reamp delay advice

[anth]

ok so I recorded some guitar tracks for our new EP - recorded a miked guitar amp (royer r-121 roughly 6-8 inches back) and DI signal for each guitar part. Now, I reamped most of these tracks with some different amps (using the royer again roughly 6-8 inches back and a room mic) and the tones are sounding great. However, I've noticed a 2-4 sample delay (visually measuring when zooming right in) between my reamped guitar and the original miked guitar. This seems to affect the groove of the part to my ears. My RME interface is set to lowest buffer and automatic delay compensation (in PT 10) is set on. Some of this delay might be due to the mic position not matching exactly the original position but I'm assuming most is due to the reamp process. I'm happy to try line them up manually however they are slightly different waveforms (peaks and troughs follow the same trend on both waves but the length of them is sometimes different - phase?) so not sure exactly which point on the original wave to line the reamped wave up to. Any advice on how to match them up/what you would do? I don't intend on combining them.. I just want the same feel.

One possible solution could be to record a single guitar strum with a miked amp and di signal and then reamp it with the same mic position and visually measure the difference in samples to give me the reamp delay for my setup? Then just use that preset delay difference for all my reamped tracks. However, timeadjuster in pro tools only allows you to push a track forward ... I need it to be pushed back to remove the delay.

Any suggestions? thanks.

[RoyMatthews]

EDIT. Nevermind. I wasn't paying attention.
_________________
"If there's one ironclad rule of pop history, it's this: The monkey types Hamlet only once."

[chris harris]

at 44.1kHz sample rate, you're getting 44,100 samples PER SECOND. If it's only off by 2 to 4 samples, this will in no way affect the groove. I'm assuming that the track is not at a tempo of 10,000 bpm.

I think either the problem is somehow different from how you're describing it, or it's in your imagination.

I reamp all the time and never have to shift anything around.
_________________
Studio - http://www.hookechosound.com
Label - http://www.wearenicepeople.com
Band - http://www.depthandcurrent.com
Twitter - http://www.twitter.com/HoodEchoSound

[T-rex]

I think, you are way over thinking this.

There is always going to be a difference between a DI signal and a miked signal. Sometime on bass it will make a difference to the phase, but I can't see the rhythm being affected that much if its really 2 - 4 samples.

However, if you really feel its off the easiest solution I think would be re-amping the original track. Mute the original amped track and reamp through the original amp with whatever tone you want from that one. Then all tracks are reamped tracks which should be together.

Visually lining stuff up, while a valid tool in certain cases, in my experience generally makes things feel much less connected as opposed to much more. If you want a really good feeling rhythm tracks, just leave them be and reamp or mute the original. You have reamped the original x number of times so you have the performance down.

[dfuruta]

If you're sure it makes a difference, move it around until it sounds best. Right? Just try some different alignments and pick the one that seems right to you. No need for a method.

[losthighway]

If you're blending these different amp 'performances' it would make sense that the waveforms would be a little different. If it doesn't sound 'phasey' not worries.

Sometimes when visually aligning things I find that more convoluted processing paths sometimes inverts the polarity of the track. Reverse it on the new track when it's sitting next to the original and see if a clearer pattern emerges between the waveforms. I also probably don't need to tell you that it's easiest to check the things if you zoom in on a spot with a louder transient so you have a good peak and valley to compare (sometimes nowhere to be found on more distorted guitars recorded at lower levels).
_________________
Mo' mics mo' problems.

http://www.myspace.com/furnaceroomstudio

[Nick Sevilla]

ProTools ADC (Automatic Delay Compensation) only works in the non HD version anyways, with INTERNAL PROCESSING.

Your ReAmping is an external device, so ProTools will never sort that out.

As to how to align things:
You mention using two signal paths, which are a DI and a ribbon microphone.

Questions for clarification on your end:

1.- What is the time difference of these two signals with regular recording? if it is the hardware reaping device, the delay SHOULD BE THE SAME EVERY TIME. Since you mention it being 2-4 samples, it is not the reap device, but something else.

2.- Does it sound "right" like that, as it is? Or do you feel a need to delay one signal of the two to make it sound better? You mention the original recording sounds right. Then, does the reaped one also sound alright?

3.- When you ReAmp, are you re-recording the same two signals, a DI and a ribbon mic? What is the time difference here? The same as when recording without the ReAmp device? Different? By how much?

4.- If you do choose to use the internal plug in for aligning the signals to "perfectly phase match each other", which I say is not the way to go, then you will have to delay the signal that arrives FIRST.

Cheers
_________________
The Song. Nothing else really matters.

[lefthanddoes]

You could fiddle with the instrument cable or plug it in and unplug it or something if you want a simple waveform to line things up with. A guitar strum will not reliably work for that.

As you can see, the consensus is not to go with the theoretical or scientific approach here, but to fiddle around with it until it sounds good. If you want to move back the reamp track more and it's already at the beginning, just chop a tiny bit off the beginning, shift it back more and use delay compensation until it sounds right. You can still use the visual as a basic guideline to get it close.
_________________
http://lefthanddoes.bandcamp.com

[Dakota]

Agreed that 2 to 4 samples isn't going to change anything about feel. 2 to 4 milliseconds can.

It's expected that re-amping is going to change the look of the waveform - the frequency response and frequency/phase response of the amp and mic and pre are going to juggle the small details in the audio.

If you want to be super militant about time alignment, put a short spurt of a very hard edged sawtooth wave on the guitar track before the performance starts, run that out with the re-amping. Use the hard edge for your line-up.

If you want to be maximum militant, also use a spurt of white noise in front as well. Then after recording, put a trim plug on both tracks, flip the polarity of one, and scoot the reamp track until you get maximum cancellation on the white noise. Then take the trim plugs off, that's as lined up as is possible.

(point of maximum phase cancellation is easier to hear with broad spectrum white noise than performance tracks)

[The Scum]

Just thinking about the physics involved, something isn't fully making sense to me. The described delay is too short.

You say the mic is 6" to 8" from the speaker. We know that sound travels through air at about 1 foot per millisecond.

To cross the 6" gap between speaker and mic, it takes about 1/2 millisecond.

We also know that there are 44.1 samples per millisecond.

So we should expect that the reamped signal would be at least 22 samples behind the original, accounting only for the air gap.

Of course, this is a guitar signal, with some alteration being done in the reamp path. It may be changing in such a way that features of the two waveforms just don't correlate, making side-by-side comparison less meaningful.
_________________
"What fer?"
"Cat fur, to make kitten britches."