whatever eq, dynamics, etc needed to make every part fit within the song.

 

[Big Jinx]

As some here and on other BBS look harshly at normalizing files, I've been doing so for some time now. Normalizing takes your requested file and raises it in volume to 0db by it's loudest point. (or any other db level below 0). So if you rip off a guitar solo, and at 1:21 in the track, you get a peak, normalizing takes that peak to 0db and brings the rest of the audio on for the ride. However, I would recommend normalizing your files first, then adding eq, comp, FX. Unsure of how many tracks you're recording, so ensure that you drop your fader levels to nominal levels prior to pressing that good 'ole play button to avoid blowing your eardrums out. This way, your files are already at their best gain levels, and your mixing can be done without worry of clipping or distortion on your master fader.

But at the same time, if you're recording between -24db and -12db, normalizing isn't really required. Could just be a compression issue. Understand that if you normalize a low volume audio file you also raise the noise floor along with it.

Many ways to get the job done here, just depends on the route you'd like to take!!

 

[philbagg]

However, I would recommend normalizing your files first, then adding eq, comp, FXand your mixing can be done without worry of clipping or distortion on your master fader.

and your mixing can be done without worry of clipping or distortion on your master fader.

Sorry, but these two don't go together at all. If you normalize before you apply your gain-based and time-based fx, you're taking away your much needed headroom. There's no need for normalization in the mixing process, and it shouldn't be done.

The thresholds on a compressor will go down far enough to work effectively.
The monitors can be turned up so that the sounds are louder in the room, not on the master buss.
If you leave yourself a good amount of headroom, and boost frequencies with EQ, it's fine. If you don't leave headroom and do this, chances are that you're going to run into problems.

I could go on...

I don't mean to rant, but I don't think the OP should be led down this path.

My 2 cents.

 

[SouthSIDE Glen]

[EDIT]I wrote this before I groked step 3 in the OPs post. My mistake as what I'm describing is basically step 3 in his post. I'll just leave it here in case there's some detail that wasn't described to him before. Sorry guys [/EDIT]

You're on the right track, Gibson, but there's one main recommendation I'd like to make make as a change in that plan. Chili wisely alluded to it, and I'd like to expand on it:

Don't make any changes to your individual tracks at all until you've heard them together. Many mix engineers like to start with what they call a "faders up" listen. By this they mean star out with the tracks you have to work with, and just play them together by pushing each channel fade up to somewhere around unity gain (the 0 mark about 2/3rds of the way or so up the fader) and just listen to that rough mix to see how the tracks do and do not fit and sound together. Then *listen* to that faders up mix as your education to start planning what you need to do to each track to make a cohesive-sounding mix.

Some engineers (including myself) like to start from there by planning and setting locations and the moving on to automation, compression, EQ and the rest of the process. Many others prefer to leave everything centered mono first, adjusting each track's EQ and compression and so forth while they are stil stuck together, because it makes it easier for them to hear phase issues, masking, etc. while they are still on to pf each other. Neither method is necessarily superior to the other, it's really personal choice and preference at that point; whichever seems more natural to you and works best for your ears.

But the key either way is to hear the tracks together first to hear how they work as a group, not only as soloed tracks, since that's how their going to wind up. Otherwise, what sounds great as a solo track may wind up being like trying to fit a square peg into a round hole when you hear how it relates to the rest of the tracks.

Oh, and I also recommend not normalizing each track. You're going to wind up having to set the relative volumes of each track as part of the mixing process anyway; peak normalization is just a waste of effort since you're going to have to un-normalize them as part of mixing anyway.

G.

 

[Dogbreath]

JinxDude
I see what you're saying about normalizing and it's uses. Can be good, yeah...but ...

If you're talking about bringing that peak back to zero, it's being tracked too hot. No?
Why not just bring all the tracking and mixing back down to the negative teens, say -15 to -18 with the peaks only going to -12, and leave yourself all that extra headroom to be used at the "mastering" stage so you can get fatter punchier and louder masters?
(that's in quotes cuz my masters suck. I'm talkin about real mastering dudes.)

But GibsonDude

I pretty much agree with 1-5.
imho...I think you could get a similar but cleaner signal by getting good with a compressor/limiter (the part I'm still working on ) and by using some lo cuts (hi pass) in the EQ to the tracks that need it.
And I almost lo cut everything.

Hope it helps mang...

 

[Dogbreath]

Yo BaggDude

most excellent post dude.

but... "boost frequencies with EQ" would be my second choice after grabbin some lo cut first.

jus sayin...

 

[Chili]

Oops, I went back and reread the OP and realized I misread the first time through.

So yeah, when you say:

1. Get a better computer so I dont have to worry about software EQ/FX/Reverb eating up processing resources and only use external devices if desired sound cant be achieved by software modules
2. Record Dry
3. "Inital" Mix Dry - using faders and pan
3. Add FX and EQ where desired
4. Add reverb to individual tracks as desired in the mix
5. Add reverb to the overall "final" mix if desired
6. Normalize?

I think that's a good plan of attack except, as others have said, no need to normalize.

I might add, depending on your computer you might not need a new one. Some plugs can eat up cpu time like reverbs, but if you run only one instance and send all your desired tracks it isn't so bad. Samples can eat up cpu time also, like VSTi's. I used to run everything with a P4!! It didn't skip a beat.

Sorry for the misread and have fun. Welcome to the site.

 

[Massive Master]

However, I would recommend normalizing your files first, then adding eq, comp, FX.

Just wanted to "+1" the point that this is exceptionally bad advice.

 

[masteringhouse]

In addition to normalizing adding reverb to the entire mix is a bad idea. You want to space tracks differently depending on where you want them to sit in the "space". Adding reverb to the entire mix just makes it sound like you are listening to the entire song from another room.

 

[philbagg]

Yo BaggDude

most excellent post dude.

but... "boost frequencies with EQ" would be my second choice after grabbin some lo cut first.

jus sayin...

I completely agree, I just thought it was easier to say "boost frequencies" rather than saying "cut any unwanted frequencies to get it sounding right and then turn it up" which is what I'd normally do. For example, not enough low end on a bass, I cut some of the high/low mids and then turn up the fader.

But, either way you do it, if the tracks are normalized, you're going to clip with additive EQ, and you're going to clip with subtractive EQ+amplify.

But it doesn't matter anyway, because as long is everything is louder than everything else, and you're mixing to +1dBFS, it HAS to sound good



EDIT: I can't rep you back

 

[Big Jinx]

Just wanted to "+1" the point that this is exceptionally bad advice.

My bad...I'm a hip hop guy who does most of his work in Reason. I use the normalize feature for instruments that are tracked out of Reason and into DAW. Should have stayed in my own house for this one, lol

 

[Massive Master]

But *why* would you do that? Why would you want anything anywhere even approaching full-scale while mixing?

 

[philbagg]

My bad...I'm a hip hop guy who does most of his work in Reason. I use the normalize feature for instruments that are tracked out of Reason and into DAW. Should have stayed in my own house for this one, lol

Whether it's a software instrument recorded down to an audio track, or a live instrument on an audio track, makes no difference. The point is headroom, and the problems are the same on both parts.

 

[shortedaman]

as far as i can tell no one has said this. i think maybe the problem might stem from the fact that; this man is not going through the normal routine of: tracking, mixing, mastering. yet he is trying to mix and master at the same time. i mean he said he recorded with his desired fx already in place. so IMO he should be bouncing a mix to a stereo track for mastering. adding his fx for his entire mix there. am i right?

 

[SouthSIDE Glen]

yet he is trying to mix and master at the same time.

If that's the case, it's like trying to put up and paint the drywall wall at the same time, it's just not a good idea. You don't want to paint until after all the nailing, taping, pasting and sanding are long done.

Mastering is the act of prepping a mix for final release. How can you prep a mix before you have a mix to prep? One simply cannot rightly master a mix until one actually has a mix to master.

G.

 

[Big Jinx]

But *why* would you do that? Why would you want anything anywhere even approaching full-scale while mixing?

Because I have made it a personal practice not to mix above 0bd on ANY fader. May be wrong, but no one would ever know if I'd have never said anything!! Guess I'll have to conduct a small experiment. Have a song I'm working on at the moment and I'll try to mix the song both ways and find out how much of a quality difference there will be on the end product.

 

[Massive Master]

Where the fader is makes no difference - The question is why you'd want to have ANY individual element in a mix anywhere NEAR full-scale.

Or are you normalizing and then reducing by 12dB...? Which is equally (actually quite a bit more) odd in itself...

 

[Big Jinx]

Yes people, he know's his stuff!!

LOL, Massive Mastering, you are completely correct!!!

Mixing in Reaper with levels from Reason and right off the bat, mix seems to have a lot more "air"....Damn...Next time I'll have my foot ready so I can directly insert into mouth prior to speaking to the pro's!!! Got the "normalization" thing from another senior member of this board a couple of years back and just ran with it....