Description image


DSLRs aren’t designed to record high-quality audio; they lack professional inputs for microphones and have an auto-gain (AGC) circuit that ruins any chance of manually setting your levels. This has changed with the recent introduction of Canon’s firmware upgrade for the 5d, which enables manual audio recording, so it’s not to say you can’t record theatrical-quality audio on your DSLR production; you just need the right equipment. Essentially, you have two options for audio recording: on-camera, or separate-system. On-camera is what any video camera user will be familiar with; you plug in your microphone(s), and the audio is recorded together with your video. Separate-system is what filmmakers accustomed to shooting celluloid will be familiar with; you record visuals to one medium and sounds to a separate recorder, and then have to sync the two up during editing (thus the need for a clapper on shoots). Both approaches have their pros and cons (briefly: separate system affords you higher quality audio recording at the expense of convenience in the editing room; on camera is the converse), and which approach you go with will depend on your production needs and whether you even have the option of on-camera recording (your DSLR may not have a mic input, or you may be unable to disable the auto gain).

For a series of very helpful DSLR audio recording tests watch Jon Fairhurst’s roundup of audio recorders (he also makes uncompressed audio samples available):

  1. Boom Mic
  2. Camera Mounted Mic
  3. Wireless Lavalier
  4. Foley
  5. Noise
  6. Conclusions

To record pro-level audio in-camera, you will need a device to connect your microphones (usually a XLR connector) to your camera (in the case of DSLRs, the audio input is a consumer-grade 1/8″ stereo jack). The consensus on “best pre-amp/XLR adapter” is the Juicedlink CX231 ($300). It offers pristine mic pre-amps, good construction quality, very low noise, and phantom power for your mics (if you don’t need phantom power, the CX211 is cheaper). To use it, you screw the Juicedlink onto the bottom of your DSLR (or attach it elsewhere on your rig), plug in your balanced XLR audio cables into the Juicedlink, and run the included 1/8″ stereo cable to your DSLR’s mic input. There are still a few issues — namely, headphone monitoring is in mono and the levels are quite low. To address these problems, Jon Fairhurst recommends a Boostaroo headphone adapter to achieve a usable monitoring volume (I’ve found the Boostaroo makes the audio approximately twice as loud, which is adequate for most situations).

If your DSLR doesn’t offer manual audio settings, the JuicedLink DN101 offers a solution to defeat DSLR auto-gain. But whereas the 5D’s new manual settings allow you to retain both channels of audio, the DN101′s hardware solution gives you one channel instead of two (it defeats the AGC by blasting the unused channel with noise). This is fine if you’re recording one microphone, but if you want two discreet channels of independently-adjustable audio and your camera doesn’t offer manual audio input levels, you’ll have to go separate-system. It’s worth noting that Beachtek also offers a DSLR solution, but its tone-based method of defeating AGC is, in my opinion and a lot of others’, unusable for pro audio (this newer version I linked to is reportedly improved, however). Additionally, JuicedLink is now offering the DT454, which builds in the AGC defeating tone as well as level meters and a headphone monitor.

As with any XLR adapter of similar size, a Juicedlink or Beachtek can also double as a riser plate on your support system. Furthermore, the Juicedlink website offers a catalog of very helpful audio recording tutorials — indispensable if you find yourself trying to turn a friend into an audio crewmember.

If on-camera audio recording isn’t an option for your DSLR — or if you have a need to untether your audio from your camera — there are a number of quality flash memory recorders available at very inexpensive price points. One of the most widely-used is the Zoom H4n (review, review), which records at (up to) 24-bit/96kHz on SD or SDHC cards. It offers 2 XLR inputs, a built-in stereo mic, and offers 4 tracks of simultaneous recording, which enables the ability to simultaneously record its own stereo pickup in addition to audio from the 2 XLR inputs; for filmmaking where foley won’t be possible, this can offer a nice mixing alternative, in that you can adjust separately-recorded ambient levels independently from your boom and lav levels. Having such a scratch track, assuming the H4n is protected by a windscreen and is well-positioned, can be a boon for post-production turnaround time on low-budget shows that have web, mobile, or low-end TV releases in mind. At $350, the H4n is a great deal. One alternative: The Tascam DR100, which others review well, but I witnessed the silver jog dial fall off repeatedly on a shoot (YMMV).

In addition to these two methods of recording, there’s a third approach: purchase a CX231 to enable on-camera audio when you want it, and add a flash memory recorder for separate-system sound when the production calls for higher quality audio or needs the recordist to roam. A top choice for this purpose is the Sony PCM-D50, which is similar to the H4n in size and functionality, but it only accepts a 1/8″ stereo input. However, when connected to a CX231, the combination offers a great combination of mic pre-amps and recording medium — you could use this recording setup for a feature film, the self-noise is so low. Videographer/reporter Dan Chung has reported good results, and my own tests to date have confirmed this (however, your audio will travel over an unbalanced 1/8″ cable for a short length, so you’ll have to be vigilant for interference). An important feature of the PCM-D50 (review, review) is its built-in limiter, which records each input at two levels simultaneously and automatically switches to the lower level if clipping results at the higher setting. For DSLR-based crews where the audio guy might not be a union man with decades of experience, this extra fault tolerance is crucial. Other advantages include 4GB of built-in flash memory (good for 6 hours of 48kHz audio), a 5-second pre-buffer (you can automatically start the recording 5 seconds before you pressed the button in case you weren’t rolling for something), and a real, rideable input level knob (instead of the H4n’s pushbuttons). Disadvantages to the PCM-D50 include its higher price, its 1/8″ input, and its reliance on proprietary Memory Stick cards (honestly, Sony, still?).