RMS Levels

I've been working on my mixing and proper mixing levels and started using a template created by member jsaras (his name is Jonas, here's his template http://forum.cakewalk.com/tm.asp?m=843953). According to him, the proper levels are:

-20dB RMS, -6dB Peak RMS before mastering and
-14dB RMS,-10 dB Peak RMS after mastering.

Using Voxengo Span, I'm getting PeakRMS left -8.1 and Peak RMS right -8.4, and RMS -20.9.

Using DFX RMS Buddy, I'm getting pretty much the same thing, -21.45 average RMS.
My Master bus is set to RMS and it gives me a peak of +1.4, which RMS Buddy does as well.

Am I in the ballpark? I still have no idea what I'm doing, but Jonas' template has helped quite a bit with workflow as well as using sends, ect., and of course reading threads for hours from this forum helps as well. Any feedback would be appreciated.

Robert Anthony

I don't understand the numbers. You're peaking at -6db before mastering and -10db after mastering? Normally, the peak is higher after mastering, often as high as -0.1db (I shoot for -1db myself).

-20db RMS is OK for dynamic music such as jazz or classical. -14db RMS would be a good target for rock and pop. However, a mastering engineer might boost that up.

If you're seeing +1.4db on the master bus, you've gone too far. While track levels > 0db are acceptable during mixing (although you still want to avoid them if you can), you absolutely cannot exceed 0db on the master bus.

I try to get my RMS around -9, -10 and peaks around -.5 when I "master" my tunes.

The problem is how the "peak" RMS is calculated. RMS means Root mean squared. Its an average of things. Peak is the highest peak level that you get. You can average the peaks but how its done will have an impact on what it says. Further it really should be the real peaks. The reason is this is used to have a notion of where your highest peak is and if its clipping. RMS gives a notion of the general over all loudness. Averaging the peaks is interesting data but its not all that useful. The range between the peaks and the RMS readings give you an idea of the dynamics of the song. The "RMS" Peak isn't telling you much. I guess I am used to looking at this this way and I am set in my ways.

Edit to add What does mean? Is it the peak average or average peak?

RMS is calculated using a small "window", usually around 50 ms but adjustable in some applications. With each window you can calculate the RMS for that window and RMS for all audio up to that point using previous RMS data. Peak RMS (PRMS) in SPAN is the maximum RMS found for any single window.

When RMS is near PRMS then most of your material has the same loudness and is probably very un-dynamic. However, most people adjust listening volumes such that RMS levels are at a comfortable sound level. If you had a Peak RMS of say 15 dB higher than RMS then it's possible that the listener will be blasted by very high volume wherever that happens to be in your track. IIRC, most material I've worked with has PRMS about 4-8 dB higher than RMS, but in some cases it may go outside that range. Our ears are more sensitive to Peak RMS than instantaneous peaks so while we may not hear a very loud instantaneous peak we will be able to perceive a very loud Peak RMS value.

I'd say PRMS should just be used as a guide to make sure that your mix isn't too outta whack. It probably doesn't matter too much as long as 2 < (PRMS minus RMS) < 12. If it's outside that range then I'd try to have a clear understanding why it's that way.

ORIGINAL: blueoneblue

So to get a handle on what everyone has said, after mastering it should be around -14 RMS (for pop/rock) and peak RMS 4 to 6 db higher, with peak on the master at or just below 0.

Thanks again
Robert Anthony

That won't be up to current "commercial volumes" if you care. It depends on the type of music too (don't know if you said that)

Most pop/rock/new country music is going to RMS above -10 (even -7, -8) and peak right up against 0dB

When I polish my mixes (I hesitate to call it "mastering") I use Elephant and the final levels in SPAN are around -11 dB RMS and -6~7 dB PRMS. Works for me.

I would not worry too much about RMS prior to mastering - just worry about getting a good mix and controlling the peak.

I try to keep -pre-master peak at -3dB or lower. This is to "leave room" for mastering. Sometimes I will stick a hard limiter at that level to make sure I don't go over, but that's only if there are a couple little peaks here and there that are close - other than that, I just back off in the mix.

ORIGINAL: MatsonMusicBox

I would not worry too much about RMS prior to mastering - just worry about getting a good mix and controlling the peak.

I try to keep -pre-master peak at -3dB or lower. This is to "leave room" for mastering. Sometimes I will stick a hard limiter at that level to make sure I don't go over, but that's only if there are a couple little peaks here and there that are close - other than that, I just back off in the mix.

Everyone is different , but I pay very little attention to PEAK levels during recording, and about 100 times more to RMS levels whilst recording.

It would take me a zillion pages of typing to explain why but basically.....

I mix at a reference level coming out of my monitors....as I mix song to song, the same volume generally hits my ears from my monitors, this reduces the effect that volume has on percieved levels....search "Fletcher Munson"...this gives me incredible "repeatability" on my mixing.

I use a system to manage levels during tracking....search "0VU =18dB" or "K20 system Bob Katz"


Basically, I track instruments leaving a boat load of "headroom" in the digial realm. I track using the "analog" part of the chain is operating at optimum levels (I dont overload the mic, I try and put the mic in the right place,I use the preamps in their optimum position"

This allows me to get a detailed and clear sound.

I track to the K20 principle, this gives me a 20db "Crest Factor" which means I can just get on with the job of tracking, and at them end of the take, check the PEAK meter to make sure I dont have any overs, which I never do , setting stuff up like this...takes the guess work out.

When I have finished a mix, I also mix at K20, My mixes have good consistency, and go to the mastering engineer with maybe 5 or 6dB of headroom, sometimes a little more, if its say a ballad, and maybe a touch less, if the song is a bit more full on. The mastering engineer loves me...he has room to work ...8).


I spent a good year , working through all this stuff, after already being a very experience "home recordist" (say 20 years at that point) when I went to make my second self released album.....I made the point of saying to myself, I want to learn why my recordings dont sound as professional as what I hear on CD....and I worked insanely hard to get as good as I could...

Now you can listen to the song in my signature, to decide for yourself If I achieved my goal, to my mind I have, and it was worth the thousands of hours, of reading, re reading, practicing, pestering, and more and more reading.....try and look past the song, if you dont like that style of music, and listen to the engineering and such.


My RMS level when mixing is at -20db....that matters zero...really, again listen to the song in my signature, and see if you think it suffers acoustically...it has been mastered, of course, and its mastered by a professional, for good reason, I sell the CD, and I want the best for my music, after working on this album day and night for a year, I have no objectivity I couldnt master myself out of a wet paper bag....8)....


Spend your time, learning how to track and mix, and getting the best out of your voice, instrument, song arrangement, room, gear...everything ....dont worry about mastering...at this point...in fact unless you are presenting more than one tune to someone, its not really mastering in the true sense....


dont mean to come over as preachy....I just sometimes wish , I could take some of the stuff I have gone through and just place it in others hearts/heads so they wouldnt have to go thru it.....


hope it helps someone...

cheers

Wiz

Everyone is different , but I pay very little attention to PEAK levels during recording, and about 100 times more to RMS levels whilst recording.

The reverse here. I am far more concerned about peaks when recording then RMS.

I mix at a reference level coming out of my monitors....as I mix song to song, the same volume generally hits my ears from my monitors, this reduces the effect that volume has on percieved levels....search "Fletcher Munson"...this gives me incredible "repeatability" on my mixing.

What does this have to do with record levels?

Basically, I track instruments leaving a boat load of "headroom" in the digial realm. I track using the "analog" part of the chain is operating at optimum levels (I dont overload the mic, I try and put the mic in the right place,I use the preamps in their optimum position"

This seems to contradict the first statement.

I am not trying to attack you on this but recording is not the hardest thing a person will do. Over the yers I only try to get a good signal and thats about it. It ain't rocket science. You are right everyone is different.

I guess what's important is that he has a method that he has arrived at that allows him to achieve consistency in his mixes.

ORIGINAL: John

Everyone is different , but I pay very little attention to PEAK levels during recording, and about 100 times more to RMS levels whilst recording.

The reverse here. I am far more concerned about peaks when recording then RMS.

I mix at a reference level coming out of my monitors....as I mix song to song, the same volume generally hits my ears from my monitors, this reduces the effect that volume has on percieved levels....search "Fletcher Munson"...this gives me incredible "repeatability" on my mixing.

What does this have to do with record levels?

Basically, I track instruments leaving a boat load of "headroom" in the digial realm. I track using the "analog" part of the chain is operating at optimum levels (I dont overload the mic, I try and put the mic in the right place,I use the preamps in their optimum position"

This seems to contradict the first statement.

I am not trying to attack you on this but recording is not the hardest thing a person will do. Over the yers I only try to get a good signal and thats about it. It ain't rocket science. You are right everyone is different.

Hey no probs....I wont take it as an attack....and dont read anything I say as negative...8)..we are just having a nice chat...8)...


What does this have to do with record levels?

This seems to contradict the first statement.

everything actually, you get used to working at a consistent volume, I can basically listen to whats coming out my monitors and know if I am not getting enough signal to tape so to speak.....8)...

See the thing is, you need to have a standardised set up, and each piece in the chain needs to operate at a consistent and proper operating condition...the guitar has to have good strings, the player has to be tight, the room has to be good, the mic goes in the right postion, the preamp gain is set so the preamp is operating optimally, the DA convertors input sensitivity is set appropriately to capture the output of the preamp (-10 or +4 etc etc ) and then the volume coming out of your monitors, is set to the right level...so you can judge that relationship between kick and bass guitar effectively, day to day, hour to hour...song to song....once you start working like this, things like compression ratios, release times, reverb decay times, etc all become easier to deal with.....IMHO....

hope that helps to understand what I was getting at...

cheers

Wiz

ORIGINAL: MatsonMusicBox

uh ..... OK .... I'll assume that was at least mostly directed to me as you copied my post ... Not sure we disagree much ... I think ... huh ....

anyway ....

You don't want to clip - we all agree - so record 24 bit and don't freak if you don't peak on ind. tracks above -6 or -12 for that matter or even more - especially depending on the source.

I was talking about overall RMS/Peak (2 buss) not ind. tracks. It is the job of the mastering engineer to "close the gap" between RMS and peak - as the mixing engineer - we need to watch clips and give them some room ... -3dB is enough. My point is that if the MIX IS GOOD and you DON'T CLIP (tracks or overall) then the RMS will take care of itself. Use your ears, not your eyes so much.

If the material is "too dynamic" (and you don't really want it to be) that is what compressors coming in or as an insert are for - or you need to re-track and play/sing/whatever smoother.

Overall - I go back to the same point - if it sounds good, it is good.

Are you talking the "40 Years" song in the video from you page (that's all I found) - I like that song a lot BTW - and nice mix and master. I'd point you to my sig too - maybe more than one way to skin a cat? LOL

Thanks!

Hey MatsonMusicBox

sorry, I didnt mean to direct it at you....its just your post I quoted...8)...take no offence..non intended...


We do see eye to eye...8)

and yep, 40 years is what I was talking about...

There are a lot of opinions on the net, and I thought If I am going to give mine, I might as well put up something people can listen to and decide if I am worth listening to in the first place...proof is in the pudding, put up or shut up kind of thing...8)...hey, and If I sell an album along the way...8)...we all gotta eat right...8)

I am gonna go listen to your song now...

and hope there was no misunderstanding.....

cheers

Wiz

ORIGINAL: John

I guess what's important is that he has a method that he has arrived at that allows him to achieve consistency in his mixes.

True but should there be that kind of consistency with all mixes? Just asking. I suppose its that signature sound.

Dont get me wrong every song is different, stylistically, arrangement, tempos affect compression attack and release times, decay times of reverbs, delay times, modulation settings etc...

but relationships between instruments, good dynamics control, space, depth, dynamic change...all those things are consistent....and are helped by a consistent approach to tracking and mixing...

it also helps a lot with ear fatigue, repeatable mixing, etc etc etc...seriously.

Try this, play one of your mixes, set the relationship between the kick drum and bass guitar...now turn the whole mix down 12 db, how is that kick drum sitting?, now turn it up past where you had it, really loud...how is that kick drum sitting...

Do it with delay repeats on a vocal, same thing, set the vocal delay level, across three different mix volumes coming out at your ears.......each will be different..why Fletcher Munson Curve...

This is part of the reason why that mix you thought sounded great last night, sounds like garbage the next day when you come in fresh...8)...I bet thats happened to everyone of us....8).....


Set up your monitoring environment correctly (speaker placement, monitoring level, room treatment) get your tracking under control (room treatment, mic choice, mic placement, gain staging, AD sensitivity settings) and you will be amazed at how much easier, after a time, this stuff gets....



but, as I said, its just one guy on the nets opinion, and that and 2 dollars will buy you a cup of coffee...8)...

cheers

Wiz

edited for incredibly poor spelling on my part..8)

ORIGINAL: space_cowboy

I am with John here. I always want the best signal to tape (or whatever it is). I try to pay as much attention to that as possible. I never ever ever want digital clipping.

I usually go for about -2-3 db below when I am testing before hitting record to make sure I dont clip.

Mixing is an entirely different process.

ORIGINAL: John

Everyone is different , but I pay very little attention to PEAK levels during recording, and about 100 times more to RMS levels whilst recording.

The reverse here. I am far more concerned about peaks when recording then RMS.

I mix at a reference level coming out of my monitors....as I mix song to song, the same volume generally hits my ears from my monitors, this reduces the effect that volume has on percieved levels....search "Fletcher Munson"...this gives me incredible "repeatability" on my mixing.

What does this have to do with record levels?

Basically, I track instruments leaving a boat load of "headroom" in the digial realm. I track using the "analog" part of the chain is operating at optimum levels (I dont overload the mic, I try and put the mic in the right place,I use the preamps in their optimum position"

This seems to contradict the first statement.

I am not trying to attack you on this but recording is not the hardest thing a person will do. Over the yers I only try to get a good signal and thats about it. It ain't rocket science. You are right everyone is different.

I dont disagree with you space cowboy, our difference is in the amount of headroom we leave...but the important thing, is that the preamp be operating correctly and in its "sweet spot".

with 24 bit, bit depth, its a moot point wether the peak is at -6, -12,-18...what is important is to A not clip, ever and B operate the analog part of the chain optimally...

There is no point turning up the preamp, to just get the needle higher...8)...and I am not suggesting anyone here does that...8)

cheers

Wiz