K Metering

www.aes.org/technical/documentDownloads.cfm?docID=65 here is a nice paper about the system.

It seems to have moved on a tad. Something about filtering out low frequencies, ballistics, headroom and how many dB between varying colors on the meter to fall in accord with K-16v2.

In any case I just remeasured and recalibrated my system.

I have a question (which I'm hoping to resolve by reading the internet) I'd like to post here on my homebase:

If you play a pink noise file, -20dBFS mono, to each of two stereo speakers to impart a 75dB c-weighted, slow response reading on an SPL meter then should you not use two simultaneously-playing opposite mono files to calibrate a mono speaker even though the summed RMS would be -17dBFS?

I'm trying to figure out which reading to use for my mixcube. Playing two mono oppositely-panned files simultaneously sounds more similar to a single mono playing back full-range. After all, in practice both stereo speakers will be on and the mixcube is just one driver.

Basically, do I compensate my mixcube by increasing the level 3dB? That's what I've been doing and it seems/sounds correct.

Thanks,

-m

Currently experimenting with 75dBC calibration in my room.

It's loud.

I tried 74dBC but it's too tame. That 1dB makes the difference.

I learned that your volume needs to be "loud, comfortably" rather than "comfortably loud". The latter allows you to be in control whereas the former puts the music in charge safely but powerfully.

-m

Hi Hexspa,

I'll take some time to look at your links but in looking for a brief definition, I found this thread on Gearslutz:

www.gearslutz.com/board/mastering-forum/76806-k-system-explained.html

"In a nutshell, it's an attempt at revising the metering system of DAWs. Currently, peak meters representing dbFS (decibels Full Scale - where 0dBFS is the maximum, the point of clipping).

Analog gear has typically used the dBVU system, where you aim for an average level in the sweet spot of your analog gear. You typically have lots of 'headroom' - maybe 30dB in good gear, which means you have room for transient peaks.

With digital - too many people are aiming too close to 0dBFS, and not allowing enough headroom. This means when they interface with analog gear (including the converters themselves) the results are often crunchier than they need to be.

The idea about "using all your bits for maxiumum resolution" has been swallowed by many people, even though it's not actually correct. Digital audio is very different from digital pictures - the analogies just don't work. Your eyes see dots - and more is better. Your ears hear a re-created continuous waveform, regardless of how many 'dots' were used to define it. You can define a sine wave with 2 dots or 1000000 dots - and it's still the exact same sinewave. So more bits doesn't necessarily translate to better resolution - unlike pictures.

Sooo - the K system is a way of metering that encourages us to apply a specified headroom to digital audio, and try to stick to it. E.g. K14 means that you define your top 14dB as headroom, so when your K14 meter is showing 0, you have 14dBFS of headroom. It uses Average RMS peak values - not actual highest peak values."
Kiwiburger

and

"Just don´forget that half of the K-system is to calibrate the sound volume from your speakers using a sound dB meter."
Gunnar

I've been doing the former by simply "mentally calibrating" my DAW meters so I simply "Imagine" that -20dbFS = 0. The part about calibrating the monitors is something I'll have to look into.

Cheers,
Rock