Calibrating monitors to the K-system - Too loud?

[Mordi]

I am reading Bob Katz' book "Mastering Audio", and I was intrigued by the K-system. The idea is that you find a good sound pressure level to master or mix at, and then mark on your monitor gain knob where that level is.

I did this:
1. Generate pink noise with RMS-level at -14dBFS.
2. Turn off one speaker.
3. Move gain knob on audio interface until SPL meter reads 83dB C-weighted.

Measuring was done using the SPLnFFT iPhone-app, which is supposedly fairly accurate. It was placed at the listening position.

The problem occurred when moving the gain knob up. When the measure read about 70 dB, I had to go get some earmuffs because of how loud it was. Am I really supposed to master or mix at 85dB? It seems way too loud for me. Could my iPhone be attenuating the loudness somehow?

Does anyone else have any experience with this?

[karbomusic]

The default K-System recommendation of 85dB assumes a fairly large room (like a small theatre) IIRC, if your room is smaller and you feel it's personally too loud you can bring it down to a comfortable level so long as you mark it and can return to it just like you would 85dB.

[ashcat_lt]

K14 at 83dbSPL should be reasonably loud, but not terribly uncomfortable for most sources. Peak levels hit 97dbSPL, which is definitely not messing around. I think it really is the sweet spot and you should just take a minute to get used to it and keep things around there most of the time when making real critical decisions. It's ok to turn it down a bit when doing things like editing and stuff, and pretty important even in mixing if only to give a new perspective and a bit of a break against ear fatigue.

I have my "living room" system calibrated to K20 at ~85dbSPL for television and movies. Big explosions shake things (peaks at 105!), typical music is uncomfortable and hypersmashed metal is kind of frieghtening.

But all this very much depends on your monitor system having the headroom to get that loud and stay clean. A clipping monitor amp or distorting speaker will be a lot harsher and more fatiguing in general.

Course it definitely IS possible that the meter on your phone is just way off. You're supposed to calibrate the meter itself, but that requires a known source, which is hard to find when you can't reliably measure to begin with. I never did that myself, but I know it's close enough for my purposes just from experience.

[James HE]

I'd rather mix at conversation levels, how else I'm I'm supposed to hear the voices in my head that tell me what to do?

85db is stupid loud to me. I'll crank it up to get a reference, but I'm not comfortable working at that level for a long period of time.

Do what feel right, it's your health, not Bob Katz'.

[Jeffsounds]

I used an app on my Android phone to check and 85db was pretty quiet to me. I mean SERIOUSLY quiet. I could whisper to someone next to me and they could hear me. I don't trust phone apps in this case.

[RJHollins]

First ... be certain that your SPL meter is set to the proper ballistics.

C-weighted, SLOW Response.

As to target volume, yes, the suggested 83dB [SPL] of One monitor [at a time]
will yield 86dB[SPL] with both speakers.

It WILL sound loud. This is a target for large Control Rooms.

Bob Katz has also posted 'revised' comments, suggesting lower target values for smaller rooms. [79 dB[SPL] is one example].

2. Understand WHAT and WHY the Katz system IS ... and what its intention is.

hint: since it is cal'd to -14dB-RMS Pink Noise ... [which I feel is TOO HIGH, and prefer referencing to -18dB], So should your TRACK levels use the same Ref Level. When you push it to 0dB ... it will be loud.

There is plenty of IMPORTANT DETAILS in Katz writing on this topic. It is imperative to read and understand.

[Mordi]

I ended up going by ear and marking a comfortable spot for -14dBFS.

Also ordered a dedicated SPL meter online, so I can measure again. Nice to have anyway.

[Judders]

Probably worth remembering that a bad room can create massive peaks and dips across the frequency spectrum, so perceived loudness may differ a lot from SPL readings taken at the speaker.

[Juan D'Fall]

85db? I sing louder than that. Typically between 90 - 100db (@1m) and I'm not a particularly strong singer. Limit your exposure and give yourself breaks while mixing if it feels too loud. And don't sit too close to the speakers. That 85db should be where your head is. If you set that level at one distance then sit closer, it will sound a lot louder. Or if you are too close to next door, you may have to mix quieter. I did read somewhere that 85db is the optimum level for mixing at but I don't know how true that is.

[RJHollins]

Originally Posted by Juan D'Fall

85db? I sing louder than that. Typically between 90 - 100db (@1m) and I'm not a particularly strong singer. Limit your exposure and give yourself breaks while mixing if it feels too loud. And don't sit too close to the speakers. That 85db should be where your head is. If you set that level at one distance then sit closer, it will sound a lot louder. Or if you are too close to next door, you may have to mix quieter. I did read somewhere that 85db is the optimum level for mixing at but I don't know how true that is.

Look up 'Fletcher Munson Curve'.

[Stella645]

Table of recommended levels based on room size.

https://www.soundonsound.com/techniq...g-levels#para7

[karbomusic]

Originally Posted by Stella645

Table of recommended levels based on room size.

https://www.soundonsound.com/techniq...g-levels#para7

Thanks for posting that, it's what I was digging for when I posted but couldn't find it.

[Philbo King]

Interesting topic. I've found that mixing for balance between tracks or stems works best at very low levels, but when adjusting for freq balance, 83 dB spl is best. If you turn it down and pieces of the mix disappear, you're not done mixing yet.

This 83 dB level is intended for optimized acoustically treated rooms, too. If you are not in a well treated room, cranking it up to 83 dB will also excite freq nodes and nulls, causing all kinds of weirdness (holes and peaks in the freq response) in the bass and low mids. That's one of the reasons near-field monitors were developed.

I find that mixing quietly really helps translation to other systems quite a bit; it's very rare that people listen to music for enjoyment at 83 dB levels (this depends quite a bit on music genre; EDM, metal, and other "loud" music types might be played much louder than 83 dB). It also helps prevent hearing fatigue that can have a big effect of mix decisions after many hours of mixing.

But it's also important to occasionally listen at high levels to check for balance between low bass, mids and high "air" levels.

Just my humble opinion...

[ashcat_lt]

Originally Posted by Philbo King

...cranking it up to 83 dB will also excite freq nodes and nulls...

Naw. Those things happen at any volume. Nearfields, being closer to you, help shift the mix of direct sound (what you want to hear) and reflections (what you don't really want to hear because it's not actually in the mix) toward the speakers themselves. It's mostly about the higher frequencies. Bass will still be wonky either way.

[emarsk]

I also tried to follow that calibration recommendation but found that it's way too loud for me.

[karbomusic]

Originally Posted by emarsk

I also tried to follow that calibration recommendation but found that it's way too loud for me.

Here's why that might be...



Ref: https://www.soundonsound.com/techniq...g-levels#para7

I personally think 83/85 is fine for my ears though I'm in a much smaller room but I'm less sensitive to SPL at my age. That being said, I still spend a good bit of time mixing at very low levels in the low to mid 70s, and sometimes as low as I can get it for balance checks because things that are poking out or too low will become more apparent when on the fringe. The noise floor in my studio is at times near 30 dB SPL so I can stand very low levels without the floor getting too much in the way.

I also check mixes at 90+ from time to time but I never mix at that level.

[James HE]

Originally Posted by karbomusic

I personally think 83/85 is fine for my ears though I'm in a much smaller room but I'm less sensitive to SPL at my age.

I feel like I got more sensitive to spl levels when I had a kid, lol!
can't stand overly loud anything these days!

I'm currently, technically in a 20,000+ cubic feet room. It's also my shop however, so at high levels, all the ratchets and wrenches and bits and hardware start rattling in their toolboxes!

[karbomusic]

Originally Posted by James HE

I feel like I got more sensitive to spl levels when I had a kid, lol!

Hehe, that might be a real biological change due to having a kid. I'm more like 1700 square feet so 20k sounds grand.

[ashcat_lt]

Originally Posted by karbomusic

Hehe, that might be a real biological change due to having a kid.

Didn't work for me. I tell my kids "If it's too loud, you're too old!"

You know, cause they're always complaining about my loud music...

[Tod]

Originally Posted by James HE

I feel like I got more sensitive to spl levels when I had a kid, lol!
can't stand overly loud anything these days!

I'm currently, technically in a 20,000+ cubic feet room. It's also my shop however, so at high levels, all the ratchets and wrenches and bits and hardware start rattling in their toolboxes!

Originally Posted by karbomusic

Hehe, that might be a real biological change due to having a kid. I'm more like 1700 square feet so 20k sounds grand.

Wow, 20,000 cubic feet! I've got a pretty good sized room and I've only got about 2,720 cubic ft.

Karbo, is that 1700 square feet or cubic feet?

You know, even with my hearing loss, I'm still pretty sensitive to volume levels in my room. I haven't checked in quite a while, but I think my mixing levels are 80dB or less, I think less.

[Philbo King]

Originally Posted by ashcat_lt

Naw. Those things happen at any volume. Nearfields, being closer to you, help shift the mix of direct sound (what you want to hear) and reflections (what you don't really want to hear because it's not actually in the mix) toward the speakers themselves. It's mostly about the higher frequencies. Bass will still be wonky either way.

You are correct. However, when working at quiet levels the Fletcher-Munson effect causes a large reduction in perceived bass anyway.

Ultimately there is no substitute for proper room acoustical treatment.

[karbomusic]

Originally Posted by Tod

Wow, 20,000 cubic feet! I've got a pretty good sized room and I've only got about 2,720 cubic ft.

Karbo, is that 1700 square feet or cubic feet?

You know, even with my hearing loss, I'm still pretty sensitive to volume levels in my room. I haven't checked in quite a while, but I think my mixing levels are 80dB or less, I think less.

Hi Tod,

Cubic, it's an estimated guess but should be fairly close. I actually like "jamming out" kind of loud, but mixing I usually keep pretty low.

[James HE]

Originally Posted by Tod

Wow, 20,000 cubic feet! I've got a pretty good sized room and I've only got about 2,720 cubic ft.

Overestimated a bit. It's steel hangar type farm building. 40' x 28' x 12', so 13,440 cubic feet. Trapezoidal shape probably knocks a bit off of that. ~12,000 square ft. or so?

ehh... I wasn't too far off. LOL

We have a larger building down the road where they used to work on tractors and stuff. Maybe I got confused. That one is well over 20,000 cubic ft. Big holes in the roof though. I do want to move in there at some point, but will take a substantial investment to do that. I feel cramped in here, believe it or not!

https://imgur.com/cqiW7ol

Studio area is only like 8x10. Yes, things get dusty. I put mics and instruments away. Clean often. Those 20 year old Event 20/20s don't give a rats ass about a little dust

[ashcat_lt]

Originally Posted by James HE

Those 20 year old Event 20/20s don't give a rats ass about a little dust

They're self cleaning. Just crank them up and they remove their own dust.

[Tod]

Originally Posted by James HE

Overestimated a bit. It's steel hangar type farm building. 40' x 28' x 12', so 13,440 cubic feet. Trapezoidal shape probably knocks a bit off of that. ~12,000 square ft. or so?

ehh... I wasn't too far off. LOL

We have a larger building down the road where they used to work on tractors and stuff. Maybe I got confused. That one is well over 20,000 cubic ft. Big holes in the roof though. I do want to move in there at some point, but will take a substantial investment to do that. I feel cramped in here, believe it or not!

https://imgur.com/cqiW7ol

Studio area is only like 8x10. Yes, things get dusty. I put mics and instruments away. Clean often. Those 20 year old Event 20/20s don't give a rats ass about a little dust

Heh heh, by my figures, that's about 1120sqft, big enough you could build a nice control room as well as a good band room right inside your building.

[RJHollins]

I use the '1812 Overture'. When those cannons go off ... the dust settles.

[ashcat_lt]

I think maybe I was a bit cavalier in my first reply. In all honestly, these are pretty much my maximum levels. I spend a lot of time with master fader down 10db or so, and like when I'm just watching sitcoms on Netflix I turn it down to less than half most of the time.

It really is my "true reference" level for mixing, and when we watch movies things shake.

[endorka]

It's been a while since I've studied the theory and performed the setup and calibration, so my thoughts could be a bit disconnected, but I use this system in practice every day.

The K system was incredibly useful, but I think many including Bob Katz recommend switching to the EBU R128 now. The concept is similar but more advanced and widely applied. For example, I can enable EBU R128 normalisation in my "consumer" JRiver media player.

A key point is that the system can evolve with you as you use it. So I initially calibrated for 83dB per speaker. Playing un-normalised already mastered music through this was often excruciating, but that's not the point. The point is to get a level in the room you mix at, where music you intend to be loud in the mix actually sounds and feels loud to your ears. And that this will be sufficiently below peak distortion that you never have to worry about "going over". That something quiet like a whisper is quiet like a whisper. That the exciting drum fill makes you feel excited, and a good groove makes you feel like dancing. So in essence the calibration is to you, not a standard as such.

So, I set the speakers to 83dB, and mixed naturally, and after the first couple of mixes noticed that the integrated loudness (LUFS) of these mixes came out somewhat below -23 LUFS. I was aiming for -23 as a good dynamic mix. Now, no harm done obviously, as relative levels were ok, and mastering sorts out the absolute volume. But over time and more mixes, I eventually settled on a speaker calibration level of 77dB, and now consistently produce mixes very close to -23 LUFS.

And herein is the beauty of the system. Say you wanted to mix or master something to -13 LUFS. Just turn your monitor controller down by -10dB, and mix / master the music to your normal listening levels where everything feels good, and you're sorted.

This also applies to music playback. If you listen to Californication un-normalized on a system calibrated to -23, it will tear your ears off. But normalize to to -23, and it fits in nicely with everything else.

Cheers,
Jennifer