Recording and Production tips

[Geir]

This is a collection of tips and tricks that I've learned the last couple of years after I started recording and producing my own music.
Examples I use is from recording with the Boss MicroBR (mbr) if nothing else is specified. I don't claim to be an expert on these things, but would like to share my experince and get some feedback and tips from you guys and gals to further improve the process of getting the music to sound as good as possible.

Recording vocals
Vocals are best recorded dry. Turn off the effects, or use just a limiter or a soft compressor. This will maybe result in one extra bounce during the production phase, but will give you much more options. If you need the effects to deliver your optimal vocal performance, be sure to record the dry signal anyway (mbr EFFECTS -  - LOC - +  INPUT REC DRY ).  Add the effect later.

Distance to mic should be kept at a minimum! The closer you get the fuller sound. If your deliverance is very dynamic on a song, compensate by withdrawing from the mic when you sing louder. This is something that's worth practicing! Using the mbr I hold it in the palm of my hand, stretch out the thumb and touch my lips as a starting position. That also gives an angle to the mic so I don't breath/spit directly into it !

Recording  acoustic guitar
Guitars are best recorded dry (see recording vocals).

Distance to mic. I find a distance of 20-30 cm from 17th fret of the guitarneck is a good starting position. If you want more bass move mic closer to soundhole and closer to 12th fret for less bass. Now, more bass and thus a fuller sound can be tempting BUT consider what you're gonna use the recording for. If it's in combination with several other instruments with a similar or lower frequency range, keep the bass low to moderat. High frequensies are easier to handle in the mix by panning them out, but bassy instruments are hard to seperate.



Mixing
There's a LOT to be said about mixing, I'm just scratching the surface here, but for me one of the most important things in the mixing process is to give everything it's own place in the mix, where it can be heard. One simple but illustrating example to start with:

You want to record a song with just acoustic guitar and a single (male) vocal track. The challenge is that often the vocals and the guitar will dwell in the same frequency range. One simple way to separate the two is to pan them in each channel, the extreme version resulting in a early Beatles kinda mix. Tho I love the Beatles, I hate this kind of mix, especially when listening with headphones.So as a rule I record my guitar twice and pan the two takes hard L/R. If recording live, with only one mic, I make sure the vocals get louder than the guitar by placing the mic closer to the mouth than the guitar. That also takes some of the bottom of the guitar which also benefits the mix.

------------------

I'll  post a followup later on mixing/mastering a "full band" recording, if anyone is interested.

[Tony W]

I'm interested. The proof is in the pudding. Your production has gone through the roof lately.

[Wartime Novelty]

I'm also interested i havent been on ina  while and this would be something great to come back to

[AndyR]

Interesting stuff... nice one Geir.

While I'm waiting for the wife to tell me dinner's ready, I'll dump some of my secrets in response - hope that's ok?

Usually, I record EVERYTHING dry. If the guitar amp has reverb, turn it off. On the synth/piano/organ, turn the reverb off. If I'm using an external amp modellor, I put minimal reverb on (in the modellor) to make it sound like the amp was mic'd in a real room. It is so much easier to control everything later if you haven't imprinted a "final mix" reverb on the tracks themselves. First of all, it means you can add appropriate reverbs when you know what they need to be, second, you can do drop-ins and cut and paste if the recording has no reverb tails on it.

The only exception to this is if I am using an electric guitar effect (in my outboard modellor, Vox Tonelab) that uses Delay - I'll record that happily, especially if it's an effect the Boss can't do (specific types of stereo panning), but mainly because the Delay in the Tonelab has "Ducking" which the Boss delays don't (ducking is when the repeats are automatically lowered in volume as the effect registers a new note coming in).

I run compression/limiting and gating (noise reduction) on all vocals as I'm recording - the compression/limiting stops nasty peaks, the gating removes the neighbours, the wife and the sound of my shirt rustling, etc, from the recording.

I also compress/limit and gate acoustic guitars if it seems necessary.

All other instruments - there are no effects on when I record, unless it's part of the instrument (eg chorus on the electric guitar and it's amp, or on the organ, or whatever).

I've just realised, if you're using the BR's amp modelling (I don't except for experimental stuff) you have a choice of recording plain signal or effected. In this case, I always record the "amp" effect, I can't be bothered with messing around later choosing an amp sound for the dry electric! But, I DON'T record any reverb.

Years ago someone told me "Unless you're a complete genius, Reverb is for mixing, nothing else... and if you are a genius, you'd still be wrong to record it during tracking...". I only tried to disobey this once, and I had to re-record everything.

====

I've got a few more minutes... I'll amplify on what Geir said about choosing a bassier sound or not for an acoustic.

It actually applies to every instrument, including the vocals. Try to get the recorded sound as close to what you think you'll need during mixing. Do this at source if possible (ie the instrument or it's amplifier), or by moving the mic around like Geir says. Guessing what you need only comes with experience, but it does come quite quickly - record and mix 2 or 3 songs and you'll start learning easily. The reason for doing this is so that minimal EQ is needed on tracks during mix down.

It's months since I used the MBR - but I think this is almost essential on an MBR. There's no track EQ on play-back is there? You can go through an effect while mixing, but this is really time-consuming if you need to EQ instruments differently. I think the MBR actually taught me to get the recorded track right in the first place. On the BR1600, I've got parametric EQ on all tracks during playback, but even then, you want to use it as little as possible. Get natural sounds, mix them naturally, they almost sound natural at the other end!

But one thing I will add to this - any EQ you do at mixing, you always try to CUT stuff, not turn it up. If you have to turn the bass up on the acoustic guitar in the mix, then it sounds flubbery. If you have to turn the bass down on the acoustic in the mix, it sounds reasonably natural. SO, when recording the track, remember you can take stuff like bass out later (as long as it's not too much), but putting in what wasn't there in the first place makes messy mixes. Same applies to mids and hi frequencies.

=====

Talking of reverb, I've explained this recently, but not on songcrafters I think. If you use reverb (some don't), think about using DIFFERENT reverbs on different instruments in a mix. You're a bit limited on the MBR, but there's enough variation to do it. Vary the type (Hall or Room), the delay time, and the tone. Doesn't need to be massive changes, but try doing this:

Lead Vocal and Lead Guitar (or whatever the lead instrument is) use the same reverb to keep them in the same place in the song.

Backing vocals have a different reverb.

Guitars have another different reverb.

Keyboards have another.

Drums have another.

- you'll be AMAZED at the difference this makes.

On a BR (MBR to BR1600) this means you have to create stereo sub-mixes and then mix them together. EG. Do the guitars down to stereo, then the keyboards. Then mix those two sub-mixes down to stereo (remember to turn the reverb off this time!) etc, etc. It's very time-consuming doing it, and needs a lot of planning on an MBR, but it is SO worth it...

Dinner's ready, I'm off... sorry for any typos - I'll check it tomorrow!

EDIT: One last thing - EQ during mixing, I always take ALL of the bottom end off of all electric guitar parts (except the lead, maybe, if there is one). Say 50Hz, mebbe 100Hz, all the way -20db,  take it all off. This has a fantastic affect on your mix. Bass guitar and kick drum have room, the bottom end of keyboards have room, evrything feels so much warmer, more natural, and LOUDER!

[danieldesete]

please keep them coming this is just what  i'm looking for.

Even If I don't know what that means
 "I always take ALL of the bottom end off of all electric guitar parts (except the lead, maybe, if there is one). Say 50Hz, mebbe 100Hz, all the way -20db,  take it all off."
daniel

[OsCKilO]

What does "mbr EFFECTS -  - LOC - +  INPUT REC DRY" mean?

[Geir]

While I'm waiting for the wife to tell me dinner's ready, I'll dump some of my secrets in response - hope that's ok?

That is more than ok !! I was kinda hoping you'd jump in on this thread and share your secrets!!

"I always take ALL of the bottom end off of all electric guitar parts (except the lead, maybe, if there is one). Say 50Hz, mebbe 100Hz, all the way -20db,  take it all off."

It just means that you damp the lowest bass sounds. It does clean up the mix as it gives room for the bass/bassdrum in the frequencey-range they naturally belong. A bassy guitar can hide a good bassline.

What does "mbr EFFECTS -  - LOC - +  INPUT REC DRY" mean?

If you choose that input-method, you will hear the effect when playing, but the mbr will record the dry input signal. That means you can add it later, when bouncing, by setting the LOCation of the desired effect to the track you recorded your guitar to. In that way you can experiment with different amps after you've recorded that perfect take !!

[Geir]

Andy, like you I do prefer recording everything dry.and I've tried a couple of times recording the plain guitar-signal and adding amp after, but now just tweak out the reverb/delay from the patches and record the guitar as I hear it. I also choose patched with much midtone or tweak the effects cutting off some bass.

When it comes to mixing a "full band" I mainly use two principals as guidlines, both aimed to give every instrument/voice it's own place in the mix and getting a good stereo balance.

Panning.I normally record every rhythm guitar twice. Sometimes I let them be a bit different, either in rhythmic pattern or the way I take the chords and sometimes I just record the same thing twice. The more alike the two takes are the more extreme I pan them. On "The stairs are falling" I panned full left/right for the two guitars. If I use a capo for one take and not the other I will typically pan ~20 L/R. This also has to do with my other principel: frequency balance. More on that later.

Bass always go in the center and with NO reverb !!

As I only use the drums from the BR's and havent yet done any advanced drumprogramming they are just stuck where the BR puts them (I never pan the drumtrack)

Lead vocals will in most cases end up in center. In cases where I double the vocals (by singing it twice) I can pan them up to 15-20L/R if they are very similar, or I can put both in center and put one of them a bit lower in the mix with a touch of reverb.

Lead guitar in center, but I occationally use the panned delay or some other stereo effect to give it a bigger sound.

Harmony-vocals. When mixing in the harmonyvocals I listen to what I've got and find where there's room for them. If guitars are panned full L/R I usually put the harmonyvox 20-30L/R if I have 2 or more. If I only have one harmony I place it closer to center. 5-8 left or right and maybe have the main vocals 3-5 the other way. It also depends how close the harmony is to the melody. If "far away" I can put both in center. .... which leads to my other principal:

Frequency.When mixing I try to avoid that two sources with the same frequency-range is put at the same place in the stereo image. I also try balancing out one source with a given frequency-range in the left with another source with similar range in the right. Let's say I've put a guitar in left channel (30L), then I put another guitar or a piano in the right (30R).  A harmony-vox in one channel can be balanced with another harmony, an organ or strings in the other.
On many of my acoustic songs I record 4 guitar-tracks. I play twice without a capo and twice with (at 5th or 7th fret). then I place one take without and one with capo at 30L and the other two takes 30R.  

To summarize, I try not placing two audiosources with the same frequency-range at the same place in the stereomix. Audio-sources with a good distance in the frequency-range can be put at the same place in the mix.

One last thing for now. If I want something to sound a bit "distant" I put it low in the mix and with a good doze of reverb. It must then have a well-defined space in the stereo-mix without any other sources competing for the attention.

I made a quick drawing to illustrate a typical mix for my songs (the "clouds" symolises reverb and how that will fill out the sonic space):

[dwardzala]

Two other tricks that I use when recording guitars:

On electrics when mic'ing an amp, split you signal out of the guitar and send one signal to the amp and the other signal DI into your recorder.  You have the option of "re-amping" the take.

When recording acoustic guitars, if you have pick-up system, you can do the same thing.  Mic the guitar and record the mic'd signal DI.  Often mixing both will give a better/fuller sound.

Also, don't eq tracks on their own.  EQ them in the mix.  Often a solo'd track that has been EQ'd sounds terrible, but in the mix it fits in its own little pocket.

[AndyR]

Also, don't eq tracks on their own.  EQ them in the mix.  Often a solo'd track that has been EQ'd sounds terrible, but in the mix it fits in its own little pocket.

+1

That links in, a little, with my "take all the bottom end off the guitars".

I've worked in situations where an engineer has asked me to turn the bass down on my amplifier. As a guitarist, I'm sure that's WRONG!!!

I want the guitar to sound and behave like I want it to. The tone controls (or most of them) on a valve amp are part of the pre-amp and together with your guitar, pedals, fingers, etc, shape the sound/voice you are making.

What the engineer actually needs to do is to remove all the lower end frequencies from the guitar signal he's got in the mix. I'd even recommend trying this if you're doing 7/8-string drop-tuned metal stuff. If you listen to the guitar part on its own, taking all the bass frequencies off does make it sound a little bit "that's not my guitar sound ma-an!!". BUT turn the rest of the band up again, and it does sound like your guitar again. Leave the bass frequncies on the guitar parts and the guitarist will think he's happy, but he'll mutter that "the mix sounds sh1t, that engineer was useless, he made the drums and bass really weedy..." - you're back to the "everything louder than everything else" struggle in the mix.

=====

Love that picture Geir - while I was reading the post, I was thinking "I do this visually in my head, what we need is a picture..." ... and there it is! It's not the same picture I use, but it's very close

I've got to go now, but before I forget - shall I mention "complementary EQ"? It's an extra weapon you can use when you've run out of spaces in the picture. If someone knows what I mean and explains it, cool... But otherwise I'll write some waffle next time I'm on (tomorrow I hope)