What kind of levels am I suppose to be hitting on my mix (pre master)
- Thread starter chadsxe
- Start date
mshilarious
Banned
chadsxe said:
Peak to RMS of 6dB is extremely compressed. The RMS value sounds about right, are you sure about that peak value
mshilarious said:Peak to RMS of 6dB is extremely compressed. The RMS value sounds about right, are you sure about that peak value
You know now that I think about it does sound weird. My snare has to be hitting harder then -12db because I know that I am using very little compression. Hmmm, I am going to have to check again when I get off of work.
But you say -18db is about correct. This just seemed low to me. What do I know. I guess thats why I am asking.
the inhabitant
WTF is going on here?
chadsxe said:
It's not low at all IMO. I've been experimenting with various levels as I have recently learned a recording levels lesson and I decided to try to keep my RMS at about -20. You have more than enough headroom with 24 bit. It's good to leave a couple bits of space for the mastering engineer.
mshilarious
Banned
chadsxe said:
That entirely depends on what your peak is. Say it's -3dB, then you have 15dB peak-to-RMS, which is a typical number for a mix to hit (and a good amount of headroom too).
Of course it varies a lot, anywhere from 20 to 12 could be normal for a mix, depending on the style of music. 12 is getting pretty hot though, that would be a loud metal mix or something like that.
Massive Master
www.massivemastering.com
A 24-bit mix that *PEAKS* at -18dBFS isn't too low.
I shoot to have the "bulk" of the mix around 0dBVU - which would translate to *around* -20dBFS or so. Then the peaks can have at it.
And I know that there are many other factors involved, but some of the better sounding HEAVY music that comes in here (from several engineers) might peak at around -10dBFS tops. VERY quiet - I'm sorry, not "quiet" - Very "NORMAL" stuff.
I shoot to have the "bulk" of the mix around 0dBVU - which would translate to *around* -20dBFS or so. Then the peaks can have at it.
And I know that there are many other factors involved, but some of the better sounding HEAVY music that comes in here (from several engineers) might peak at around -10dBFS tops. VERY quiet - I'm sorry, not "quiet" - Very "NORMAL" stuff.
Massive Master said:
Well thats good to hear it from the your mouth because I think you are going to be the one I go to for mastering.
Now I am really curious to see just were I am peaking.
Another question...
When using the word "head-room" are we talking about the room between RMS and peak or the room between peak and clipping?
mshilarious
Banned
chadsxe said:
Again, without knowing the peak value, I have no idea. More information is required . . .
If the peak really is -12dB, then you have -6dB peak-to-RMS, that is already very very loud. I've heard pro masters that sound good around those levels, but I don't recall too many home mixes up there that don't have serious issues.
Why don't you post your mix in the Clinic and ask for comments?
the inhabitant
WTF is going on here?
Yeah, I try to keep my peaks out of the top two bits. And I don't think that's overkill. I might even end up with my RMS at around -24 when all is said and done. And listen to Massive Master because he knows what he's talking about (not to mention he's a good mastering engineer). When I first learned about proper recording levels he sort of confirmed everything I learned and made it clear.
Think about this: in the old days of tape, engineers kept RMS at around 0 dBVU which left 24 dB of headroom on most consoles, and that was with a high noise floor. So now with 24 bit which has a low noise floor and more than enough resolution, I don't see why -24 dBFS RMS would be overkill at all. I've read that some fast transients can go as high as 20 dB over what the meters show. That stuff might not show up as an obvious audible clip, but it can show up as some slight distortion that may not be very audible, but the ensuing loss of fidelity is.
Think about this: in the old days of tape, engineers kept RMS at around 0 dBVU which left 24 dB of headroom on most consoles, and that was with a high noise floor. So now with 24 bit which has a low noise floor and more than enough resolution, I don't see why -24 dBFS RMS would be overkill at all. I've read that some fast transients can go as high as 20 dB over what the meters show. That stuff might not show up as an obvious audible clip, but it can show up as some slight distortion that may not be very audible, but the ensuing loss of fidelity is.
the inhabitant
WTF is going on here?
You have 6 dB per bit.
SonicAlbert
Super-Sonic "Herb" Albert
Peaking at -12 seems pretty safe to me. But if I read your post correctly, it seems like you have already squeezed a lot of the dynamic range out of the music. The valleys could be lower.
the inhabitant
WTF is going on here?
SonicAlbert said:
Yeah, I haven't put together complete mixes yet with these new levels so I'm not sure if -20 / -24 is where I'll end up at. That was just a guess. I've been experimenting with laying down bass and a couple of distorted guitar tracks (lead and rhythm) which is pretty compressed stuff. I'll do a few that peak at -12 or so and a few at just before clipping and then make a mix out of each group and pull down the fader on the hot mix so I can monitor them both at the same volume just to see if I can hear the difference. And I absolutely can. I'm not sure what the real RMS is; I'm just eyeballing it.
But your point brings up another thing that was wrong with my old mixes. I was trying to make them as loud as my CD's, or at least close. My new mixes are only going to have moderate compression where needed and some of it on seperate tracks that I can blend in under the non-compressed ones. I'm not going to touch the stereo buss at all. So I imagine I will end up settling below -24 if you are right about that. Obviously I'm still learning a lot from people like Massive and from books. I'm having a break-through lately. The album I'm just finishing will have drum tracks that were recorded too hot and I'm going to re-amp any guitar track that was recorded too hot so I think it will sound OK. But any future albums...
I can't wait.
flatfinger
Use every dam bit!
Is'nt looking at the RMS of the entire mix a little misleading? I mean the relationship between amplitude and loudness of that amplitude vary because of the way the ear works , right? An amplitude value of " X" is going to sound louder in the mid frequencies than in the low because of the loudness contours and fletcher- munson.
I think "genre generalizations" are about as close as you can expect to get. Why haven't any of you guys shared you secret preset for this
!!!!
I think "genre generalizations" are about as close as you can expect to get. Why haven't any of you guys shared you secret preset for this
!!!!
SouthSIDE Glen
independentrecording.net
Yes, extremely!flatfinger said:
RMS is a form average over time. The key question is how long of a length of time the RMS measuremet is measuring. The RMS of the last 5 seconds of a song (not counting fade out) can be entiely different from the RMS of the whole song. The RMS of the bridge can be quite different from the RMS of the second chorus depending upon the mix arrangement. The exception is if the sonic density of the song remains basically the same through the entire song.
I really, Really, REALLY wish people would not be so obsessed over RMS numbers. It's just so damn silly.
......then again I also really, Really, REALLY would like to spend a month backpacking in the mountains with Julie Bowen, but that ain't gonna happen anytime soon either. *sigh*
G.
flatfinger
Use every dam bit!
SouthSIDE Glen said:
Ok! as a newbie here I will graciously agree to stop obsessing! ................... Now can some body tell me what the perfect Crest Factor should be for barbershop Rap?????
the inhabitant
WTF is going on here?
Yeah, I think you guys are right. I've been eyeballing my RMS for the sake of discussion but haven't really given it much thought otherwise. I guess it's just a nice way for me to describe what my mixes "look" like. I've never stopped to think that everyone's RMS is probably different for good reasons.
BRIEFCASEMANX
Winner chicken dinner!
can someone explain to me why it matters for the mastering engineer where peaks are hitting as long as it's not distorting? Why is it going to help the mastering engineer to have a mix with lower volume? I never understood this.
I guess I'm wrong but I thought using up more of the available bits would yield a slightly better quality sound for the mastering engineer to work with.
I can see him needing some headroom for adding EQ or whatever he does without it distorting, but -12db peaks? Why would the mastering engineer NEED that much headroom? I'm not trying to be argumentative but trying to learn something new hopefully.
ORRRR are the -12db peaks simply a function of the way it was recorded, instead of actually turning down the master fader to get that much headroom, i.e. the person didn't hit their converters very hard when tracking, and didn't turn the gain up on the channel faders or plugins when mixing and ended up with those levels through a conservative approach.
I guess I'm wrong but I thought using up more of the available bits would yield a slightly better quality sound for the mastering engineer to work with.
I can see him needing some headroom for adding EQ or whatever he does without it distorting, but -12db peaks? Why would the mastering engineer NEED that much headroom? I'm not trying to be argumentative but trying to learn something new hopefully.
ORRRR are the -12db peaks simply a function of the way it was recorded, instead of actually turning down the master fader to get that much headroom, i.e. the person didn't hit their converters very hard when tracking, and didn't turn the gain up on the channel faders or plugins when mixing and ended up with those levels through a conservative approach.