Recommend a mini-portable audio recorder?
 

Any recommendations for a mini-portable audio recorder? I was thinking of picking up a couple of Zoom H1s for a documentary but then noticed this part of the market has really exploded.

Tascam DR-03
Tascam DR-07
Tascam DR-08
Tascam PR-10
Zoom H1
Zoom H2
Alesis PalmTrack

I have the DR 07, and own a and H1 that I just bought, and had to send back because of a battery drain problem. It seems the early models of the H1 had some drain issues, though they record fine.

I use both as a recorder for my Sign Video ENG 44 mixer, and they both work great for that purpose. The H1 has a very simple record interface, and slider switches to choose various setting. The DR 07 gives more recording choices, and you need to access menu. I haven't used either extensively standing alone using their on board mics. My brief experience with that type of recording favors the H1, but as I said, I did not test against each other, and can't support that with anything but limited trial in similar situations.

Build wise, I think the DR-o7 presents a much more solid, but heavier piece of equipment.

Sony PCM D50

Jim the Sony PCM D50 is really nice but it's around US$500. The others recorders I mentioned are around US$100, the idea being to get a handful of inexpensive mini recorders and stick them close to the subjects. I already have a sound guy with a full kit, but he can only do so much, so we're going to add several US$100 mini-recorders for better flexibility and coverage.

Got it! I guess when you said a "couple" you were meaning a "bunch"!

Most of these small recorders work well as long as you DO NOT USE the built-in mics, thus you have to add the cost of a lav for each person you want to record.

My preference is for the Olympus LS-10/11 or Sony PCM-M10, both record very good audio and will run for almost two days on 2 AA batteries.

If price is an issue, then the Zoom H1 + Lav might be the cheapest way to go.

I use the H1, H4n and Olympus LS-10. All of them will give you great quality audio, but the feature set differs. If you even think of using a pro quality (XLR) microphone, then the H4n is a good bet. The H1 has a smaller display that makes it harder to monitor, and chews through AA batteries at an alarming rate. Build quality isn't the best either, so keep it well protected. The LS-10 is a solid chunk of recorder with a much better battery life, easier to read display and tactile controls that make it a little friendlier to use. I've had more audio drift with the LS-10, which is the only reason I use it less often than the others. I don't think any of these units will keep perfect sync over long interviews, so take my advice with a grain of salt.

All Non Time-Code recorders will drift, but if the drift is consistent you can correct for this afterwards.

I use the Olympus LS-10 for long-form live events without any problems, but I always adjust the sync on the WAV files afterwards using the cheapest audio editor I own ( GoldWave $19 to buy ) because it has a very accurate time adjustment feature called TimeWarp ( good to 0.0001 seconds when processing hours of audio ). Depending on the camera I am shooting with, the sync has to be adjusted for between 2-20 frames of drift every hour recorded. I can completely adjust an hour of audio in less than 1 minute, so it`s a pretty quick process.

Just FYI, timecode or the lack of it has no effect on the 'drift' of a file based recording system such as current digital recorders. Back in the day of DAT, and analog before it, LTC provided a speed reference for playback. But it doesn't do that for file-based systems. All timecode does is establish a single reference point, a timestamp marker of the first audio sample and matching marker on the video frame during which that sample occured, to use to align the audio and video files, functionally identical to a manual slate. To eliminate drift, one must provide a common timebase to feed the sample clocks in both the audio recorder and the video camera so they're slaved together,and the clocks in the two devices are running at identical speed as the recording is made. Any difference, no matter how little, such as a video device clock running dead-on at 48.0000000000000000000 kHz but an audio clock running at 47.999999999999999999 kHz, will cause the two files to playback at very slightly different speeds when played together as they're merged in post. It's that playback speed difference that causes the drift. The proper fix when shooting is to provide a clock signal from a single device that sends genlock sync to the camera and wordclock sync from the same clock to the audio recorder. A set of tuned Lockit boxes, although physically separate devices, qualifies as a "single clock" since they are so accurate and can be tuned exactly to match each other, that's why they cost a few bucks. Of course that means the recorder has to accept external wordclock sync, not something found on consumer recorders like the popular Zooms, while the camera has to accept genlock sync, again not found on consumer and prosumer cams. Another way to provide a common clock is to use the video camera's internal sync signals, found in its composit video output, with an audio recorder that has a video blackburst input to drive sync on its sample clock, again, found in some pro-grade recorders but not present in the cheaper consumer-level models.

Those Zoom H1 recorders look cool for what they are, and they will power the new Rode Lavs when you get the MiCon-2 connector.

I mentioned timecode because the cameras/recorders I've used that feature proper TC usually have a very accurate internal clock, so that even without genlock they will maintain frame accurate sync for a very long period of time.