Precision Limiter: K-System confusion

[iguana]

Hi,

I've recently delved a bit into the K-System specifications, and with the addition of the K scales to the precision limiter, I've run into an oddity. I'm pretty sure it's just my ignorance, perhaps someone who's more acquainted with this system can enlighten me:

As far as I understood, the various K scales differ in the placement of their respective 0dB point, where 0dB on, say, a K-14 scale is supposed to be at -14dBFS. According to Bob Katz' book, a calibrated monitor should output 0dB on each scale at 83dB SPL, giving you 14dB of headroom in this case. So far, so good.

Now here's where I get stumbling: if I feed pink noise at -20dB RMS into a channel strip of my DAW and place the precision limiter right behind the generator, it correctly reads -20dB RMS in PK-RMS mode, but +3dB on a K-20 scale! Why is that? Isn't it supposed to read 0dB?

Perhaps one of the mastering whizzes around here can shed some light on this, as I'm not fully sure to which spec I should calibrate my monitors to... thanks!

-jan

 

[csl]

I'm not sure but, I think there are two standards for peak metering, one by steinberg and the 'classic' industry standard, with one being 3dB louder than the other (the steiny one by the looks of things).

I seem to recall folks complaining about the PL when it first came out because it adhered to the steinberg standard rather than the industry standard, or something (which the K system is based on). So, the peak will appear be 3dB out when comparing scales, as they're based on differing standards. I don't know why they're different.

 

[iguana]

I seem to recall folks complaining about the PL when it first came out because it adhered to the steinberg standard rather than the industry standard, or something (which the K system is based on). So, the peak will appear be 3dB out when comparing scales, as they're based on differing standards. I don't know why they're different.

You're right (although it's RMS, not peak , and I have to do a self-RTFM here: it's right there in the manual.

Quote:
"When the meters are in the K-modes, the displayed RMS level is 3.01dB higher when compared to the same signal level in the Peak-RMS mode. This is done to conform to the AES-17 specification, so that peak and average measurements are referenced to the same decibel value with sine waves."

So, mea culpa... but then again, which reading am I supposed to calibrate to 83dB SPL? Peak levels of pink noise are unuseable...

 

[MASSIVE Mastering]

You calibrate *your monitors* to 83dBSPL.

The K-System relies on a specific calibration of your monitoring chain. Without knowing where your monitors hit a standard level with a -20dBRMS level, it isn't going to do you any good at all.

 

[iguana]

You calibrate *your monitors* to 83dBSPL.

The K-System relies on a specific calibration of your monitoring chain. Without knowing where your monitors hit a standard level with a -20dBRMS level, it isn't going to do you any good at all.

Yes, perhaps my wording wasn't clear - in order to calibrate my monitors, I need pink noise at an RMS level of -20dBFS. Now, do I use an AES-17 pre-compensated scale (such as the PL's K-Modes) or a non-compensated scale to determine what IS -20dBFS? The difference is 3dB...

-jan

 

[anha]

There is a lot of information on Bob Katz Digido website including a -20 dBFS RMS pink noise 44.1 audio file. Just look at downloads.

http://www.digido.com/

 

[csl]

I seem to recall folks complaining about the PL when it first came out because it adhered to the steinberg standard rather than the industry standard, or something (which the K system is based on). So, the peak will appear be 3dB out when comparing scales, as they're based on differing standards. I don't know why they're different.

You're right (although it's RMS, not peak , and I have to do a self-RTFM here: it's right there in the manual.

Quote:
"When the meters are in the K-modes, the displayed RMS level is 3.01dB higher when compared to the same signal level in the Peak-RMS mode. This is done to conform to the AES-17 specification, so that peak and average measurements are referenced to the same decibel value with sine waves."

So, mea culpa... but then again, which reading am I supposed to calibrate to 83dB SPL? Peak levels of pink noise are unuseable...

oh yeah, peak differences would be silly. whoops. :lol: