K-System Metering

Can someone explain this K-System Metering to me in simple words? I read the article 2 times and here my questions:

1. If my master meter in Sonar 4SE is peaking at -14 is this using the K-14 system?

2. Do meters like EAS Inspector or Span use the K-System Metering method?

Edit: Typo

In addition to what Andrew explained, another important property is that the master out level should be callibrated to the actual SPL levels coming out. This is not too difficult:

1) You take a full scale white noise, that is, if you play it with no attenuation, the meters in sonar should show the RMS level of -3 dB, for either channel. The K meters show 0dB RMS for the full scale sine wave, whereas Sonar would display that as -3 dB, as Andrew mentioned. So whatever value Sonar shows, a true K meter would show 3 dB more.

2) You pull the master out fader down to -17 dB.

3) Solo only the left channel, and tweak the volume control on your speaker, or on your soundcard until the SPL at the monitoring position is about 82 dB. Use the flat wighting (type C) on the SPL meter. Repeat for the other speaker.

4) When both speakers are on, you now know that whenever the master out displays RMS at -17 dB, the sound level will be at roughly 85 dB SPL. This way, the loudest peaks will hit at roughly 100 dB. Lovely.

If you want to test the mix under extra loud or extra soft conditions, you may want to add another bus layer ("pre-final", just before the true master out) and use its fader to tweak the absolute volume. At any point, you can doubleclick it (snap back to zero) to return to the standardized volume level, which was carefully set with the (most likely automated) final fader.

The point of K-meters is portability: anyone who uses callibrated K-14 system, when playing your mix will hear the same absolute loudness as you.

I might be mistaken (because I don't own one) but the UAD-1 might have something (one of its plugins) that uses K-metering, either built-in or available as a download from their site. But here we're talking serious $$$. If I'm wrong someone will set me straight...

Dan

ORIGINAL: Otis

K-system meter

Cool... I wish I could just buy a DX or VST plug that has these feature.

I'm surprised no-ones made a vst version.

Good point - come on you budding coders!

ORIGINAL: Dickie

BUT, SONAR's meters do not have the 3dB RMS 'correction' required by K-system meters, so this means that the RMS values for K-14 need to be 3dB LOWER - i.e. loud = -17dBFS, maximum = -13dBFS.

Andrew..thanks for this... but are you sure that is the right way round, as unless I am having a "moment" this conflicts with what tonyd said...

1) You take a full scale white noise, that is, if you play it with no attenuation, the meters in sonar should show the RMS level of -3 dB, for either channel. The K meters show 0dB RMS for the full scale sine wave, whereas Sonar would display that as -3 dB, as Andrew mentioned. So whatever value Sonar shows, a true K meter would show 3 dB more.

So which is right.
Do the Sonar meter read high or low compared to K-14 meters.

Your post has just sent my brain into a spam ... but I've just recovered (unless I'm having a 'moment', of course). SONAR's RMS meters read 3dB lower than a true K-system meter. So if a piece of music mastered to K-14 bounces around the -14dB level on K-system meters, they'll bounce around at -17dB on SONAR. So there's no contradiction. Make sense?

Edit: Brain into a "spam", I'm obviously getting too much, I meant "spin".

I'm using the PSP Vintage Meter to replicate a K-14 meter. I have the VU integr. time set to 600ms, the 0VU refer. level set to -14.0dBFS. The meter display was left clicked to show -40 to +6.

The loudest part of my song on the Vintage meter is hitting -10dB. The meter in Sonar is peaking at -5.7dB.

Do I have this setup correctly or am I way of base?

Edit: K-14

ORIGINAL: Andrew Milne


Those are good settings.


If you've set 0dB reference to -14dBFS, then you should aim for the peaks to be no higher than +4dB. -10dB is way too low. How do you have SONAR's meter set up, and is the RMS or peak value peaking at -5.7dB?

They are setup for peak with 300ms rise and fall. The -5.7dB was peak value. I re-read Bob K's article and decided to do my mixes at K-20 and my mastering at K-14.

Should I set Sonar's meters to RMS? Should my Master fader stay at zero?

Hi all,

At the risk of increasing any confusion, as the Vintage Meter uses the VU averaging method, as opposed to the RMS method, it will read around 1 dB lower than an RMS meter. (The exact difference depends on the audio in question; it could be a touch less than 1 dB lower, or even up to 2 dB lower i.e. on an exposed drum track with no accompaniment. But the 1 dB value is a good rule-of-thumb.)

Anyhoo, IMHO meters are just an aid and it's better to judge the final loudness of your music against other musics whose loudness (or lack of) you want to match, rather than relying on statistics like VU and RMS that are only approximations. The CD Honor Roll on Bob Katz' website is a great starting point for that purpose.

Bye,
Steve

ORIGINAL: stevejay
...If you compare reading on an averaging meter (VU or RMS) set to a 300 ms integration time and then a 600 ms integration time, you notice that the meter reads higher with the former setting than with the latter (as well as being more active).

Hi Steve - thanks! I'll look that up in my Bob Katz manual...I'm just trying to do a little something on vocal tracks to judge dynamic ranges vs how much more compression I might need to add - I know, trouble already but it's just a fun experiment. Like you said depending on the setting the meter will be more or less active - that's what I want to do, dampen the meter a bit to respond to vocals the way my ears does.

Kind of OT from the general intent of this thread about K system...oops

Like you said depending on the setting the meter will be more or less active - that's what I want to do, dampen the meter a bit to respond to vocals the way my ears does.

Kind of OT from the general intent of this thread about K system...oops

Hi again,

No, not at all OT. The shorter 300ms integration time will make the meter more active (which makes it harder on the eyes) but the important thing here is that it results in a meter that reads a bit higher than an averaging meter with a 600 ms integration time. (Try the comparison on a section of a song.) This integration time setting has everything to do with K-System metering.

Bye,
Steve