Question about doubling tracks when recording:

[Mapleg4]

When you guys record guitars on a computer and you are to record 2-3 tracks of rhythm guitar for a single song, do you set up 2 or 3 layers/tracks/channels and record the riffs on them simultaniously, or do you record 2-3 tracks of the same riffs after one another? :icon_scratch:

 

[swarfrat]

"Doubling" would be playing the same part - whether you play it again, or cut/paste/delay/pitch shift, etc.. Recording different parts would be "layering". Both are done, but of late I myself have been paying attention to really simple parts and how they support the song, and trying not to create stuff that can't be done live without looking like a Southern Rock band onstage. (I love southern rock, I really do. 38 Special is still one of my favs even though they don't seem to get no respect. But one of the hallmarks of southern rock appears to be having 4 guitars onstage, plus bass/drums/keyboards, and maybe even a fiddle/banjo/mandolin plus a harmonica, plus whatever else they can think of. It can get incredibly busy.)

 

[dmraco]

I do at times for vocals.  I will have one track perfectly clean, no effects, the other I will add delay, verb and tweek the EQ.


FOr guitars, I will often have both a six and twelve for acoustic stuff.  For electric I ususally only have one.

I read Randy Rhodes would tripple is guitar solos!!! arty07: arty07: arty07:
I consider myself lucky to get it correct once!

 

[AndyG]

"Doubling" would be playing the same part - whether you play it again, or cut/paste/delay/pitch shift,

Actually, in order to get the true "doubled" sound, you have to play the same part again on a different track.  Any kind of processing will give you a faux-doubled effect, like chorus.

Doubling isn't necessary for everything, but it will give you a thick, rich tone that most people find pleasing.  Depending on the style, what I have done in the past is simply double the part, then pan hard left/right.  To go over-the-top heavy, play the part 4 times, and each pair gets panned left and right.  This works particularly well in bands that have 2 guitar players, or if you change your tone slightly (different guitar/amp/whatever) between takes 1&2 and takes 3&4.

For bass, I always record 2 tracks (DI and amp), but only 1 take.  Doubled bass can really mess up your "bottom end".

Doubling vocals was REALLY common in the 70's and 80's.  Not so much anymore, but you will sometimes hear it, for example, in the chorus of the song rather than the verse.

The really nice thing about DAW recording is that you can record as many tracks as you like, try all sorts of ideas, then sift through it and decide what you want to keep in the mix.  Just because you recorded it doesn't mean you have to keep it!  Back in the days of, say, the Beatles, they had to pre-mix, or bounce parts together in order to free up tracks to record more stuff.  That forced those artists to make decisions as they went along.  But there was a lot of great music recorded at that time, so maybe it wasn't necessarily a bad thing    :glasses9:

 

[Doughboy]

I never quite understood the reason to physically double a guitar part. Why not bounce the same track on to a 2nd track, pan them hard left & right & maybe put a 10ms delay on one track?

Why would anyone want to spend hours physically doubling or quadrupling a part when then can just bounce or cut & paste it onto as many tracks as they want to?

 

[AndyG]

I never quite understood the reason to physically double a guitar part. Why not bounce the same track on to a 2nd track, pan them hard left & right & maybe put a 10ms delay on one track?

Why would anyone want to spend hours physically doubling or quadrupling a part when then can just bounce or cut & paste it onto as many tracks as they want to?

The reason doubling sounds the way it does is BECAUSE the part was played twice .... and the variences between the 2 takes will be different throughout.  If you copy and paste a track, and pan left and right, it will still be mono.  If you add the delay, it will either phase against itself in mono, or it will simply sound delayed.

What gives doubling its unique quality is the infinite randomness between playing a track twice .... we're talking variations of microseconds and microtones.  There are doubling plug-ins that approximate this randomness, but for the most part they just end up sounding like chorus.

 

[stubhead]

I agree with the above, though I have had enough bad experience with tuning differences that as a practical matter, I would only double a track with the same guitar, especially on acoustics. Change the tone around if you want, change the mike, but same guitar. What a lot of people don't realize is that all guitars and keyboards by their nature are slightly out of tune, to accommodate the "tempering" necessary to play in all the different keys. And this seems to get magnified by using different guitars to play the exact same part - and if you don't know why, it can get really aggravating. The best two-guitar harmony bands like the 1st edition Allman Brothers and certain lineups of Thin Lizzy spent weeks rehearsing parts so that the guitar players knew each other's bends and rhythms.

http://en.wikipedia.org/wiki/Equal_temperament

http://en.wikipedia.org/wiki/Just_intonation

I've noticed that a lot of the very best guitarists in the world play a very few number of guitars at any given time, and learn one guitar's ideosyncrasies well. When I read about a CD where the producer and artist shipped in 40 guitars and 30 amps I kinda know what it'll sound like...  :sad1: Economy in overdubbing is a great virtue. Unless you're as good as Clapton & Allman or Jimi Hendrix, it's better NOT to try quadrupled harmony leads.

 

[Mapleg4]

I never quite understood the reason to physically double a guitar part. Why not bounce the same track on to a 2nd track, pan them hard left & right & maybe put a 10ms delay on one track?

Why would anyone want to spend hours physically doubling or quadrupling a part when then can just bounce or cut & paste it onto as many tracks as they want to?

That's what I try to do. I'm working on some demos (with Line 6 Pod Farm) that also have some orchestra samples on them, as well as clean lead guitar on top of distorted guitar, and because of that, I need 3 guitar tracks, one in the middle and one to each side. It still sounds like one guitar and I actually like the effect I just got from duplicating one track. It still sounds like one guitar and actually simple in a way, but it sounds really fat.

 

[AndyG]

That's what I try to do. I'm working on some demos (with Line 6 Pod Farm) that also have some orchestra samples on them, as well as clean lead guitar on top of distorted guitar, and because of that, I need 3 guitar tracks, one in the middle and one to each side. It still sounds like one guitar and I actually like the effect I just got from duplicating one track. It still sounds like one guitar and actually simple in a way, but it sounds really fat.

From a straight physics point of view, if you copy and paste a track, all you are doing is increasing its volume by 3db, regardless of which way they are panned.  A clean guitar and dirty guitar will obviously add a nice texture, even if it is the same take, because the difference in tones will fill things up.
Personally, if I am going to leave a guitar in the middle (which I almost never do, 'cause that's where I put the vocal), it will be a single take.  I like putting the instruments on a virtual "sound stage", where, when wearing headphones, you can alomst picture where each player is standing.
Again, when recording, there is no right or wrong, as long as you get the desired results.  Even if you copy/paste a track and add delay .... that's not wrong if you like the result .... its just that the result won't sound like double tracked guitars.

 

[OzziePete]

That's what I try to do. I'm working on some demos (with Line 6 Pod Farm) that also have some orchestra samples on them, as well as clean lead guitar on top of distorted guitar, and because of that, I need 3 guitar tracks, one in the middle and one to each side. It still sounds like one guitar and I actually like the effect I just got from duplicating one track. It still sounds like one guitar and actually simple in a way, but it sounds really fat.

From a straight physics point of view, if you copy and paste a track, all you are doing is increasing its volume by 3db, regardless of which way they are panned.  A clean guitar and dirty guitar will obviously add a nice texture, even if it is the same take, because the difference in tones will fill things up.
Personally, if I am going to leave a guitar in the middle (which I almost never do, 'cause that's where I put the vocal), it will be a single take.  I like putting the instruments on a virtual "sound stage", where, when wearing headphones, you can alomst picture where each player is standing.
Again, when recording, there is no right or wrong, as long as you get the desired results.  Even if you copy/paste a track and add delay .... that's not wrong if you like the result .... its just that the result won't sound like double tracked guitars.

I reckon this is invaluable to remember when doing your own original work......If you start working to strict 'right or wrong' principles you begin to go down the track that typified EMI in their pre-Beatles era. White coats, Business shirt and tie,  lots of restrictions etc. because "we do not do that here"..

 

[stubhead]

If it's just fat and/or controllable effects levels of effects - in the mix, - then copy->paste works great. You have a clean tone, usually kind of "flat" which really means "scooped mids" because in the post-1950's history of electric guitar, most of it has been through systems that had a huge emphasis on midrange. Play your guitar through a PA with tweeters or horns, you'll see... :evil4: And, on the other tone you pile in some overdrive, and cut the screeching highs and mud-filled lows. It's actually a really good idea to experiment EQ'ing the signal for bass and treble cuts before it hits an overdrive - you want to try all the variables in signal path, eventually.

From my reading of interviews, I'd guess that over half of touring guitarists now used a bi-amped system live. If my memory serves me, I believe it was Steve Morse who began playing through two (or three) amplifiers on all the time, then it quickly spread to his bud Eric Johnson, who passed it on to his bud Alex Lifeson, then BLOOEY! all over the place. The usual setup is to have a Fender Twin or such to produce clean highs and lows, and a Marshall-y amp set for midrange overdrive. Overdrive sounds best in there, and if you try to overdrive a clean amp, the bass frequencies load up a disproportionate amount of the signal. Which is why rack compressors have a hi-pass filter, to keep the bass spectrum from hogging all the juice. No two-channel amp can do this, it takes two amps. I believe the latest PODs will let you run two separate paths. I used to use a rack system with a split signal, clean and dirty with different EQ curves, a poor man's way to duplicate this.

So anyway, yes, copy another track of a solo, then pull it out, cut the lows and highs, then dirty it up and add it back together. See whether you like it, try it some more. There's some real fun things you can do with your clean, "stringy" sound up loud and a super-cranked out pile-driving fuzz-sounding tone mixed back in - but very faintly. It adds, like, FUR to the BEAR - but without taking over. I love mixes that add faint, odd, nearly-hallucinated stuff mixed way back - "Paul IS dead...." :toothy12:

 

[AndyG]

Its funny ... there is such a fine line between "rules" and "established practices".

When you are working to record someone else, and they say "I want to sound like .........", it helps to know the established practices to know how to get what your client wants.  The joy of recording at home is it gives you the opportunity to experiment, and try things that develop a "sound" and "style" to your recordings, and ultimately, to your songs.

I've said it before, that if you give the same tracks to 10 different mix engineers, you'll get 10 different mixes, none of which will be "right" or "wrong", but your ear will find a favorite.  There is no rhyme or reason why, it just is.  That is the beauty of the art of recording.  Having said that, knowing what the "established practices" are will go a long way to getting clients to want to work with you in the future.  But there is certainly nothing wrong with developing your own little trade secrets to get the sound only you know how to get !  :glasses9:

 

[mayfly]

If I want a heavier sound, I always double the guitar tracks.  Then I pan them hard left and hard right and I have a huge sound.

This technique is also pretty fun when you change things in the tracks just a little - like starting a line on one guitar and finishing it on another.  or trading back and forth in just a very minor way.  Really makes the track special.

 

[Superlizard]

I never quite understood the reason to physically double a guitar part. Why not bounce the same track on to a 2nd track, pan them hard left & right & maybe put a 10ms delay on one track?

Why would anyone want to spend hours physically doubling or quadrupling a part when then can just bounce or cut & paste it onto as many tracks as they want to?

The reason doubling sounds the way it does is BECAUSE the part was played twice .... and the variences between the 2 takes will be different throughout.  If you copy and paste a track, and pan left and right, it will still be mono.  If you add the delay, it will either phase against itself in mono, or it will simply sound delayed.

What gives doubling its unique quality is the infinite randomness between playing a track twice .... we're talking variations of microseconds and microtones.  There are doubling plug-ins that approximate this randomness, but for the most part they just end up sounding like chorus.

^ what this guys sez - "manual" doubling is by far the coolest sounding.

 

[Patrick from Davis]

This is kind of where a re amp (Reverse DI) helps.  While I do not suggest using the re amp for quality tracks, it really helps out when doubling by making it easier to find the desired amp sounds/combinations.  It is usually pretty easy to get a "big" sound by using one amp that has a nice mid heavy sound, and one that has the clarity of top and bottom to fill the rest out.  Then you can futz with mixing levels and get the desired sound.  The drawback of doubling (if used excessively, like tripling, quadrupling,...) is that eventually the resolution of what you were playing is lost.  To the point that the amps don't sound right any more.

There are several other outboard ways to make things sound bigger as well, but the recording equipment has to be there.  Apogee's soft limiter seems to be able to get rid of a lot of the digital transients that tape doesn't pick up, so you can keep pushing the levels.  There are other transformer based solutions that do this as well, but then you start to get into making a home studio, not just a room to record in.  Also the mic (really the mic combo) that you use makes a large difference as well.  Good luck
Patrick

 

[rightintheface]

for some heavy riffs or sections I really want to stand out I will double track like this:

one guitar played through orange, double mic'd with a 57 and beyer m81

same guitar played through marshall into the same cab with same mics/placement

So essentially four tracks of guitar. I'll usually keep each part "together" and panned so you can hear the individual takes. It sounds pretty awesome most of the time. But it's a track-dependent thing. sometimes I track a double part with completely different guitar/amp/cab/mic setup. Just depends what you're going for. I have often fallen into the trap of recording loads of guitar parts, thinking it sounds really really good, and it almost always ends up sounding "cluttered" in the mix and I end up ditching all but two of them - the more tracks playing the same riff, the less definition you tend to get. in general anyways.

Experiment! have some fun, there's no right or wrong way to do it  :icon_thumright:

 

[The Norwegian Guy]

I'm using three different tracks for rythm.
One "main" track which has zero pan at all.
Two panned 50-60% (L/R) tracks where I can get creative with f.ex different pitched harmonics and stuff like that.

I'm using different sounds on each track because I think it sounds a bit more awesome. And it helps developing your timing and attack because you have to be dead on, on every single note to make the three tracks sound cool.

 

[TonyFlyingSquirrel]

I record up to 4 rythym guitar parts per side, each with a different amp model, ie; bogner uber, peavey 5150, deizel vh4, and a diesel Herbert.  I don't keep them all, I only keep the best of each side, and end up with 2.  This way I can let the combination of the performance and the character of the amp be the determining factor.

The only other thing I do is sometimes I record in dual amp mode, playing through the uber and Herbert models simultaneously, and I'll mix them together, you have to be very careful when doing so, and back off of the gain and low end considerably so you'll still have clarity.

If I'm recording dual harmony leads, I'll do a take panned left, and a take panned right.  If I'm tracking a solo that will not have any harmonies, rather than recording a stereo track, I'll do 2 individual mono tracks. I've found that most larger studios will request that you submit your project that way, especially if you are going to do an .omf export such as I find myself doing from Sonar on a PC to Logic or PT on a Mac.

 

[TonyFlyingSquirrel]

Another helpful tidbit when using Pods or Podfarm and crunch tones, and I stumbled upon this the hard way, consider using the same cab, and same ambiance (distance) and often, the same mic set up or else you can run into phasing issues just as you would encounter in a real micing setup, even in software.  For this reason, I use the same cab set up in each and every crunch preset, and my results are far more consistent now.
I have found that the tread plate cab with about 15% distant micing with the 421 mic model puts my guitar right where I want it in the mix, still allowing my bass player to get all over his 6 string bass, and have keys in there as well, with plenty of gain, plenty of crunch, but not too much hair that it no longer sounds organic.

 

[elfro89]

AndyG has said it all really.

Panning left and right adding a small delay is fine but that is a totally different sound to having a layered guitar. Copy and pasting with pitch shifting and all that other studio magic to try and achieve the same sound won't work at all. You need to record the same part twice or several times depending on how thick you want it to get it sounding proper, which can be bloody difficult to do exactly spot on. Def Leppard have the guitar layering down perfectly, testament to how great Mutt Lange and the band are, Here is what I mean if you listen to the guitar tracks. Impossible to recreate accurately without layering properly; even if you used a 10 channels from 10 different amps. I think the distorted guitar tracks were layed about 4 times, they are so perfect it's hard to believe you are hearing so many layers.

http://www.youtube.com/watch?v=AsvFK8gWIIQ