Recording and Production tips

[AndyR]

Another point i forgot to mention is if you are grouping tracks.

say 4 guitar tracks (2 left 2 right)

to there own bus and you are EQing dont just put the fx on the bus header do it to each track individually it will sound far better that way

+1 !!!

As I understand it, what you're saying is:

- If you're going to bounce 4 guitars to a sub-mix for later mixing with the rest of the band,
- do the Effects/EQ on the guitars individually before or as you bounce them...
- Try not to have to Effect/EQ the sub-mix afterwards

I learnt this on my "If We Should Sing Together" (I had to EQ the sub-mixes quite a lot, and put up with some severe compromises - I can't hear them now, months later, but I do know the mix could have sounded a lot better).

However, I seem to recall that it's quite time-consuming to do on the MBR, and a bit of a "shot in the dark". To get the Effect/EQ on to a guitar track, you need one of the effects patches, this can operate on one track, a stereo pair, or the master. This means you need to effect and bounce every track individually. So, on the MBR, for 4 guitars, it's 4 times this lot:

1. Select guitar track to work on
2. Make a guess for the Effect/EQ (in relation to the other 3)
3. Bounce it to a clean track

Then mix the 4 together with the appropriate levels and panning...

- You can do it...  , but you can see why I didn't go back and do it on that song (at least, not for all of the tracks)... mainly because:

DON'T FORGET YOUR EARSDON'T FORGET YOUR EARSDON'T FORGET YOUR EARSDON'T FORGET YOUR EARSDON'T FORGET YOUR EARSDON'T FORGET YOUR EARSDON'T FORGET YOUR EARSDON'T FORGET YOUR EARSDON'T FORGET YOUR EARSDON'T FORGET YOUR EARSDON'T FORGET YOUR EARSDON'T FORGET YOUR EARSDON'T FORGET YOUR EARSDON'T FORGET YOUR EARSDON'T FORGET YOUR EARSDON'T FORGET YOUR EARSDON'T FORGET YOUR EARSDON'T FORGET YOUR EARSDON'T FORGET YOUR EARSDON'T FORGET YOUR EARS DON'T FORGET YOUR EARSDON'T FORGET YOUR EARSDON'T FORGET YOUR EARSDON'T FORGET YOUR EARS DON'T FORGET YOUR EARS  DON'T FORGET YOUR EARS  DON'T FORGET YOUR EARS  DON'T FORGET YOUR EARS  DON'T FORGET YOUR EARS  DON'T FORGET YOUR EARS  DON'T FORGET YOUR EARS  DON'T FORGET YOUR EARS  DON'T FORGET YOUR EARS  DON'T FORGET YOUR EARS  DON'T FORGET YOUR EARS  DON'T FORGET YOUR EARS  DON'T FORGET YOUR EARS  DON'T FORGET YOUR EARS DON'T FORGET YOUR EARS   DON'T FORGET YOUR EARS   DON'T FORGET YOUR EARS  DON'T FORGET YOUR EARS.

                                 that really is enough. Goodbye. henwrench

My ears (brain, wife, and every fibre of my being) said "f*** it, you've done enough, it sounds OK already..."

=====
This one is VERY important, and I don't follow it enough:

Always press record!!!!!

It's not just "make sure you're recording when you think you're recording" - nothing worse than getting psyched for a take, doing the take, and then finding the stoopid engineer (me) had pressed the wrong button...

It's this as well: why not record your practice and "working out the part" takes? I have to get into the right frame of mind to do this, but it has enormous advantages. Depending on the style of music, the take when you first get the "right idea" for the part is often good enough for the recording. All the subsequent attempts are "learning" to play that part consistently, and then the real take can take ages... But often, that first inspirational "take", with minor fixes, is probably the best you'll play it... if you don't press record everytime you play along with the thing, no-one (including you) will ever hear the best performance.

[pjd1]

Some great information on here Geir and lads ..

Dunny

[AndyR]

Got some new stuff to post -

I was putting this in the interesting "Loudness war" thread in General Discussion, but then realised it was useful here. So I came over and expanded it here instead.

"Loudness" in this context is about getting your finished recording LOUD. To do this you've got to compress the peaks so that you can turn everything else up. The downside is that you loose dynamic range and "natural" sound.

I always master the final mix, but nowadays I tend to use one of the BR presets without fiddling with it. Partly for speed, but partly because when I fiddle with the settings I usually overcook it. After a while I reasoned that the presets are a good starting place, and if you like what a preset's sound is anyway, just use it! So I listen to the final mix through the presets and pick the one that sounds best (to me) for that song...

HOWEVER - there's something else I do, which seems to mean that at least three of the presets always sound pretty good.

I apply "mastering" to instruments and submixes. On the MBR I'd use the MASTERING TOOLKIT to do it, but on the BR1600 I don't need to - I have indepedent EQ and Compression available on all tracks.

For example, I compress the rhythm electric guitars together as a group if they need it, I do the backing vox together, and so on.

Usually EQ has taken place on individual instruments (see previous post) when I was creating the sub-mixes, but I will apply further EQ if necessary. HOWEVER, I try to stick to the rule that I only cut on a sub-mix. If I boost, there's more of a danger that I add something unpleasant that I can't hear at the moment... so if the sub-mix seems to need a boost in the low end, for example, I'll consider recreating the sub-mix and fixing it in the individual tracks.

The reason for "mastering" the submixes before mixing them together is this: different parts of the arrangement are going to peak at different places in the song. When you've mixed them altogether, all you can do is apply compression to the whole thing. So, for example, to fix one set of instruments, you have to squash the whole song. Sounds ok at first, but after a couple of days you notice that the detailed acoustic guitars, or whatever, have disappeared....

By gently smoothing the submixes where necessary, you've removed the nastiest peaks and left as much of the dynamics as you can. Then when you combine the sub-mixes, a) they seem to fit together better, b) you don't need to use nearly so much compression during the mastering of the final mix.

Basically all you're really doing is fixing the parts before you mix them together. But as well as smoothing the individual instruments, you're recognising that if you put three or for instruments together that each go:

dum-DEE-dum-dee-DAH!!

Then the sub-mix is going to have possibly more dynamics for DEE and DAH!! than you intended (and far more than your final mix can cope with...)

Folks are always kind enough to say my recordings sound "professional". I'm not so convinced meself , but I think one of the reasons they approach it is because I learnt about this trick.

A note of warning: don't compress willy-nilly. I try to compress things IF they need it. I'll get it wrong, but in the main I find a little gentle compression in the early part of proceedings goes a long way.

For example, if I'm mixing three acoustics together to produce a stereo acoustic part that is going to be the "carpet" in-and-around everything else, I'm listening for a nice carpet with appropriate detail shining through. Ideally that involves no compression at all, but often it means the part that supplies the detail has to have a) the bottom taken off in comparison to the others, and b) it needs to be compressed a bit so that it can be turned up enough so that it will compete with the rest of the band. If you turn it up without compression, then the acoustic mix will still sound good, even fabulous, on its own. But when you mix it with the band, the detail guitar seems to stick out too far and interferes with the vocal. So you turn the acoustic sub-mix down, but then the other two acoustics disappear.... So you remix the acoustic sub-mix, and then you can't hear the detail enough!!!... SOLUTION: compress that detail guitar during the mix of the accoustic sub-mix.

Always try to keep track of what compression got applied at each stage though - each time you compress, you're squashing more of the original performance out of the part. This can be a good thing if you played the original part like a chimpanzee with boxing gloves on, but more often it's a degradation that you probably don't want.

And one last thing about compression, I'm not sure we've said it anywhere in this thread. It's an effect, a tool. You can use it in two basic ways - to produce a specific "effect" or to "improve" the percieved performance. If you want the "effect", eg a vicious pumping bass part, use a lot. If you want the "improve", use it sparingly...

And like Henwrench says - use your ears!!

[Wartime Novelty]

I try to use compression as little as possible on everything.

Mainly due to the fact that if you've got it all sounding good but the final product is a little quiet after normalising i'd prefer to turn the speakers up and keep the airiness of the track than compress it and not have to turn up.

However i will use slight compression if its neccesary mainly on vocals, snare drums and clean guitars.

other than that i do try to avoid it to keep the life in the track.

I also far prefer Limiters to compressors as i find them much more natural sounding

[stoman]

Here is another quick little contribution that may or may not be of value for some of you:


HOW TO BEEF UP A MIX

My mixes sound weak and quiet. Why?

First of all: A mix should have a lot of both dynamic range and head room. Making it "loud" is part of the mastering process. So if you don't already do that, please do the mixing and the mastering in separate steps (and store them as separate projects)!

However, there still are a couple of dynamic processings you have to get right in the mix to get a good balance.

The most important thing is to get rid of peaks. E.g. if a rhythm guitar tracks plays at an average level of -12 dBFS but has some sporadic peaks at up to 0 dBFS, you can never get that track to be louder than -12 dBFS on average. Remember that everything exceeding 0 dbFS will be clipped when rendered down to a 16 bit wav file (as used on CDs).

How do we get rid of the peaks then?

If you use a DAW, or if you have an audio editor like Audacity, take a look at the wave form and find out where the peaks are and what the average level is. If you only have analog equipment, look at the VU meter and identify the peaks that way. Then use a compressor and set its threshold to the desired average level. Adjust the ratio according to the length of the peak. If the peak is very long, use two compressors with modest settings in a row instead of setting the threshold and ratio extremely low! Use a short attack and release on the compressors as you want them to catch the peaks, but nothing else.

In some situations it may be reasonable to edit the audio file and smoothen the peaks manually instead, as a compressor always adds a little bit of color/distortion to your track. In a DAW you can also use a lookahead limiter to tame the peaks. But please don't use a limiter for anything else in your mix!

Another way to deal with peaks in a DAW is using volume automation. Many DAWs allow you to draw automation curves for most settings (like volume, EQ band frequency, panning, bypass, ...).

And there we have another candidate for beefing up our mixes: Most homerecorded mixes have a consistent volume level on all tracks throughout the whole song. That's boring! Use automation to feature different elements of the song in different parts. E.g. bring up the lead vocals in the verses to make them sound more intimate, or make them build up (along with other elements) throughout the song. Maybe turn down the volume of the bridge, etc. ...

And the third thing to pay attention to is the frequency balance of all instruments. Along the time axis there is always one instrument (including vocals) playing the dominant part. That instrument must be featured, i.e. its characteristic frequencies must be clearly heard. If there are other instruments with similar characteristic frequencies, turn them down, pan them differently, and reduce the disturbing frequencies with an EQ!

These are, IMO, the most important aspects of creating a solid, balanced mix that can easily be mastered (and thus be made as loud as you want it).

More on mastering later.

Regards,
  Steffen

[j.g.]

This is a brilliant course in recording technics. Although an old thread, most of this is applicable today with all the modern equipment now available. Thank you for all of you for taking the time and trouble to add your personal tips, hints and information.

[drumbum]

Thanks J.G. This is all great stuff. There is fantastic info here.  Look forward to more recording..!  db

[AndyR]

Just typed all this in my latest song in answer to a query from Ferryman.

Decided it a copy of it here was a good idea.

Blimey, you should write a book on writing, recording and producing music..... Seriously, could you pm me some of the lessons you learned about limiting if you get time?

I just listened again on my new monitors. They're not fully burned in yet but it sounds even better than last time!

The quickest way is to share where I got them from...

Have you come across this guy on youtube?

Produce Like a Pro

His name's Warren Huart. He's an English musician/producer based in LA.

I subscribed several years ago because he had an in-depth demo/review of the Yamaha THR100-HD amp (which is what I use, don't remember whether I already owned it by then - but he would have sold it to me!).

He's really enthusiastic and keen that anyone can learn. I fancy he might rub some folks up the wrong way, but I quite like his delivery/personality - he's sat in his own studio, it feels like he'd make sure you had a cup of tea etc etc and would let you fiddle with the knobs and faders on his kit.


Anyway, didn't watch him too often back then - he generates too much content for me to keep up with!

Then a few weeks back, he posted this:
How to get LOUD MIXES!
... just when I wanted to know

From there, I wanted to know "what on earth is parallel compression?"
And he posted this a few days later:
Secret Digital Mixing Trick for Analog Sound - Parallel Compression
(having cleaned the unwanted transients, you're highlighting the performance's transients and bringing them back in using a copy of the track and heavy compression - a technique I sometimes use on vocals is a variation on this... it's also called New York Compression... but this guy demos it and actually makes it happen in front of you).

I had to translate his DAW work to the BR1600 (although, he's almost convinced me to switch, and it would be much quicker that way), but flip!! Much of this was used on Practical Gardener... Mrs R had said "you're putting loads more effort into this one, is it special?" so I showed her the videos and explained what the techniques were - she started getting interested too.
For example, I've been terrified of over compressing or limiting the lead vocal... no more! That's the first thing I do once I've got the take. Then I can add parallel compression to bring out the juicy bits.

I'm not going to do this, but I am tempted to go back to Edith Finally Gets It and remix... that could get seriously WOAH!! if you did this to it.

I've spotted other videos I need to watch:
How to clean up a Muddy Mix
8 Tips for Amazing Low End

Sounds like the sort of thing a chap needs to know, eh?

Check out the descriptions in the videos - he describes everything, links everything.

Just noticed, he also has an "academy" you can join via his own website.

[Greeny]

Cool! Thanks.

Update: I knew the name was familiar. He appeared on the Andertons TV channel I watch regularly.

[Frank53]

Just bumping this thread 'cause it has some excellent information. There is so much to learn in this field, and I have but scratched the surface. If I may, let me contribute a little trick I use for acoustic guitar (though it may be very common).

Recording an acoustic guitar part twice, and panning left and right works well to add depth, and keep them from competing with centered vocals, but since there will invariably be some subtle differences in how the part is played each time, it will always sound like 2 guitars. This is what I do to get a fuller sound from one acoustic guitar.

Record the part once, in mono, using whatever method you like. Copy the entire track to a second track. If you pan the 2 hard left and right, it will essentially sound like 1 guitar, centered. That doesn't help. Now, take the copy and delay it by about 15ms. The brain can't quantify the delay, but it gives the guitar a fuller sound, and opens up the middle of the field.

I will use this for e.guitar in a mix as well. Whatever reverb you use on the guitar, apply at least twice as much to the copy, and reduce the level by 3 to 6dB. It sounds like the copy almost disappears, but when you pull the fader for the copy all the way down, there is a stark difference.

I'll play with this, depending on the song. Panning the copy toward center (if they are the same level) will move the image toward the other side. A shelving cut below about 200Hz can contribute to giving the copy a more distant sound, when its level is attenuated.

Just something that seemed to work for me. Carry on...