Just a quick question about gain staging and what levels people set there kick drums and bass to.
I've read that it's best to set your kick at -8db and then your sub bass to -10db, mixes seem to come out at -6db on the master which is perfect but I find they're lacking in top end energy and have a bit to much bass.
I've also worked with the kick peaking at -17db and the bass at -24db, which seems to give a nice balanced mix, but I'm not really sure why!.
Wondering if anyone has any rules they use in terms of starting levels, or good tips for gain staging?
Gain Staging
Re: Gain Staging
I don't understand this completely either but i always get better results at -18db. Depending on how your sound card is calibrated -18dbfs in the box = 0VU
http://en.wikipedia.org/wiki/DBFS
but what i definitely do not understand is why tracks get mastered to 0dbfs. Cause in this cause wouldn't there be all kinds of digital clipping? ugh... i hate this subject, so i generally just do what i like.
http://en.wikipedia.org/wiki/DBFS
but what i definitely do not understand is why tracks get mastered to 0dbfs. Cause in this cause wouldn't there be all kinds of digital clipping? ugh... i hate this subject, so i generally just do what i like.
------------------------------------------------------
http://soundcloud.com/kirkwoodwest
http://soundcloud.com/kirkwoodwest
Re: Gain Staging
from wiki:
"A digital signal which does not contain any samples at 0 dBFS can still clip when converted to analog due to the signal reconstruction process. This possibility can be prevented by careful digital-to-analog converter circuit design.[2]"
i just realised when i record with my soundcard (a complete mixdown run through outboard gear) at about -6dbFS it sounds better in the high freq. guess i need to update my soundcard (saffire dsp 24). which supports the theory i read on gearslutz not to record too hot into semipro converters.
on the other hand i keep my levels at about -6db ITB. but to achieve that there is no rule how at what db the bassdrum and bass sits. just trust your ears!
i also mix into a comp ITB a lot at the moment. just like the sound of it.
http://en.wikipedia.org/wiki/Full_scale
"A digital signal which does not contain any samples at 0 dBFS can still clip when converted to analog due to the signal reconstruction process. This possibility can be prevented by careful digital-to-analog converter circuit design.[2]"
i just realised when i record with my soundcard (a complete mixdown run through outboard gear) at about -6dbFS it sounds better in the high freq. guess i need to update my soundcard (saffire dsp 24). which supports the theory i read on gearslutz not to record too hot into semipro converters.
on the other hand i keep my levels at about -6db ITB. but to achieve that there is no rule how at what db the bassdrum and bass sits. just trust your ears!
i also mix into a comp ITB a lot at the moment. just like the sound of it.
http://en.wikipedia.org/wiki/Full_scale
techno made me do it
-
NoAffiliation
- mnml mmbr
- Posts: 253
- Joined: Sat Jul 08, 2006 4:38 am
Re: Gain Staging
@ the OP
couldn't resist this topic...
gain staging is important but you are missing a few things. one is that the numbers don't correspond to actual loudness of a signal. we could take two kicks both gain staged to -18 that could have a fairly drastic difference in perceptual loudness. so if you read that someone says put the kick at -8 or or whatever but your mix is unbalanced when you do that, it's most likely because you haven't treated the sounds the same way the person who wrote that if they in fact got a good result.
the main idea behind gain staging is having a close to equal amount of prefader energy between all the tracks. engineers back in the day would watch VU meters and use the trim on the console to get all the tracks sitting about average -18. but see in those days before the computers, the engineer was tracking the instruments through the eq's and compressors and then set the prefader trim as close to 0vu average as possible.
in a computer workflow gainstaging is not as simple as just using a trim plug first in the chain. the point is to rebalance the signal to it's original energy level after each process. so if i put some heavy compression on a kick drum to get it to fit the mix i reblance the signal back to what the original recording was...
i use a plugin called Satson which has two options for the metering, -18 or -14 calibration. Bob katz suggests K-14, -14 = 0 for a final mixdown but this is a little weak for todays techno market. most releases that i like are around -11ish with a 6 crest factor.
just remember that the numbers dont equal loudness, thats why gain staging is handy because you can directly compare the sound of a compressed kick with the dry version by switching on and off the trim plugin so you can actually judge if it's louder or not. all of the plugins in any chain are pre fader so you want the gains at the end of the line to be even. you will see quickly how the levels of different frequency ranges react on the meter. a kick will hit -18 alot easier and quicker that some high hats.
for loudness if you check the fletcher munson curve and think about the average spl of a club system, you have about 35 db of spl between the bass and the high end. +6 db (dont confuse with spl) is about double perceptual volume so you can use the fletcher munson curve in accordance to how loud you think your song will be played and find the levels.
however if you do good per track compression and eq and gain staging you should be able to find the mix by ear just as easily if the signals are balanced
couldn't resist this topic...
gain staging is important but you are missing a few things. one is that the numbers don't correspond to actual loudness of a signal. we could take two kicks both gain staged to -18 that could have a fairly drastic difference in perceptual loudness. so if you read that someone says put the kick at -8 or or whatever but your mix is unbalanced when you do that, it's most likely because you haven't treated the sounds the same way the person who wrote that if they in fact got a good result.
the main idea behind gain staging is having a close to equal amount of prefader energy between all the tracks. engineers back in the day would watch VU meters and use the trim on the console to get all the tracks sitting about average -18. but see in those days before the computers, the engineer was tracking the instruments through the eq's and compressors and then set the prefader trim as close to 0vu average as possible.
in a computer workflow gainstaging is not as simple as just using a trim plug first in the chain. the point is to rebalance the signal to it's original energy level after each process. so if i put some heavy compression on a kick drum to get it to fit the mix i reblance the signal back to what the original recording was...
i use a plugin called Satson which has two options for the metering, -18 or -14 calibration. Bob katz suggests K-14, -14 = 0 for a final mixdown but this is a little weak for todays techno market. most releases that i like are around -11ish with a 6 crest factor.
just remember that the numbers dont equal loudness, thats why gain staging is handy because you can directly compare the sound of a compressed kick with the dry version by switching on and off the trim plugin so you can actually judge if it's louder or not. all of the plugins in any chain are pre fader so you want the gains at the end of the line to be even. you will see quickly how the levels of different frequency ranges react on the meter. a kick will hit -18 alot easier and quicker that some high hats.
for loudness if you check the fletcher munson curve and think about the average spl of a club system, you have about 35 db of spl between the bass and the high end. +6 db (dont confuse with spl) is about double perceptual volume so you can use the fletcher munson curve in accordance to how loud you think your song will be played and find the levels.
however if you do good per track compression and eq and gain staging you should be able to find the mix by ear just as easily if the signals are balanced
Re: Gain Staging
i use my ears, i only ever use the VU reading as a very rough guide.
in an analogue mixer i work at very roughly +6 dB for both bass and kick.
alot depends on how you write too, i tend to use my Kick as a bassnote, with only release tails of preceeding bass notes harmonically layered, i'm not a tiny clicky kick or side chain guy.
i just look for wayward spikes on the mixdown waveform, and reduce them manually in Soundforge ( theres not usually many because i mix carefully )
i usually end up with an RMS value of -12 dB measured at the loudest part of the track.
i agree with No Affiliation that -14 is too low, -11 is too squashed for me
the 1 dB difference is just a tiny tweek on the gain knob on the DJ mixer.
dont take everything Bob Katz says as Gospel, this is Techno.
in an analogue mixer i work at very roughly +6 dB for both bass and kick.
alot depends on how you write too, i tend to use my Kick as a bassnote, with only release tails of preceeding bass notes harmonically layered, i'm not a tiny clicky kick or side chain guy.
i just look for wayward spikes on the mixdown waveform, and reduce them manually in Soundforge ( theres not usually many because i mix carefully )
i usually end up with an RMS value of -12 dB measured at the loudest part of the track.
i agree with No Affiliation that -14 is too low, -11 is too squashed for me
the 1 dB difference is just a tiny tweek on the gain knob on the DJ mixer.
dont take everything Bob Katz says as Gospel, this is Techno.
-
NoAffiliation
- mnml mmbr
- Posts: 253
- Joined: Sat Jul 08, 2006 4:38 am
Re: Gain Staging
-11 is a bit flat, move d's stuff usually rides around -12 - -11.5 and i use him a reference
i also agree that if you are in dynamic control the mix can be done by ear. there is something to be said that engineers in the past were able to use the meters to create consistently loud tracks across an album without any computers, so the meter has value in that sense of consistency, especially without today's tools
but it's actually best to do it by ear since creating a loud mix isn't all about the numbers, the numbers can mislead people, i know i used to get tricked all the time and realized my bass was showing the tracks to register at the right RMS but the result was still not as loud as needed.
a little tip i made for myself was to reference a track i know sounds good and check out the crest (difference between peak and average) factors of the individual frequency ranges by placing an eq before voxengo SPAN and filtering so i could reference specific ranges. one thing that's obvious with most music today in the low end has 3-4 less db of dynamic range than higher end signals
i also agree that if you are in dynamic control the mix can be done by ear. there is something to be said that engineers in the past were able to use the meters to create consistently loud tracks across an album without any computers, so the meter has value in that sense of consistency, especially without today's tools
but it's actually best to do it by ear since creating a loud mix isn't all about the numbers, the numbers can mislead people, i know i used to get tricked all the time and realized my bass was showing the tracks to register at the right RMS but the result was still not as loud as needed.
a little tip i made for myself was to reference a track i know sounds good and check out the crest (difference between peak and average) factors of the individual frequency ranges by placing an eq before voxengo SPAN and filtering so i could reference specific ranges. one thing that's obvious with most music today in the low end has 3-4 less db of dynamic range than higher end signals
Re: Gain Staging
Nice topic, always interested to see what sorta levels everyone else gets.
I have worked to numbers a bit before and my tracks have been a lot easier to mix.
Now I don't really care too much about level's, as long as the master isn't peaking, and my recording signals arn't too weak then life is good.
I have been a bit obsessed with steevio's mention of overdriving pre's recently. Have been doing it quite a lot quite enjoying not giving a sh!t about levels. After coming from a college that wanted us to all have a firm grasp of the theory > PFL > Add gain to 0dB > PFL off > yawn....
It's quite refreshing to be 'feeling the music'
I just wish I could mixdown my tracks!
I have worked to numbers a bit before and my tracks have been a lot easier to mix.
Now I don't really care too much about level's, as long as the master isn't peaking, and my recording signals arn't too weak then life is good.
I have been a bit obsessed with steevio's mention of overdriving pre's recently. Have been doing it quite a lot quite enjoying not giving a sh!t about levels. After coming from a college that wanted us to all have a firm grasp of the theory > PFL > Add gain to 0dB > PFL off > yawn....
It's quite refreshing to be 'feeling the music'
I just wish I could mixdown my tracks!
Re: Gain Staging
analogue mixing is a different process really, you can do what you like with an anlogue mixer, its only when it goes back inside the box you've got to watch your recording levels.lem wrote:
I have been a bit obsessed with steevio's mention of overdriving pre's recently. Have been doing it quite a lot quite enjoying not giving a sh!t about levels. After coming from a college that wanted us to all have a firm grasp of the theory > PFL > Add gain to 0dB > PFL off > yawn....
It's quite refreshing to be 'feeling the music'
I just wish I could mixdown my tracks!
if you've got an analogue mixer, forget your digital recording techniques.
absolutely everything on my mixer is above 0 dB, channels, busses, masters. usually between 0 and +9dB
this also means you get much less noise. if you use digital recording levels in an analogue mixer, you'll get a hiss-fest.
of course you need a hot signal from your kit, or the opposite will happen. you want to avoid a low signal from your gear being boosted by the gain on the channels, you want maximum output levels on your synths/drum machines.