What RMS to shoot for to keep clients happy (going for a commercial rock sound)?

[danbronson]

I offer "quick mastering on the cheap" to bands I record and/or mix (I'm no M.E.)...basically just EQ, compression and hard limiting tailored to the song to sweeten things and bring up the volume to competitive levels. I'm not a professional (doing this out of my basement) and the bands I record don't have much of a budget, so this is usually the best they can do.

But what are competitive levels these days? I can slam a mix and leave it sounding like garbage (but appease the deaf rockers) or leave it wide open like a Steely Dan recording and appease the audiophiles (not that I record any, ). Obviously the less I mess with it, the better. But at the same time, most clients will not be happy if I don't give them a recording that sounds at home in their iTunes playlist. I'm always second guessing my choices in this regard and want to put out 'masters' that will appeal to more people more of the time.

So what kind of RMS do you guys generally shoot for (in just the loud parts of the song like the choruses)? Obviously it'll depend on the material in question, but maybe there's a ballpark average?

A quick look at the loudest and quietest rock masters in the last year that I'm aware of (Death Magnetic and Chinese Democracy) reveals -7.5 dB RMS -14.5 dB RMS respectively (just in the loud parts of a couple songs I looked at). So maybe the average of about -11 dB?

[Farshad]

But Death Magnetic sounded like crap! dfegad

[droogs]

the loudest cd ever is from eminem
and it is still sounding very clear and defined

Analyzed folder: E:\TRAX\Eminem-Relapse\
----------------------------------------------------------------------------------------------
DR Peak RMS Filename
----------------------------------------------------------------------------------------------

DR13 over -18.68 dB 01-eminem-dr._west_(skit)_(produced_by_dr._dre_and_eminem).mp3
DR6 over -7.13 dB 02-eminem-3_a.m._(produced_by_dr._dre).mp3
DR6 over -6.84 dB 03-eminem-my_mom_(produced_by_dr._dre).mp3
DR6 over -6.94 dB 04-eminem-insane_(produced_by_dr._dre).mp3
DR7 over -9.36 dB 05-eminem-bagpipes_from_baghdad_(produced_by_dr._dre_and_trevor_lawrence_jr.).mp3
DR6 over -7.90 dB 06-eminem-hello_(produced_by_dr._dre_and_mark_batson).mp3
DR9 -0.00 dB -11.80 dB 07-eminem-tonya_(skit)_(produced_by_dr._dre_and_eminem).mp3
DR6 over -6.15 dB 08-eminem-same_song_and_dance_(produced_by_dr._dre_and_dawaun_parker).mp3
DR6 over -7.09 dB 09-eminem-we_made_you_(produced_by_dr._dre_and_eminem).mp3
DR6 over -6.65 dB 10-eminem-medicine_ball_(produced_by_dr._dre_and_mark_batson).mp3
DR13 -5.88 dB -20.69 dB 11-eminem-paul_(skit).mp3
DR6 over -7.12 dB 12-eminem-stay_wide_awake_(produced_by_dr._dre).mp3
DR6 over -8.07 dB 13-eminem-old_times_sake_ft._dr._dre_(produced_by_dr._dre_and_mark_batson).mp3
DR5 over -6.49 dB 14-eminem-must_be_the_ganja_(produced_by_dr._dre_and_mark_batson).mp3
DR8 -0.00 dB -10.82 dB 15-eminem-mr._mathers_(skit)_(produced_by_dr._dre_and_eminem).mp3
DR6 over -7.90 dB 16-eminem-deja_vu_(produced_by_dr._dre).mp3
DR7 over -8.03 dB 17-eminem-beautiful_(produced_by_eminem).mp3
DR6 over -8.68 dB 18-eminem-crack_a_bottle_ft._dr._dre_and_50_cent_(produced_by_dr._dre).mp3
DR14 over -17.52 dB 19-eminem-steve_berman_(skit).mp3
DR5 over -7.43 dB 20-eminem-underground_(produced_by_dr._dre).mp3
----------------------------------------------------------------------------------------------

Number of files: 20
Official DR value: DR7

[voidtunes]

Id shoot for -10 but if the song starts to fall apart back up a bit!

[Waltz Mastering]

Originally Posted by danbronson

So what kind of RMS do you guys generally shoot for but maybe there's a ballpark average?

I've never looked at a numerical re-out of a rms value when mastering...but from analyzing a couple modern rock releases in the past, an average level in the ball park of -11 would seem sane.

So maybe the average of about -11 dB?

[mobius.media]

You have to look specifically at the loud parts like the choruses. Measuring the song as a whole is pointless imo.

-8 to -12 on choruses to my taste but some pro releases go -6.

[dave-G]

Originally Posted by Waltz Mastering

I've never looked at a numerical re-out of a rms value when mastering...but from analyzing a couple modern rock releases in the past, an average level in the ball park of -11 would seem sane.

I'm with you. I've never shot for the number while doing the actual work .. If the client wants "loud", I just push until the music says stop. On some material, that can get into the crazy-loud, minus-single-digit range, while other stuff really can't be pushed as far.

In short: if you have to go for loud, go by ear; let the music find the number.

-dave

[danbronson]

Thanks for the responses guys. That really helps a lot.

droogs, I just took a look at Same Song and Dance off of Eminem's Relapse and I'm seeing an average of -11 dB RMS. A quick listen back to Death Magnetic and it's quite a bit louder. Where are you getting numbers like -6.15 dB?

[Masterer]

Please do not bother comparing Eminem to metalica to the stuff you are working on. Both of those records sound pretty bad for different reasons and attempting to match the level of either will result in disaster.

For what it's worth, Metallica sounds louder because of the crowded midrange... but that doesn't really matter.

What does your record sound like when you make it louder? That is the only question that really matters.

Dave G gave the best summary advice. Follow that and you'll be O.K.

[chironomidae]

I just check to make sure I don't have to reach for the volume control when the song I'm working on is mixed with a selection of songs in a similar genre.

[audiomichael]

Originally Posted by FRP

you MUST first adjust your monitor to give 85dB at this level.

Why?

[danbronson]

Originally Posted by FRP

bob katz, who surely knows something about mastering audio, give some indications in order to stop loudness war.

there is 3 value for each type of material

RMS should stay -12dB for commercial pop/rock K12
-14dB for commercial dynamic song (indie/ jazz ) K14
-20dB for acoustical classic music K20

you must first adjust your monitoring to give 85dBspl at this level (-12 ,-14 or -20 depending on your choice)

these are indications but I m just a young soundi , recodring with poor mics and cracked Le in a 10m2 room.
just quick mastering for the band to listen on theyre itunes like you do.
and I notice since I stay -12dB for RMS, I have enough loudness and no bad surprise when listening on bad hifi.

ok, I must sometime explain to the musicians why they don't sound as loud as mettalica. It takes a little time but they sometimes ask me for more dynamic mastering so it worth.

and i always keep a K20 version.
I m young so maybe in a few years, people will ask for more dynamic sound.


finally whatever you choose for RMS value ( -11, -12, -14 or else) you MUST first adjust your monitor to give 85dB at this level.
I realise this is one of the most important step in mastering ( but i m a begginer)

In the order , I first choose the dynamic texture I want, then caliubrate the monitoring and then I can start mastering.
That's the only way I can realy keep the balance of my mix.

I used to adjust the RMS with my limiter at the end of the master but now I adjust my limiter depending on the RMS I chose at the begginning.
Obviously still limit at the end of the chain but I feel it much better.


remember we, as sound technicians, are at the frontline in the loudness war. We deal with producers and musicians and we still turn the knobs.
So we can do something about it.

Im young and cheap so my secret weapon still : "if you want more loud, go and pay a mastering studio, I will not do that".
It works most of times (in fact everytime) as my master seems acceptable for the price, and bands who pay for mastering studio dont bother me with loudness.


and I ask to show it on the CD to inform consumers


recorded by S. Omebody in NowEre Studio
mastered by S. Omonelse at Elsewhere Studio
Dynamic : K14


even if they don t know Katz s mastering standard, if they just could ask "what is dynamic ?"

Interesting. Are you referring to the average throughout entire songs or just the loud parts of the songs?

...

I gave this a try and I must say, 85 dB feels like it will be fatiguing after less than an hour. I'm monitoring on Adam A7s that are each three feet away from me. I think in a more treated room, with the monitors further away and a less peaky high end (Adams are a little bright), 85 dB would be fine though.

EDIT: I found this paper by Bob Katz that explains the K-System. I'm still not totally 100% on everything but I get the idea and will set up my monitors for K-14. So I guess I just set things up so that pink noise at -14 dB (from 0, digital clipping) plays through each monitor at 85 dB (C weighted with slow averaging)? Then from there I make the averages of my songs end up at 0 VU (-14 dB)? Someone please correct me if I'm wrong.

Also, FRP, I should correct you about one thing. K-12 is not for pop/rock, it's for broadcasting. From the article: "The K-14 meter is for the vast
majority of high-fidelity productions for the home, e.g. home theatre, and pop music (which includes the wide variety of moderately compressed music, from folk music to hard rock). And the K-12 meter is for productions to be dedicated for broadcast."

[bob katz]

There is no single metering level that will tell you your loudness or your "competitive level..." Even with an accurate loudness meter it still wouldn't tell you (much).

One recording may sound sufficiently "loud" at the same meter reading as another which sound either significantly louder or softer for many reasons. The amount of distortion, high frequency response character, rhythmic character (e.g. the amount of snare in the mix), the nature of the song, whether it is acoustic or electric. So many factors. In the end only experience and good judgment on the part of the mastering engineer can answer this question. In fact, if I told you a flat RMS level to "go for" or for an approximate, it would be misleading and be as much as 3-5 dB off depending on the nature of the song, the part of the song that's being referred to, etc. Nor would it answer the overall picture of whether it is "loud enough" to satisfy your clients.

The short and the long of it is to educate yourself what is out there, listen to it on a wideband system, calibrate your ears to that, note the monitor gain which is used, and compare it objectively to what you are putting out and to the long term standard that you have also known. And work on your own skills to create good sounding masters.

When you know the answers to all those questions then you will know what you need to know, which is far more than "what RMS to shoot for". I hope this helps.

BK

[chironomidae]

Originally Posted by danbronson

Interesting. Are you referring to the average throughout entire songs or just the loud parts of the songs?

...

I gave this a try and I must say, 85 dB feels like it will be fatiguing after less than an hour. I'm monitoring on Adam A7s that are each three feet away from me. I think in a more treated room, with the monitors further away and a less peaky high end (Adams are a little bright), 85 dB would be fine though.

In a smaller 85 dB is too much - you might go down to 82 dB. The important thing is to establish a standard for yourself.

[danbronson]

Originally Posted by bob katz

There is no single metering level that will tell you your loudness or your "competitive level..." Even with an accurate loudness meter it still wouldn't tell you (much).

One recording may sound sufficiently "loud" at the same meter reading as another which sound either significantly louder or softer for many reasons. The amount of distortion, high frequency response character, rhythmic character (e.g. the amount of snare in the mix), the nature of the song, whether it is acoustic or electric. So many factors. In the end only experience and good judgment on the part of the mastering engineer can answer this question. In fact, if I told you a flat RMS level to "go for" or for an approximate, it would be misleading and be as much as 3-5 dB off depending on the nature of the song, the part of the song that's being referred to, etc. Nor would it answer the overall picture of whether it is "loud enough" to satisfy your clients.

The short and the long of it is to educate yourself what is out there, listen to it on a wideband system, calibrate your ears to that, note the monitor gain which is used, and compare it objectively to what you are putting out and to the long term standard that you have also known. And work on your own skills to create good sounding masters.

When you know the answers to all those questions then you will know what you need to know, which is far more than "what RMS to shoot for". I hope this helps.

BK

Thanks Bob, your response is much appreciated!

This makes sense. Ears should be the final judge. The K-System makes sense to me too. Music should be standardized so that volume levels don't get out of control. I suppose at the end of the day, what you say about experience and good judgement holds true. A good mastering engineer would use the numbers and his/her ears to find the appropriate level not only for an individual song, but for all music.

Thanks for the responses everyone, I've learned a lot in this thread and gained a new perspective on monitoring and loudness. Hopefully I can put it to good use...

[Coyoteous]

- 11

[Jesse Graffam]

Originally Posted by Farshad

But Death Magnetic sounded like crap!

Yep, and lots of what was wrecked happened during mixing, because you can see pre-clipped waveforms that are under the levels of other tracks in the mix, even after it was mastered.

The only reason the Guitar Hero versions came out so good was that Activision has a lot more power & control over the business relationship and had the position to be able to "ask nicely" to have Rick Rubin piss off, and get the tracking sessions from before it was mixed. The ME obviously doesn't have that option, ever, and if they reject the producer's work flat out, then they won't be mastering the album.

Anyways...

Originally Posted by droogs

the loudest cd ever is from eminem
and it is still sounding very clear and defined
[..]
Official DR value: DR7

I've done DR4 just to mess around with the meter, and it still doesn't sound anywhere near as bad as Death Magnetic. I did it just to test DR meter/rating. Don't do this for any other reason than to just screw around.

Honestly I would shoot for K-14 if you're using a proper VU meter which doesn't have true peak. (you could end up over-compressing your average and over-limiting your peaks to hit -14dB"K" without hitting -0dBTP. whew.)

Originally Posted by danbronson

K-12 is not for pop/rock, it's for broadcasting.

Yes, and it's outdated. The K-System is now 10-some years old, and using technology and ideals that are much older. Not necessarily bad, but the modern broadcaster is really starting to standardize stuff for the very long haul. Which is great.

The latest proposed standards (EBU) look more like -23dB LU (loudness units, a more advanced version of LKFS/ITU-R BS.1770-1) for ~max loudness, with a -1 to -2 dBTP (true peak, 4x oversampled peak detection) maximum peak inter-sample value.

And as Bob mentions above, it's only a quick indicator of the ballpark figure, when it comes to actual loudness to the human auditory system. It's more-so a great new proposal from an engineering standpoint. There are devices further down the broadcasting air-chain that deal with controlling loudness in much more advanced ways than just increasing/decreasing gain.

Originally Posted by chironomidae

In a smaller 85 dB is too much - you might go down to 82 dB. The important thing is to establish a standard for yourself.

Totally. I even use seperate standards for my room, and for headphones. My headphone listening is more like the equivilent of 78dBs-ish.

Originally Posted by danbronson

The K-System makes sense to me too. Music should be standardized so that volume levels don't get out of control. I suppose at the end of the day, what you say about experience and good judgement holds true. A good mastering engineer would use the numbers and his/her ears to find the appropriate level not only for an individual song, but for all music.

Totally true, and yet the "funny" part is that if people actually followed the K-System to a tee (the Papa meter of course) we would still be way better off than where we are at right now, for 80%+ of the "new" music being released.

[bob katz]

Originally Posted by uncajesse

The latest proposed standards (EBU) look more like -23dB LU (loudness units, a more advanced version of LKFS/ITU-R BS.1770-1) for ~max loudness, with a -1 to -2 dBTP (true peak, 4x oversampled peak detection) maximum peak inter-sample value.

I am a member of the EBU P/Loud committee and I welcome the upcoming standards! TC Electronic has implemented an excellent version of a loudness meter available for Pro Tools and for the System 6000. The RADAR history display is quite informative. In due time someone will implement the recommended filtering in a K-system meter, but I've come to the feeling (like TC) that we must eliminate peak metering from the user's eyes, that it is misleading and completely unnecessary, and a simple peak overload indicator in conjunction with a loudness meter will be our best bet for standardization.

Basically, a "K-20" (or K-23 if you will) meter WITHOUT THE PEAK SECTION and with appropriate loudness weighting.

BK

[IIIrd]

Originally Posted by danbronson

Thanks Bob, your response is much appreciated!

This makes sense. Ears should be the final judge. The K-System makes sense to me too. Music should be standardized so that volume levels don't get out of control. I suppose at the end of the day, what you say about experience and good judgement holds true. A good mastering engineer would use the numbers and his/her ears to find the appropriate level not only for an individual song, but for all music.

Thanks for the responses everyone, I've learned a lot in this thread and gained a new perspective on monitoring and loudness. Hopefully I can put it to good use...

+1....ears

[Waltz Mastering]

Originally Posted by bob katz

I've come to the feeling <snip> that we must eliminate peak metering from the user's eyes, that it is misleading and completely unnecessary, and a simple peak overload indicator in conjunction with a loudness meter will be our best bet for standardization.

Originally Posted by bob katz

There is no single metering level that will tell you your loudness or your "competitive level..." Even with an accurate loudness meter it still wouldn't tell you (much).

This is a little confusing. I'm not sure how peak metering is misleading, but I think, trying to eliminate it will present a whole new set of problems.

The "peak metering" thing probably deserves a thread of it's own, which I'm sure would get rather heated; )

[narcoman]

Originally Posted by uncajesse

The only reason the Guitar Hero versions came out so good was that Activision has a lot more power & control over the business relationship and had the position to be able to "ask nicely" to have Rick Rubin piss off, and get the tracking sessions from before it was mixed. The ME obviously doesn't have that option, ever, and if they reject the producer's work flat out, then they won't be mastering the album.

The reason has nothing to do with this at all !!

It's more to do with the multis provided and the stems mixed out. I can assure you - it's entirely accidental that the Guitar Hero version sound better.

[bob katz]

Originally Posted by Waltz Mastering

This is a little confusing. I'm not sure how peak metering is misleading, but I think, trying to eliminate it will present a whole new set of problems.

The "peak metering" thing probably deserves a thread of it's own, which I'm sure would get rather heated; )

Tom. The reason that peak is misleading is that it encourages people to normalize to the full scale peak, which produces extreme loudness variation. The ability to Peak normalize to a single short sample peak was introduced with digital recording and is an entirely artificial construct from the point of view of loudness and compression.

Remember that a compressed signal only sounds louder than the original if you normalize the new peak level to the same or a higher level than the original peak level of the uncompressed material. As soon as you do that, the loudness (average or RMS level) is made higher. And this is why peak normalizing all compact disc masters is essentially a bad idea, from the standpoint of loudness standardization and audio compresion.

If this still is not clear to you, in order for us to be on the same page I'd like to point you to Chapter 14 in the second edition of my book, which discusses how peak normalization is the catalyst for the loudness race.

BK

[Waltz Mastering]

Hi Bob, I didn't realize the want to eliminate peak metering was tied in to the need to stop the war on bad sound. That wasn't mentioned in the post that I quoted and I do not have the book, so that's were some of the confusion lied.

You went on to say, "a simple peak overload indicator in conjunction with a loudness meter will be our best bet for standardization."

When in a previous post you said "Even with an accurate loudness meter it still wouldn't tell you (much)." These two statements seem to contradict each other.

To me, a measurement of loudness is the difference between peak and rms, so it would be hard to have one without the other(loudness meter, rms meter and peak meter)no?

[Jesse Graffam]

Originally Posted by bob katz

I am a member of the EBU P/Loud committee

Not sure if that's different than the group, but I joined it a few weeks back too, and am looking forward to contributing after I test the waters a bit and watch what y'all are doing already. (is it a down-season right now? the list is slowwwww, and using the wiki to communicate instead of a forum like SMF seems ehm... like the 18th century :P)

I'm mostly done with my own implementation of the LU meter recommendations, and it's pretty neat. I haven't tried the TC meter yet, does it have realtime adjustments for the integration times of various things like mine? I find that it's much nicer to see the effect of the integration times (and threshold) of the secondary short-term "gating" meter/signal that underlies the LU meter, with the realtime adjustability.

Another thing that I added and am playing with and has solved one of P/Loud's latest concerns is by allowing samples that peak higher than the final/visible LU meter to effect the change in loudness the meter shows, even if the short term "gating" meter is under the threshold. So far in my testing it seems to have almost completely solved the issues with content that is "too dynamic" in the very short term, while also being "bouncy" in mid-term... which is what was causing the short-term gating meter to not show a high enough LU meter due to not rising above the gate threshold in the very short term. pretty darn simple "fix" too.

I have it as stand alone and a Winamp plugin now, but I'll probably slap it into the VST 2.4 SDK after I feel good enough about how it's working to show it to P/Loud guys. It's too bad TC won't build a VST version which is pretty easy... cos I'm not buying ProTools or a System6000 just to compare my meter to it.

[Kayo]

Originally Posted by bob katz

I am a member of the EBU P/Loud committee and I welcome the upcoming standards! TC Electronic has implemented an excellent version of a loudness meter available for Pro Tools and for the System 6000. The RADAR history display is quite informative. In due time someone will implement the recommended filtering in a K-system meter, but I've come to the feeling (like TC) that we must eliminate peak metering from the user's eyes, that it is misleading and completely unnecessary, and a simple peak overload indicator in conjunction with a loudness meter will be our best bet for standardization.

Basically, a "K-20" (or K-23 if you will) meter WITHOUT THE PEAK SECTION and with appropriate loudness weighting.

BK

Bingo!
I think that’s a great suggestion, Bob!
A peak meter type indicates the instantaneous level of audio, and can be well covered with the overload indicator. I completely agree. As long as there’s an option to switch it on or off, if need be, as it’s not entirely useless either.

On the other hand, the perceived loudness RMS meter is of greater importance. Works more like the human ear detection averaging all the incoming. Thus, I prefer to use RMS, Vu and then peak because the aforementioned, more accurately relates to what a human listener will experience in terms of relative loudness. The RME metering is comprehensive and covers most of what I need, although the TC-RADAR looks inviting and hopefully ported to the TC Powercore.

FD

[bob katz]

Originally Posted by Waltz Mastering

Hi Bob, I didn't realize the want to eliminate peak metering was tied in to the need to stop the war on bad sound. That wasn't mentioned in the post that I quoted and I do not have the book, so that's were some of the confusion lied.

You went on to say, "a simple peak overload indicator in conjunction with a loudness meter will be our best bet for standardization."

When in a previous post you said "Even with an accurate loudness meter it still wouldn't tell you (much)." These two statements seem to contradict each other.

To me, a measurement of loudness is the difference between peak and rms, so it would be hard to have one without the other(loudness meter, rms meter and peak meter)no?

Now I understand your confusion was because I forgot to tell you that the loudness meter has to be calibrated relative to full scale (0 dBFS). So, for example, a loudness level of -23 LKFS (the new loudness weighting and metering standard proposed by the EBU) will implicitly have a certain loudness level of "circa -23 dB" relative to full scale peak level. So that then we can hide the entirely-misleading peak meter, people will not think that every program has to have its peaks hitting full scale, and all we need is a simple peak overload indicator "just in case". With a -23 LKFS loudness level, however, it will be highly unlikely that any peaks will exceed full scale and likely that a peak limiter will provide some inaudible protection.

[jamescater]

My 02cents for what its worth

Pro mixers have years of experience and and their mixes are so balanced across the frequency spectrum that the RMS is high with fewer peaks.

I have found that my mixes can only be pushed so far and this is generally 2db less than the biggest hits.

As was mentioned beforfe I find it pointless to measure the RMS of the entire song, becuase your song may have bigger breakdowns of longer intros etc.
I measure the RMS in the second or last chorus over at least 4 bars.

On this basis I have found that the big hits usually vary between -7 and -6 RMS.
As Mr Katz has clearly stated this does not necessarily equate to the same loudness by ear, but it does give you a commn theme.

As a result I ty to aim for -8db RMS for my mixes. I do start to ask qustions if my mixes if have such high peaks that this is not possible.

On a good day my mixes will give me the option of pushing to -7db RMS, but I have yet to try for -6, so for the time being -8 is my standard.

[princeplanet]

I seem to be asked to get -7.5 as this competes with the big releases. They lose their hard-on if it's even .5db less than "XYZ"... Mind you, it seems most mastering folk either cannot or will not achieve this. I religiously A/B volume compensated to see at what point the deterioration is just noticeable and then back off .3. Depending on LF, sometimes the loudest my tools allow will vary +/- 1.5dB.
When I say "noticeable", I should say noticeable to the typical client for a short burst only. Obviously, as soon as it gets under -12.0 I notice the subtle artifacts, particularly over a longer listen. But I can honestly say that I notice less artifacts at -7.5 than many commercial releases. A trend I think I notice is that many of the ME's that do the bulk of their compression and at least some of their limiting in analog come off more compromised than a simple ITB limit/clip approach. The in demand ME's of course have got it down, but if they're in a hurry, the ol' cookie cutter often renders a passable but flawed result.

Getting it "as loud as possible" (which we all hate to have to do) takes a lot of listening. This adds quite a bit of time to the process. These days I think anything less than an hour per song is cookie cutting. The idea of home recordists sending their stuff off to NY to get their album rolled in 3 hours is stupid beyond belief. A lot of the high end pros will not like reading this, but their ya go. I mean, Even the perfectly mixed albums by the pros need more time than that, a day or 2 is not unheard of....

[Kayo]

Originally Posted by jamescater

As a result I ty to aim for -8db RMS for my mixes. I do start to ask qustions if my mixes if have such high peaks that this is not possible.

On a good day my mixes will give me the option of pushing to -7db RMS, but I have yet to try for -6, so for the time being -8 is my standard.

Do you mean raw mixes at -8db RMS!?
I'd find that way too loud, for an incoming mix, for mastering. Not that, it’s not possible to master at -8db RMS. I think it’s just unnecessary for a mix to be everything, all at once. Much prefer a spotless stage by stage transition to top form.

My personal suggestions are, try bring the mixes in at -14db RMS or maybe -12db RMS [raw mixes].
If I’m mixing, I aim for -14db RMS. Nice, clear and pristine and allows for mastering to achieve it's best. No matter what the genre of music it is...

The rationale being… years of experience.

Ciao,
FD

[sanddigger1]

Are you talking about max RMS or average RMS ? If it's average -12 is already too much.