Two Separate Digital Recorders Can't Be Sync'd

Elmer23

New member
Hello... The other day I was experimenting with two different digital audio recorders by having them record in different areas. My intent was to combine them in Audacity, with each unique recording to be used as a separate channel in a stereo mix. I did a clap at the beginning and end (when both machines were moved next to each other) so I could easily sync them in software, and left them recording ambient sounds for about two hours (I had actually got caught up in something else while the recorders were running). When I synced the claps at the beginning of each recording, it was clear the recorders were about 2.5 seconds out-of-sync with each other at the end of the two hours (the final claps were almost exactly that amount of time off).
What might be causing this, and is this kind of variation expected between different recorders? The equipment used was a Zoom H5 and a Philips DVT6500, both set to 44.1kHz, 16-bits.
 
Digital clocks aren't always exactly the same. One clock was running faster than the other. The longer the recordings the more they drift. That's why pro digital gear uses wordclock connection when using them together, to ensure that their clock are synced perfectly. Sometimes you can sync consumer/prosumer gear through S/PDIF connections.
 
Ya, that is normal, and maybe with even two of the same.

Here's two interfaces synced from the spdif of my word clock, so you did pretty good : )
 

Attachments

  • P1120308_crop_700.jpg
    P1120308_crop_700.jpg
    130.3 KB · Views: 18
I think the import thing is that he did test it : ) But ya, I wouldn't think it would break a 5-min tune
 
Digital clocks aren't always exactly the same. One clock was running faster than the other. The longer the recordings the more they drift. That's why pro digital gear uses wordclock connection when using them together, to ensure that their clock are synced perfectly. Sometimes you can sync consumer/prosumer gear through S/PDIF connections.

Thank you for the clear explanation. It's actually reassuring to know even pro gear requires timing synchronization!
 
You always have to find some way to sync the clocks when you use two machines.

If these recordings are the left and right of one stereo recording, you are pretty well screwed. If you time stretch either one, the image will drift throughout the recording.
 
Ya, that is normal, and maybe with even two of the same.

Here's two interfaces synced from the spdif of my word clock, so you did pretty good : )

That screenshot is brilliant; I'd never have expected so much variation with solid-state equipment. Much thanks for posting, garww. It really put things in perspective.
 
iqi616,
I haven't tried time-streching either one. But I'm partial to the idea of breaking them up into smaller segments and getting those lined up as best I can, since the recording ran much longer than I expected anyway. Both rooms were recorded with a stereo mics, so I'll convert the respective tracks taken in each room from stereo to mono, listen for interesting points, sync the segments, cut, and combine them into a new stereo track.
In actuality, this was a kind of dry run for a bigger endeavour, involving several recorders placed over a wider a area. I knew variation in timing existed and could be a problem, but had no idea what to expect, or if what I was getting falls within normal tolerances. I'm quite happy to see it does, but it's still a challenge for someone with my limited knowledge of editing. So even if this doesn't work out, at least I'll come away with some knowledge.
 
You always have to find some way to sync the clocks when you use two machines.

If these recordings are the left and right of one stereo recording, you are pretty well screwed. If you time stretch either one, the image will drift throughout the recording.

Yeah, I was planning on converting each stereo track made by its respective recorder to mono, then recombining them as a single left-right stereo track to capture the ambience across both rooms. So I shouldn't assume a linear drift? That makes things all the more complicated. Now I'll be keeping my recordings shorter for when I need to sync two devices.
 
Well, if they were actually synced they wouldn't be so far off.

I guess I'm accepting "synced" as a relative term here. Because I know absolutely nothing about word clocks I had just assumed that, like the internal clocks of the recorders, there will be variance. If anything, it does tell me to expect variation, as you've already said...But I can't go anywhere beyond a naive and purely theoretical assumption of what I was seeing in that shot.

Quick edit: The tracks look to me about 15 minutes out-of-sync with each other (just by quickly eyeballing). That's enough for me to stop worrying about timings on my humble equipment! At least I know what to expect...But there's a need for more experimentation to find out how my clocks are drifting out of sync and hopefully be able to fit that data to some sort of curve. But that's a new ball of wax.
 
Last edited:
No, if the clocks are synced, they are synced. There will be no drift. If that weren't the case, digital audio and video simply wouldn't work.
 
No, if the clocks are synced, they are synced. There will be no drift. If that weren't the case, digital audio and video simply wouldn't work.

So garww posted a screenshot of unsynced tracks? Hey, my average 1.25 seconds per hour isn't so bad after all. I'm glad I don't have to worry about this stuff on a regular basis since my Philips doesn't have SPDIF in.
But I appreciate what you are saying: "synced" should be taken as "synced," and there's less-than-little tolerance for error when evaluating timing equipment.
 
Last edited:
Well, if they were actually synced they wouldn't be so far off.

You mean they could be less far off : ) The discussion is fairly old about MFG & mags suggesting we only get decent clocking with internal. I suspect the same round of testing on my analog wordclock would be better. This was just black channel testing on a budget and one mid-line interface.

In my case, I just need to sync some random s/pdif with my Akai DR8. So, I wanted to see how random that could be. haha
 

Attachments

  • P1120294_crop_700.jpg
    P1120294_crop_700.jpg
    131.7 KB · Views: 3
You mean they could be less far off : )

If they were actually synced they would not be off at all.

The discussion is fairly old about MFG & mags suggesting we only get decent clocking with internal.

The findings published by Sound on Sound are that internal clocking is optimal and external clocking is only slightly worse. It counters the unsubstantiated claims of the superiority of expensive external clocks. Both qualify as more than just decent.

Does Your Studio Need A Digital Master Clock?

Sound on Sound said:
Overall, it should be clear from these tests that employing an external master clock cannot and will not improve the sound quality of a digital audio system. It might change it, and subjectively that change might be preferred, but it won't change things for the better in any technical sense. A‑D conversion performance will not improve: the best that can be hoped for is that the A‑D conversion won't become significantly degraded. In most cases, the technical performance will actually become worse, albeit only marginally so.
 
Back
Top