Serif Sound
New member
Big Money Dilly said:
Where should the mix be peaking during/after mix process?
- Thread starter Big Money Dilly
- Start date
Big Money Dilly
New member
Seriously, when i say -2.8 i dont know if im talking dbFS. I have been asking a few times what this is. Like i said. im using cubaseSX3 and the meter im looking at was hitting -2.8. Can someone tell me if that is dbFS or dbVU or what lol (once i get that i can start making sence of the numbers im reading)
Also. just to make positively sure, Massive -
(a) you said -14dbFS is that peak ur talking or average levels?
(b)And you mean to say record at this level instead of recording higher and then turning it down manually to -14 right?
Cheers guys, this is starting to get a bit clearer.
Also. just to make positively sure, Massive -
(a) you said -14dbFS is that peak ur talking or average levels?
(b)And you mean to say record at this level instead of recording higher and then turning it down manually to -14 right?
Cheers guys, this is starting to get a bit clearer.
Serif Sound
New member
Big Money Dilly said:
N
neilwight
Guest
big money,
everything inside the digital realm is DBFS. it stands for decibels full scale which has 0dbfs as the maximum ceiling level.
dbv, dbu etc all relate to the physical world outside the digital realm.
regarding the dbfs levels mentioned in the thread, these will all be peak levels as this is generally what DAW meters are set to measure (though some can be set to measure RMS if wanted). RMS is of no usefulness when mixing or infact when doing much at all. theres no situation i can think of when anyone would actually need to know the RMS level of a track or stem.
everything inside the digital realm is DBFS. it stands for decibels full scale which has 0dbfs as the maximum ceiling level.
dbv, dbu etc all relate to the physical world outside the digital realm.
regarding the dbfs levels mentioned in the thread, these will all be peak levels as this is generally what DAW meters are set to measure (though some can be set to measure RMS if wanted). RMS is of no usefulness when mixing or infact when doing much at all. theres no situation i can think of when anyone would actually need to know the RMS level of a track or stem.
Last edited by a moderator:
ZeroLatency
New member
In Cubase the meters read dBFs. All digital equipment uses that, unless it has been designed to look like it uses dBVu.
dBFs and dBVu are different in many ways. dBFs refers to the digital domain and dBVu to analog.
0dBFs is much as you can crank it before it starts clipping, and goes down in negative numbers from that. The meter is usually represented by LED type bar meters.
dBVu meters have needles that bounce around. They have a scale usually from -20dBVu to +5dBVu and are used mainly in analog recording, because in the analog realm you can record over 0dB.
With dBVu meters the audio has to move a physical object (the needle meter), so it cannot accurately gauge a lot of transients. Example: If you have a drum spiking at 0Vu most likely you are recording around +9Vu. You can't do this with digital equipment. You have to use dBFs, which accurately reads the transients so you don't clip.
So if someone is talking about analog peaking at -4dBVu it translates to about an average of -14dBFs probably with peaks around -6dBFs in the digital domain.
I beleive in the digi world the average should stay around -14dBFs. You will have peaks that register around -6bBFs.
dBFs and dBVu are different in many ways. dBFs refers to the digital domain and dBVu to analog.
0dBFs is much as you can crank it before it starts clipping, and goes down in negative numbers from that. The meter is usually represented by LED type bar meters.
dBVu meters have needles that bounce around. They have a scale usually from -20dBVu to +5dBVu and are used mainly in analog recording, because in the analog realm you can record over 0dB.
With dBVu meters the audio has to move a physical object (the needle meter), so it cannot accurately gauge a lot of transients. Example: If you have a drum spiking at 0Vu most likely you are recording around +9Vu. You can't do this with digital equipment. You have to use dBFs, which accurately reads the transients so you don't clip.
So if someone is talking about analog peaking at -4dBVu it translates to about an average of -14dBFs probably with peaks around -6dBFs in the digital domain.
I beleive in the digi world the average should stay around -14dBFs. You will have peaks that register around -6bBFs.
Big Money Dilly
New member
Yes i meant that the master mix on the master channel 'peaked' at -2.8dbFS (hurray i understand it
). I didnt check the avarage levels.That helps too. That means i can have the attacks of my instruments going through as usual without having to compress them too much now, cos ill be worrying about the avarage levels more now.
OK so now i shall record my synth and vsti's so there average levels are all about -14dbFS with peaks rising to whatever (but not over 0dbFS). After recording at these average levels, i can then adjust the volume faders of each track so that the overall mix is created. Then in the end i can boost the overall level up to just under 0dbFS. Thats what i have gathered!
Thank you for your help guys. i will have a go at this sometime over the next few days, hopefully the weekend! Cheers!
ZeroLatency
New member
Adjust the Out Ceiling on your limiter to -0.2dBFs. Sometimes clipping can accure at -0.1dBFs.
Last edited:
Serif Sound
New member
Big Money Dilly said:
of the peaks (transients)
...up to just under 0dbFS. Thats what i have gathered![/QUOTE]
,,,and "Bob's your uncle!" ( :
_andrew
N
neilwight
Guest
zerolatency,
im not sue i agree with much in your two posts apart from the ensure your peaks are at -0.2dbfs.
clipping can occur in low quality converters below 0dbfs so ensuring you arent quite touching this is good practice, especially when making CDs.
myself and most MEs i know set at -0.5dbfs for best practice.
following on from my other post which no one appeared to read.
the -14dbfs was refering to peak value and NOT RMS.
using RMS has no practical use in mixing environments (or in much else either).
you would record into yur DAW with peak values registering about -14dbfs (in 24bit) on each track. this ensures a good signal, headroom etc. it makes mixing easier and helps maintain quality.
once you have finished mixing, the summing of these peaks from various tracks will give a reading on the 2buss (main output) of greater than -14dbfs but still well below the 0dbfs ceiling.
you then render this mixdown for mastering and send it off leaving all the headroom intact (it could be anything from -6dbfs and lower). mastering will eat this at the last stage.
if you are self mastering, then you would just import these final mixdowns with bags of headroom present into your mastering DAW and then process accordingly, eating up the headroom at the end until peaks were reading -0.5dbfs.
at no time should you ever reference anything using RMS, its meaningless in application. its an approximation of average values and is subject to many variables. it does not relate specifically to perceived volume levels and indeed a collection of tracks all with the same RMS level will all sound entirely different both in terms of loudness and quality.
many people seem to think that working with or to RMS values will ensure they achieve good results or tracks as loud as commerical releases but this is incorrect. perceived level is more to do with content, dynamics,crest factor sounds, variations between loud and quiet passages etc and much much more.
its very complicated and the only tool able to accurately gauge is the human ear.
how loud your track can go with processing will depend on its content and how well it was mixed along with how much trade off with quality can be accepted.
im not sue i agree with much in your two posts apart from the ensure your peaks are at -0.2dbfs.
clipping can occur in low quality converters below 0dbfs so ensuring you arent quite touching this is good practice, especially when making CDs.
myself and most MEs i know set at -0.5dbfs for best practice.
following on from my other post which no one appeared to read.
the -14dbfs was refering to peak value and NOT RMS.
using RMS has no practical use in mixing environments (or in much else either).
you would record into yur DAW with peak values registering about -14dbfs (in 24bit) on each track. this ensures a good signal, headroom etc. it makes mixing easier and helps maintain quality.
once you have finished mixing, the summing of these peaks from various tracks will give a reading on the 2buss (main output) of greater than -14dbfs but still well below the 0dbfs ceiling.
you then render this mixdown for mastering and send it off leaving all the headroom intact (it could be anything from -6dbfs and lower). mastering will eat this at the last stage.
if you are self mastering, then you would just import these final mixdowns with bags of headroom present into your mastering DAW and then process accordingly, eating up the headroom at the end until peaks were reading -0.5dbfs.
at no time should you ever reference anything using RMS, its meaningless in application. its an approximation of average values and is subject to many variables. it does not relate specifically to perceived volume levels and indeed a collection of tracks all with the same RMS level will all sound entirely different both in terms of loudness and quality.
many people seem to think that working with or to RMS values will ensure they achieve good results or tracks as loud as commerical releases but this is incorrect. perceived level is more to do with content, dynamics,crest factor sounds, variations between loud and quiet passages etc and much much more.
its very complicated and the only tool able to accurately gauge is the human ear.
how loud your track can go with processing will depend on its content and how well it was mixed along with how much trade off with quality can be accepted.
Last edited by a moderator:
Big Money Dilly
New member
Hi Neil...so ive been thinking that the -14dbfs was reffering to averaga values. so if u have a plucky instrument the pluck might hit -8dbfs and the average being at -14dbFS. Am i wrong in thinking this then? So i should be recording each track at -14dbFS at the with average levels below -14dbfs?
(On a side note, if recording should be at -14dbfs, then that means all the samples you use, say drums which usually are recorded quite hot/sampled hot. will need to be turned down quite significantly..isnt this bad for quality if in the end your gonna boost these via mastering?)
btw i hope this thread is helping alot of other people too
Cheers
N
neilwight
Guest
thats exactly right.
peaks of -14dbfs with RMS value way below this depending on what sort of sound.
at the mixing stage you shouldnt be concerned with volume levels against commercial releases (these will have been quiet at the mixing stage too in most cases) and as such if you were to play a track against a commercial one there would be a huge gulf in terms of volume.
you just ignore this at this stage and turn up your amp/monitors to provide an adequite level.
with respect to using a lot of samples. if you record them from vinyl or whatever then you just use the same principle and get a 24bit capture with peaks of around -14dbfs.
if you are using samples already in a digital format then indeed you will need to turn them down alot. using the track gain rather than the fader, which will be kept at unity, will be much better for the mixing stage. in cubase track gain is found on the channel window (clicking the "e" on a track will bring up the main window, look for the gain above the fader, it will have a round knob beside it with a window beside reading 0, turn this down by entering a value in here with a - value).
how this impacts the audio will depend on the samples themselves (are they 24bit wav or MP3), as with everything in music you need to learn/understand and then apply and adjust according to your working methods and uses. that being said turning a few samples down will work much better than turning everything else up.
peaks of -14dbfs with RMS value way below this depending on what sort of sound.
at the mixing stage you shouldnt be concerned with volume levels against commercial releases (these will have been quiet at the mixing stage too in most cases) and as such if you were to play a track against a commercial one there would be a huge gulf in terms of volume.
you just ignore this at this stage and turn up your amp/monitors to provide an adequite level.
with respect to using a lot of samples. if you record them from vinyl or whatever then you just use the same principle and get a 24bit capture with peaks of around -14dbfs.
if you are using samples already in a digital format then indeed you will need to turn them down alot. using the track gain rather than the fader, which will be kept at unity, will be much better for the mixing stage. in cubase track gain is found on the channel window (clicking the "e" on a track will bring up the main window, look for the gain above the fader, it will have a round knob beside it with a window beside reading 0, turn this down by entering a value in here with a - value).
how this impacts the audio will depend on the samples themselves (are they 24bit wav or MP3), as with everything in music you need to learn/understand and then apply and adjust according to your working methods and uses. that being said turning a few samples down will work much better than turning everything else up.