I've just had the weirdest R44 experience ever. It was totally my fault, but I still don't understand how what happened could have happened.
Last Thursday I recorded a certain rehearsal, and so did a video guy. My mikes were well placed while his weren't, plus his camera records through tinny sounding built-in electrets and uses heavy, heavy ALC. Having done this before, we planned for me to send him a 48 kHz mp3 file of my recording, which he could then sync up with his video in Adobe Premiere. There's never been a problem with this arrangement, and that arrangemend worked fine for the first half of this rehearsal, too.
Unfortunately, I should have changed the R44's batteries during the intermission, but I didn't--and I didn't notice the battery indicator, and didn't have an outboard battery with me. Totally, totally, my fault. It's been years since I messed up a recording like that. The recorder shut down during the second half of the rehearsal about four minutes before the end.
I swapped in a new set of batteries and got the final two+ minutes. But it was clear that I would need to rely on my friend's recording for the ~90 seconds that I'd missed. It was just a rehearsal recording, so that's embarrassing and a bummer, but no tragedy.
Getting ready for this bit of audio surgery, my video friend sent me his 48 kHz WAV file, which I converted to 44.1, while my recording of the second part of the rehearsal existed in two 44.1 kHz files: one long one (from the end of intermission until the batteries died), and one short file (the end). The long file wouldn't load into Adobe Audition, and when I looked at its WAV header I could see why: It indicated a 0-length DATA chunk. I repaired the file header, and then the file loaded OK and sounded good. I found suitable edit points, dropped the missing music in, evened out the volume levels and sent my friend a 48 kHz mp3 file of the result.
However, he soon reported that it was several minutes longer than his video recording. And this wasn't a 44.1 vs. 48 kHz thing; I was careful about converting back and forth, and both his recording and mine play back at the same musical pitch. Yet mine takes place over a distinctly longer time span than his, and his matches with his video, while mine doesn't.
You think that's weird? Try this: The ratio of time scales for the two recordings isn't constant. Depending on which musical "landmarks" I measure between, the ratio of elapsed time for the same passage in each of the two recordings is anywhere from 1.377:1 (for a segment near the beginning) to 1.168:1 (for the recording as a whole up to the moment of battery crap-out).
My recording sounds plausible to me at the pace it has, but then so does the camera recording, and it matches the video in duration. I've looked closely at the waveforms in my recording and I don't see any repeated samples or groups of samples, nor do I hear any digital chattering. Plus I can't imagine why there would be, under the circumstances. In the end, I had to tell the video guy to use the sound track from his camera for the entire second half of the rehearsal, although it doesn't sound very good.
This is just so freakish that I had to post the story. Has anyone else here experienced anything like it?
--best regards