Skip to main content

I have a question regarding getting a stereo acoustic guitar track out with one mic. Up until now i've been doing most acoustic guitar direct in with no real problems, other than a thin sound. What I ususally do is duplicate the track bump the new track up a couple ms and hard pan left and right. Last night i set up a RODE k2 (that i just got) about 20-25cm from the 12th fret looking at where the neck and body meet. I tried my usual trick and found that there was defenitly some phase issues and comb filtering when doing this. .adjusting the amount the 2nd track was offset drastically changed the sound. So I tried something else. I duplicated the track and ran the new one through a pitch shifter and detuned it about 17 cents. .seemed more full sounding than the other method but too chorusy. .also the stereo image seemed to float left and right. I've alwas noticed things like this before but never to this extreme. .i guess it's the new mic picking up frequencies I'm not used to hearing. .

Is there another method other than just playing the part twice that would minimize these problems. ?

I've heard a little about the 3:1 rule. does this apply to offseting 2 identical tracks. .(ei, i noticed that at 33.33 ms the sound didn't change much but the timing just isn't tight enough. .

also when i say ms, i might mean samples I'm not sure off hand what the offset slider in nuendo is calibrated in.

any help would be, eh. helpful 8)

Comments

David French Mon, 01/24/2005 - 19:03

Getting stereo by the method you're using is IMO not a good idea. When you combine a sound with a delayed copy of itself, comb filtering will always result. The first notch frequency will be 1/2*t Hz where t is the delay time in seconds. Each additional notch will be at 3, 5, 7, ... times the frequency of the first notch. After about 35 milliseconds of delay, your brain no longer integrates the two sounds as one and you start to hear a distinct echo instead of comb filtering.

The only good way to get a stereo guitar sound is to record using two mics. You said you have a pickup on the guitar. Can you record with the mic and the pickup at the same time?

Playing it twice, one with one set of voicings and one with another, can sound very nice.

The 3:1 rule is about multi mic placement, not delay times.

The slider in Nuendo must be using ms, because a coupe of samples is an extremely short ammount of time.

imagineaudio Mon, 01/24/2005 - 19:47

David French wrote: Getting stereo by the method you're using is IMO not a good idea. When you combine a sound with a delayed copy of itself, comb filtering will always result. The first notch frequency will be 1/2*t Hz where t is the delay time in seconds. Each additional notch will be at 3, 5, 7, ... times the frequency of the first notch. After about 35 milliseconds of delay, your brain no longer integrates the two sounds as one and you start to hear a distinct echo instead of comb filtering..

What would happen if you did this and then detuned the guitar on the second track, you would no longer have 2 identical waves, right? would this minimize the comb filtering?

David French wrote: The only good way to get a stereo guitar sound is to record using two mics. You said you have a pickup on the guitar. Can you record with the mic and the pickup at the same time?.

I used to do this, however, the pickup is all f*'d now......there is a plug that goes from the pickup to the electronics that some how has come unplugged, and the ring that holds the jack in place has come loose leaving the jack buried inside the electronics, and i can't find a way in......but thats another subject

David French wrote: The 3:1 rule is about multi mic placement, not delay times..

thats what i figured but i had to ask

David French wrote: The slider in Nuendo must be using ms, because a coupe of samples is an extremely short ammount of time.

Yeah, im going to agree with you here, too

I guess i could run one track through a slight delay and the other through a chorus effect, this should help, eh?

took-the-red-pill Mon, 01/24/2005 - 20:00

I'd have to agree with the learned one.

Back in the day when I thought a comb filter was a hair care product, I wanted to double my acoustic. I placed one where you did, near the 12th fret, and then another about 3' out, more of an overall sound. The close one was a decent condenser. The second one was an SM57 because that's all I had. I remember treating them differently on compression, but I can't remember how.

I then played the guitar twice through, for 4 tracks, and panned hard left and right. so I had:

Left speaker:
condenser, 1st pass
SM57, 2nd pass(quieter)

Right speaker
condenser, 2nd pass
SM57, 1st pass(quieter)

Well, I must have fluked it off, because there didn't seem to be any bad stuff going on in the way of phase issues. It sounded very rich and full. Even a crappy mic(not that a 57 is crappy) run into a second channel ends up sounding better than 1 mic and 'created' effects, In my uneducated opinion.

If you held a gun to my head and told me I had to choose between a)recording it once and trying effects later, b) recording it in stereo with 2 mics, or c) recording the guitar with one mic, 2 times through, I'd definitely take the 3rd option. It seems to sound the best to my ears. It also leaves you with the most options. You could change the mic, the mic position, the guitar, the room, the compression, the player, your sock colour...you get the picture. And you've still only sucked up 2 tracks if you're limited in that way.

happy overdubbing
Keith

David French Mon, 01/24/2005 - 20:29

imagineaudio,

The more you do to the copy, or to either sound for that matter, the less correlated they will be and the less comb filtering will occur. Try everything you can think of that still sounds good.

Keith,

SOunds like you separated the potential offending pairs in the stereo field. I wonder what happens when you go to mono? As for 'a', 'b', or 'c', I wouldn't write off 'b'. A carefully spaced pair can sound wonderfully wide and exciting too.

anonymous Mon, 02/07/2005 - 06:03

Delay Calibration

Sliders in Neudo and Cubase are not Calibrated in ms.
they are calibrated in 10 th of milliseconds:
i.e.:

0.35 = 35ms,
so this is point there all your problems come from.
Haas effect is not about noticing delay then it comes to 35ms, but noticing delay then raising volume up to critical(delayed channel) (there is graph, but use your ears)
no comb filtering occur while (5ms - 100ms panned to oppoite).
comb occur only in mono.
You can achieve best minimal comb if you will try different delay times (around 10ms)
note thad delayed channel MUST be somthing about -10db lower,
usually. so comb filtering is minimal.
use panpot of original channel to move stereo image, not try to compensate with delayed channel.