logical Steps? or absolute nonsense?

  • Thread starter Thread starter tvaillan
  • Start date Start date
tvaillan

tvaillan

New member
Are these logical steps that a sane audio engineer would adopt?

1- record tracks.
2- pan to taste and fix volumes.

Now, repeat for all tracks,

3- export each track one by one
4- apply fx if needed and then normalize the wave.

What i'm talking about here would be 'corective' EQ'ing and Excitation such as the BBE Sonic Maximizer, not compression, stereo imaging or reverb. The goal is to bring out specific instruments' signature frequencies.

6- re-import all into mixing software
7- re-pan tracks and adjust volumes to taste.
8- export everything as single stereo WAVE
9- import into 'post mixing' software and preform the following post mixing steps:
10- apply corrective eq'ing if necessary
11- carefully apply multi band compression,limitting ..
12- add mild overall ambiance reverb
13- stereo expansion
14- export to WAVE and MP3
15- post to MP3.com

Let me know what you guys do!

Am I on the right track?
 
well yes and no :)

I'd miss 3,4,5?,and 6. Do your EQ in realtime, DON't sonic max or stereo image.

I wouldn't do 12 and 13 either.

cheers
JOhn
 
I have to agree with John. The over all reverb MAYBE but that would be something that's song specific and not sure I'd even do it then (maybe on something like a CS&N kind of thing where it's a solo guitar and voice) and never the stereo expansion plugs. Yikes!
 
so how about this?

1. record
2. normalize all waveforms
3. 'mix' appropriatelly . (panning, volume)
4. apply realtime EQ'ing and/or aural exciter on the tracks that need it.
5. go back to step 3. repeat as necessary. if too much EQ'ing is necessary, do another take instead of butchering the frequency spectroum and phase relationship.
6. export stereo wave and import into 'post mixing' software.
7. lightly 'touch' the mix with EQ if it's needed.
8. apply multi band compression if needed
9. apply ambiance reverb if called for
10. call it a day.
 
This forum rocks! (sometimes literally)

Thanks for the input by the way.

And ...

Is there a reason why sonic maximize and stereo imaging is a bad thing? Is the effect better aproximated with creative use of short delays?
 
Why no stereo enhancement?

I've heard some great recordings that used this sort of processing. Unfortunately, I have no clue how to use it But if I did, I would. Tosca "Suzuki In Dub" is and amazing sounding recent recording that has extremely open and detailed spatially enhanced sound. I'm certain it can be done well. But how?

barefoot
 
Seems like normalization of the raw tracks is a bad idea for two reasons. When you mix all these tracks together you're gonna have to cut the volume on each from 100% so as not to clip the mix, so why boost and then cut when you could just leave it alone? Second- you might need some headroom in the raw track if you apply any EQ boost. Normalize last, if at all.
 
Regarding normalization I definitely agree. I have read some very good information WHY NORMALIZATION IS BAD unless you are in the 32 bit realm. Besides, when you normalize aren't you admitting that your recording levels were not correct in the first place?

Normalization is wonderful for voice TV and audio production, because it's a time saver and you are under deadlines, and audio quality is not as important as recording a music CD.

I still have a BBE 462 Sonic maximizer for sale. There are much better processors available today.

I love the Waves Stereo Imaging. Just don't overuse it. But I have also learned that there are a lot of horrible stereo enhancement programs so be careful.
 
There is no reason to normalize individual tracks. Normalizing the final mixdown is not necessary either unless you just didn't pay any attention to what you were doing.

Now compressing is often done at the track level, and limiting at the final mixdown. Or whatever, there are no "rules", it just depends on what is necessary. What is necessary will be dicatated by your eardrums. Personally I've found that you should really try hard to MIX in volume, instead of relying on tools to add volume for you later. You can get a really loud mix if you just pay attention to your spikes and listen for WHY they are happening. If you mix big, you can master big!

I would also avoid against panning right off. I have to literally force myself to not pan shit out until I'm really at the right stage in the mix.

You're really trying to come up with steps that don't necessarily exist, or else you're leaving out 1000 steps that "could" exist. Either way it's fruitless. I would revise your procedure as such:

1) Track/Re-track
2) Mix
3) Repeat 1-2 until the mix is done
4) Perform any necessary "mastering" processing
5) Post to mp3 clinic; Repeat 1-4 as necessary
6) The song is done, have a beer
7) Repeat steps 1-6 for the rest of your life

Saying that "adding reverb" is a "step" opens the door to every single kind of processing that could be done. Are you going to EQ, add reverb, compress, delay, fade, resample/convert, degrade, offset, shift?????? You never know. If there was a procedure, we'd all be following it!

Slackmaster 2000
 
Slackmaster2K said:
I would also avoid against panning right off. I have to literally force myself to not pan shit out until I'm really at the right stage in the mix.
Click to expand...

Slack, if I understand you correctly, you're saying to get the mix right in "mono" (down the center) first, then splay it out in stereo afterwards?

barefoot
 
Yes, as a general rule of thumb it will produce a better mix...and a mix that translates to mono better. I read about it around here and in magazines and when I finally started trying it, it definately improved things.

It's hard though because if you're recording yourself, you tend to start mixing everything together and panning shit as you track....well, at least I do :)

Slackmaster 2000
 
wow . thanks for the response

i'm glad this didn't turn into a battle.

I've learnt some thing, thanks guys and gals.

Summary:

So normalization may cause problems, it can cause clipping. I'll shoot for -3db as peak when recording and leave it alone from there. If the sound isn't what i want it to be , I'll record again and again until the levels are fine. I'm ok with that, it actually makes perfect sense.

The more I learn, the more it seems that all the fancy plugins and do-dads are just there to help fix problems. It's best to be happy with the recording then it's easier to work with.

When mixing everything, it may turn to mush ...

A small tad of excitation (bbe sonic maximize + others) may be in order sometimes, but in small doses. This is to make certain instruments 'come out' ... but not all instruments can 'come-out'.

EQ is for fixing problems. If the recording is good than there is no need. If after that the whole mix doesn't sound so good, one 'may' use a bit of EQ (cutting is better than raising) to bring out certain signature frequencies for that specific track.


Practice makes perfect.

The best tool is my ear.

You learn every day.

I'm happy now. Thanks again.
 
Yo Tevaillian "da" Alien:

If you really want to see the antithesis of "logical steps," open the manual for the DAW 2816 by Yamaha.

I was just studying the manual as I have a session to do tomorrow with the unit. I'll get it done but after that, "...slowly I turned...step by step...until I had the engineer who wrote this manual by the neck...poco moco...."

Green Hornet:D :D :p :p
 
whats the difference between normalizing your final mix and using a multi-comp or limiter on your final mix?

if you recording and mixed perfectly, you wouldn't need either. if you recorded well and mixed well, but the final mix output was low, you'd need to normalize. if you record ok, and mix ok, you'll need the multi-comp to smooth out your dynamics and push your average volume level up.

I wouldn't normalize an entire set of songs, but i would normalize an individual track if it didn't need the multi-comp.
 
But why the heck would you mix for a low level? You wouldn't get a hot mix going but have it peak at -6, that wouldn't make much sense. I suppose in a more typical situation where your peak might be at -0.5db or so it might be beneficial to normalize, but not that beneficial. Besides, in many or most cases a bit of limiting (with a decent limiter) is going to help more than hurt.

Slackmaster 2000
 
any good reference links?

any good links that detail out all of the differences between limiters, compressors, etc? THANKS!
 
on the subject of normalizing etc

my guess is that the following would be an application of multiband compression and not normalization., and 'may' serve to demonstrate the difference between both.

Say a singer is singing to the point that things shatter .. then she transistions into a softer more 'whispery' style of vocals.

Wouldn't this be an application of compresison, since if you didn't do anything to compensate for the dynamics of the vocalist, the 'whispering' would get lost in the mix?

Normalizing would simply bring her highest level to 0db's and the rest of the volcals are scaled by the same factor and therefore the more 'whispery' part of the tune is still lost in the mix.
 
Yes, but not necessarily multiband compression.

Slackmaster 2000
 
you're right.

thanks folk.

This thread made the world of difference for me.

Now that I have that question answered , it's time to dive into building some drum triggers/pads for my drummer.


hrmm...
:confused:
 
Also keep in mind that everytime you run a plug on a track your killing some of the fidelity of that track just by running another process on it and resampling the wave. That said, there's absolutely no reason to normalize a single track. I'd also not do anything else to a track unless it's really necessary. If you search the site, you'll find a thread that was done a while ago that shows just how plugs can adversely effect your sounds.
 
You must log in or register to reply here.
Back
Top