Minimum recording levels for studio and field audio recordings? (help settle a disagreement)

[Khashyar Darvich]

Thanks for your feedback, Ty.

I just spoke with this other studio engineer friend of mine, and he mentioned to me that although ideally you want to have your average highs at tone (which is -12db’s on my FCP and digital camera meters), that an audio meter scale and the levels that you set depend on the RANGE of that audio meter scale.

For example, the audio meter on Final Cut Pro ranges from 0 to about -96. My engineer friend (not the same person who is working on my film) said that this means that zero is at about 96 db’s, and that an average reading of -24 to -30 db’s on a scale of 0 to -96 db’s is very acceptable. My friend mentioned that -30db (on a scale of 0 to -96) would be able 66 dbs, which is loud enough to increase the gain and not have artifact issues.

What are your thoughts about this?

Are audio levels relative based upon what the range of that scale is? Does -24/-30 on an audio meter scale of 0 to -96 sound like a good level?

Thanks,

Khashyar

[Ty Ford]

I just spoke with this other studio engineer friend of mine, and he mentioned to me that although ideally you want to have your average highs at tone (which is -12db’s on my FCP and digital camera meters), that an audio meter scale and the levels that you set depend on the RANGE of that audio meter scale.

**Your language “average highs” is confusing. Let’s get specific. All audio is comprised of RMS (root mean squared) and Peaks. RMS is equivalent to average. Different meters may or may not indicate both. Both are important; especially peaks readings. Peaks of -12 are OK with really good gear. MOST camcorders are not that good. If you peak at -12 on MOST camcorders and then reset and record with peaks at -5, and listen carefully, you will hear the diffference in noise. I’ve done this with a PD150. The hiss level at the lower record level was obvious.

**The RANGE of the audio meter scale is very important. There is a horrible lack of meterig standards in the industry today, especially with prosumer cameras. What’s important is knowing where -20 dB is and where 0 is. Theoretically, you can set tone at -20 and use the dynamic range all the way to 0. Continuous tone doesn’t have peak content, BTW. So, after setting tone at -20, expect most of your audio to be well above -20 if your meters read peak.

**Different people have different RMS to peak ratios in their voices. Setting levels properly and riding gain requires vigilance. Good audio is not “set and forget.”

For example, the audio meter on Final Cut Pro ranges from 0 to about -96. My engineer friend (not the same person who is working on my film) said that this means that zero is at about 96 db’s, and that an average reading of -24 to -30 db’s on a scale of 0 to -96 db’s is very acceptable. My friend mentioned that -30db (on a scale of 0 to -96) would be able 66 dbs, which is loud enough to increase the gain and not have artifact issues.

** “Zero is about 96dB” What does that mean? SPL (sound pressure Level) measurements and equipment level measurements both use decibels (dB) as a unit of measurement, but they are different. What you don’t know, is whether the 0 to -96 are really decibels or just some range the manufacturer decided to use.

What are your thoughts about this?

Are audio levels relative based upon what the range of that scale is? Does -24/-30 on an audio meter scale of 0 to -96 sound like a good level?

**Without knowing how good your gear is (camera mixer, mics) — how noisy they are — the only way to tell is to compare different record levels for noise as I mentioned above.

**The -15 peak recording I make on a Sound Devices 744T will be quieter than a -15 peak recording on any camcorder on the market. Why? Because even the most heralded camcorder makers spend their money on the picture and not the sound.

**Further, what you hear in the headphone output of a camcorder may not be what’s on the tape because many headphone circuits in camcorders are horribly noisy. You really have to get the audio out of the camera to be able to make really good comparisons, assuming the other gear you have has better noise specs.

**And don’t get me started on HDV audio, which isn’t even 16-bit, 48 kHz audio to begin with. It’s MPEG.

Regards,

Ty Ford

Ty Ford’s “Audio Bootcamp Field Guide” was written for video people who want better audio. Find out more at https://www.tyford.com

[David Jones]

Not wanting to open up a big can of worms here… But!
Just because someone sits behind an impressive looking console in city X, and brags about having a $15k microphone,
does not mean that said person knows his or her business when it comes to audio recording.

[James Sullivan]

I have heard that Audio Mixed at -20db from a professional VO Session brought into FCP did not line up at -20 and I think we are onto something in the Scale of the FCP meters VS. the rest of the world.

Interesting

James