3.20.2017

Accidentally passed a landmark. Also came to understand why people think camera preamplifiers are so noisy....

I should have noted that the Visual Science Lab blog hit (and passed) the 3200 post milestone. I did the math, that's millions and millions of words and thousands and thousands of photographs. How do you get to 3200+ posts? One post at a time.

On a different note you may have noticed that I'm diving deeper and deeper into the discipline of recording sound. Every good videography needs to know about audio. It may be the most important component of video and probably the hardest to get just right.

I have one little kernel of opinion that I want to pass on. I've been told for years now that still cameras which feature video capabilities all have very noisy and low quality preamplifiers in them and that the only way to record "professional" sound is to skip using the camera's input and start recording to an external audio recorder which, presumably, has cleaner preamplifier stages.

Hmmm. In theory I'm sure that the good digital audio recorders do have somewhat better circuitry and, perhaps, demonstrably better noise but I also think there is a prejudice floating around that has more to do with noisy headphone amplifiers than noisy camera preamps.

I've noticed for some time now that when I monitor the sound coming into my Sony cameras (including the RX10 series and the A7rii) with headphones I get some low level hiss and noise. It's there whether I've matched the microphone to the camera inputs or not. I always freak out when I hear it but I'm usually on a remote location and don't have another option.

The funny thing is that when I get back to the office, import the video files to my editing software and then listen to the output on my studio headphones I don't hear the same, obvious noise. What I hear is fairly clean and accurate audio. I have a short attention span so I hadn't tested my hypothesis until recently. My hypothesis is that the "dirty" audio is a result of crappy headphone amplifiers; not only in the consumer, all purpose cameras but even in most of the separate digital audio recorders I've worked with.

Recently I bought and received a new microphone. It was a well reviewed Aputure Diety shotgun microphone. I was anxious to test it out and wanted to give it every opportunity to excel. That would mitigate any post cognitive dissonance I might have had about spending yet another $360 on my always expensive occupation.

To that end I ran the microphone into the Zoom H5. It's a portable audio recorder that is well known to have low noise preamplifiers. The Zoom H5 supplied phantom power to the microphone and the recording I did was right in the optimum level area for voice (between minus 6 and minus 12 Db, as shown on the meters). There was no indication of overload and the levels were high enough not to be anywhere near a noise floor.

When I monitored via the headphone jack I heard a similar kind of noise that I often hear with signals coming from the camera headphone jacks. A high frequency hiss that's not terrible but not optimal. I was taken aback. All other owners/reviewers of the H5 were effusive in their praise of this model's low noise. Ditto concerning the noise profile of the microphone.

I moved the audio file to my computer, plugged in some Audio Technica headphones and took a listen to that set up. The noise I was hearing went away. It dawned on me that the real culprit in many cases might not be the camera but the camera's headphone circuitry. How could this be?

Well, I looked no further than to the car industry for an analogous comparison. Takata airbags were defective across a range of manufacturers and models, from Toyota and Honda to BMW and Ford. Seems like one airbag maker supplied a lot of different companies. By the same token the headphone amp is probably a feature on a small chip. Easier to spec a universally used microprocessor than to create a custom one for each camera line. And, an inexpensive product to make and sell.

The engineering/marketing rationale for using a noisy preamplifier chip is probably that, historically, so few people who purchased "hybrid" video/still cameras ever ended up shooting video, and the ones that did probably didn't use external microphones. Few would complain about the sound and, if they did complain then customer service could tell them that while it may affect monitoring it would not have an effect on the sound being recorded. Everyone saves money, no sound quality on the video files gets sabotaged.

But, of course, the product manufacturers omit any caveats about monitoring performance and so the urban legends are born and spread. And the legend in our industry is how awful the audio is on our cameras.

My experience reminds me to test, test, test and not to rely on urban legends.

I'll tell you right now that the audio I get from the Zoom H5 or the Tascam DR60ii is better than the audio I get from the RX10iii but in the same breath we are talking about 88 versus 85 and not 95 versus 42. You can do good work with the built in audio circuits of most current Sony and Panasonic cameras. I had good luck, audio-wise, with my Nikons as well. It's more a question of maximizing every step of technique than it is searching for the one "magic" solution.

Do your own tests. Listen to the output of a good system. Listen to the way it sounds through a set of monitor speakers, or through good headphones plugged into a high quality playback source. Don't blame your camera right off the bat.

4 comments:

Craig Yuill said...

This is good to know. There is much info/opinion around about the quality of mics and pre-amps, but little about headphone output. In fact, this might be the first time I have read about this issue.

Mind you, I don't have a headphone jack on any of my cameras (nor do I have an external recorder), so the issue for me is mostly mute. I do, however, have a couple of external mics that are powered via the camera. Although my main video camera (a Nikon V1) does have a pretty good stereo mic, there is a low-volume hiss present during quiet scenes. I haven't used the two external mics very much yet, but the tests I have done revealed that the mics record audio where the low-volume hiss is greatly reduced (or even eliminated). My camera's preamps seem to be okay. I have a feeling, however, that the severity of the pre-amp "issue" might depend on which camera you are using.

BTW, do you recommend any particular headphones or speakers? Also, what kind of an video/audio output device would you consider to be good for hearing the audio?

Tom Judd said...

Off topic for this post, but a rig like that shown has to be intimidating to a lot of subjects! Have you noticed this?

Kirk Tuck said...

Hi Tom, It just looks big in the photograph. I've got the camera set up on the rig for a series of interviews with professional actors. They are used to having microphones positioned on them and working with A/V nearly every day. No sweat for them. And, if you compare the small size of this set-up with a traditional camera, like an FS7 or an Arriflex Alexa, or the BetaCams of yesteryear it's downright tiny. People generally know they are being recorded when they are surrounded by light panels and there are microphones hanging in front of them... :-)

bpr said...

Hey Kirk, isn't it possible that back in the 5DII days the audio intake circuits were a bit crappy since nobody foresaw the vDSLR boom it kicked off. I'm sure later generations of the more video oriented devices will be more capable.

Just a thought because, as we all know, in any line of work once things get to be received wisdom it's pretty hard to change them,