Synch Drift in Zoom H2 PCM file?

[David Ennis]

There's a post in another forum by someone who tested the H2's 48/16 PCM recording against a video that was recorded simultaneously. The H2 file was about ten frames slow after 45 minutes.

Can anyone confirm or contradict? My H2 is still in the box, and tomorrow is the last day to call for an RMA. Yes, I suppose I could plan on stretching it for every project, but my HiMD recorder had led me to expect perfection even after 90 minutes.

[Wayne Brissette]

Originally Posted by David Ennis

Yes, I suppose I could plan on stretching it for every project, but my HiMD recorder had led me to expect perfection even after 90 minutes.

I don't care what gear you have, unless the same PLL (clock) providing the clock, there will be some drift. Now, the H2 seems a bit high, but you will see some drift. As you get further away from the consumer marketspace, the equipment has tighter tolerances and thus less drift. No two clocks are perfectly in sync with each other.

Wayne

[Steve House]

Ancient proverb says: Man with a watch always knows exact time but man with two watches is never really sure.

[Sean Kiely]

For what it's worth, I have a H4, and it definitely drifts. I need to adjust speed to 100.5% and adjust by a couple of frames every 20 minutes on FCP to synch with audio from my Panasonic HVX200. It's a pain in the ass at first, but once you know what to do it's a two minute exercise.

The H4 has amazing sound quality for the price - and on that basis I'm going to buy another couple of the cheaper H2's for work, they'll be invaluable, even a little disposable if an elephant eats one.

So, if you have more time than spare cash, I say keep your H2 - but if you need to get the job done without fuss, and don't mind spending the extra few bucks, aim for maybe a Marantz or similar.

[David Ennis]

Thanks guys.

Wayne and Steve, my point was that my Sony HiMD and all my camcorders recording the same event show NO relative drift after 90 minutes of continuous recording, so I can't quite agree based on my experience. This man with "four watches was still sure of the time."

Sean, apparently the H2 is greatly improved over the H4 in this regard. 10 frames in 45 minutes is on the order of 0.01%. And, in fairness their primary target market was not videography. Thanks for your comments. If it sounds good along with the convenience and economy I'll keep it, make the adjustments and call it all good.

[Wayne Brissette]

Originally Posted by David Ennis

Wayne and Steve, my point was that my Sony HiMD and all my camcorders recording the same event show NO relative drift after 90 minutes of continuous recording, so I can't quite agree based on my experience. This man with "four watches was still sure of the time."

David:

Unless you have the exact same circuit providing the clocking there will be drift. You have the best situation however in that you found two devices which are nearly identical. If you take two or three recording devices, record 90 minutes worth of audio onto them, then place them all into a DAW, you'll start seeing the drift. It's there and it can become very apparent depending on the device. That's why when you record with multiple devices you have a master clock that drives all the devices. This input clock is usually only seen on professional audio gear however, and consumer devices tend not to have a way to externally clock them.

Regardless, I'm not doubting that you aren't seeing this with the Sony MD, but realize you simply have some devices that have better tolerances with their internal PLL circuits than most.

Wayne

[David Ennis]

Wayne, I'm talking about four devices, not two--a Pany GS200, a Sony VX2100, a Canon GL2 and My HiMd recorded from the board at the event all showing perfect synch at frame level with the video. If it's chance, that's a lot of luck.

One time I taped a school event with seven cameras, adding our video club's four Sony TRV22's to the above. Same result. No adjustments necessary.

Is there a finite drift in the audio? Of course. But not any of a magnitude that matters in terms of synching with any of the video tracks or each other. So my experience makes 10 frames after 45 minutes seem monstorous.

[Andrew Plumb]

David, something I'm not clear about then. Since you have one (in the box), have you confirmed the 10-frame drift after 45 minutes you mention at the top of this thread, as reported by someone else? Or would that disqualify you from exercising the RMA option?

All it should take is setting up the H2 and a camera, set the audio sample rates to be identical (i.e. 48 kHz), hit record on both. Clap once at the start, let them record for 30+ minutes, clap again at the end and measure the time between claps.

[David Ennis]

Andrew,
No I haven't confirmed it with my own unit. The poster I referred to had done exactly as you said and posted screen shots of the audio tracks. It was plausible because of previous reports about the H4. It seemed unlikely that he had a defective or nonrepresentative unit. As you probably know, B&H will take a return as long as the piece is pristine and the packaging materials are all there and in good shape so I wasn't worried about that. Plus, I've pretty much decided to keep it anyway. So I will test mine.

[Andrew Plumb]

Looking forward to seeing how it goes.

Make sure you do explicitly set the audio sample rates to the same thing. Many recording devices targeted at musicians default to 44.1kHz to match CD standards, not the 48kHz rate (or some multiple like 96kHz/192kHz) to match DVD standards.

[David Ennis]

Well, I just ran the test and got basically the same results as the other guy.

Sony VX2100 aimed at the TV (so I could look at lip synch as well measure events on the audio timelines. Zoom H2 a few feet to the cam's right. I set the H2 to 48Khz, 16 bits. I aligned the first hand clap in Vegas and measured the Zoom's delay on the second clap 33 minutes later to be 8 frames (I didn't try to get any more precise). My unit is apparently a little worse than the other guy's.

[Roger Shore]

As some of the posters have noted, if you have two clocks involved, and no time code to sync to, all remote audio recorders are likely to show some 'drift' - especially noticeable on longer takes.
I posted a possible solution here: http://www.mfbb.net/myvideoproblems/...yvideoproblems
which I found useful.

But it does depend on the remote clock being just a slightly different frequency, rather than 'drifting'. Should be OK for most crystal clocks though!

[Bob Huenemann]

I have been following this issue since it showed up in H4 threads.

It doesn't matter if the clocks in the two devices are correct or not!

Think about it. All that matters is that they both play back at exactly the same rate at which they recorded.

If the two inputs are being fed to some DVD creation software that makes assumptions about the clocks, I suppose drift could be introduced.

Short of that, it sounds like the H4 and H2 play back at different rates than they record. That's a separate issue. And it should be easy enough to test. Record a track from WWV, for example. Then play it back against WWV. Or any other stable clock.

[Bob Huenemann]

It has been noted that the H4 has a low level click or pop when running from batteries that is absent when running from the power supply. Has anyone checked the H2 to see if it has a similar issue?

[Dan Bridges]

Originally Posted by Bob Huenemann

It doesn't matter if the clocks in the two devices are correct or not!

Think about it. All that matters is that they both play back at exactly the same rate at which they recorded.

Typically though the H2 & H4 don't play back the recording. Instead it is transferred digitally (file copy from SD card or via USB port) to a multi-track editor in the computer. So you end up syncing the DV cam's digitally-transferred (via firewire) "48Khz" datastream with the H2/H4s "48Khz" datastream. And that's where any difference between the actual freq of each device's "48Khz" sampling rate becomes significant.