Topic: Does sample rate affect the accuracy of track syncing when making a matrix?

[Roving Sign]

Does a higher sample rate allow for a finer grain control of track sync?

[sygdwm]

i dont think so. i assume you are using 2 decks and that both are set at the same bit depth/sample rate? i guess i could see if you were zooming in to the nanosecond it might be "finer grain control", but there is no need to zoom in that far to synch.

[morst]

I would think so, however anything over 48KHz is unlikely to be audible to humans. Plus, the 2GB filesize limitation on WAV makes recording 88.1 or 96K impractical unless you have a recorder which does seamless splits.

I am very happy with 24-bit/48KHz personally. The increased bit depth makes it unnecessary to run levels so hot, which makes mixing sources MUCH easier!

[Roving Sign]

I may not have phrased my question too well...

And maybe its really a question about Audio Editing Softwares...

Im really wondering if - when time syncing (and combining) two wav files - is the time shift ability of some software infinite? - or is there a relationship to sample rate?

As I've thought about it a bit more - Im thinking the answer is "no"

Time shift is probably infinite and not related to sample rate...

[SmokinJoe]

I think the best that a human can hope for (using our usual tools like a computer sound editor) is to align the tracks within 1/1000th of a second.  CD quality is 44 times finer grained than that.  In audacity if you zoom WAY in you can see dots for individual samples.  If you are good enough to align those dots, then maybe it matters, and if that's your case, you are a much better man than I.

I think the more difficult task is figuring out the alignment by ear (when we frequently have a hard time getting closer than 10/1000ths), and I'm not convinced that having 96 samples per .001 seconds is easier to align than 44.

[DSatz]

The means certainly exist to align two recorded tracks to within a small fraction of a single sampling interval--Cedar Audio's been doing this for years with 0.01 sample precision. But this capability isn't made available as such in conventional editing software. You'd have to take a somewhat roundabout route of upsampling, aligning, matrixing and then downsampling again.

Just taking 44.1 kHz for a moment--a one-sample timing discrepancy would equal an adjustment of between 1/3" and 1/4" in the relative distance of two microphones from the sound source. As far as I'm aware, sharpness of localization depends most of all on midrange and upper midrange timing relationships, so I'd be surprised if even finer corrections would often be needed.

On the other hand for M/S recording it's hard to imagine why it should ever be necessary to adjust by more than a tiny amount, even if that does give a sonic benefit--because it's just not that hard to get right in the first place. Last night I hung an M/S pair and by choosing appropriate mounting hardware, got the centers of the two capsules to align in the vertical plane within maybe 1/16"; I doubt that I'll have to time-slip those two channels any.

--best regards

[morst]

Just taking 44.1 kHz for a moment--a one-sample timing discrepancy would equal an adjustment of between 1/3" and 1/4" in the relative distance of two microphones from the sound source. As far as I'm aware, sharpness of localization depends most of all on midrange and upper midrange timing relationships, so I'd be surprised if even finer corrections would often be needed.

dangit! I think you just blew my mind again! If my mics are not aimed squarely at the source then my left and right channels may need to be "time slipped" ?! Zoinks! I usually run ORTF since that's the only mounting bar I have for the KM140's, and freqently record from right of center at one local venue, since that's the best place I've found to clamp.
Good thing I save all the master files before I process them!!
"someday I'll get it right!"