Something I noticed was that the guitar and vocals all kind of sit in the same sonic territory, plus you've got that keyboard effect there too in the chorus that's just not necessary -- all elbowing into each other -- I'd also like the bass and the drummer to decide who's anchoring this - maybe that's a mix thing too. Just general comment was that sonically it was all a big wash that got tiring by the end of the song. Also is there a way to treat the backup vocals differently from the main vocal to give them more distinction or make them more effective? I'll let others give you a more technical opinion -- that was just my general impression on two listens.

Good song. Bass and groove have a "Rio" by Duran Duran thing going. Song sounds a bit slammed (I've gotten hammered in this forum for the very same thing), so you might want to ease up on your levels going into your 2-bus.
Also, did you multi-take your vocals and blend the two or three tracks? Good job, as it makes the lead vox full and pro sounding.

CrazyLuke, post: 436903, member: 48048 wrote: Song sounds a bit slammed (I've gotten hammered in this forum for the very same thing), so you might want to ease up on your levels going into your 2-bus.

One common mistake that people makes is to ask for help on their mix while all they are doing is trying to make it sound like a master.
No professionnal mix are like that. If you squash the dynamics like that and send a mix with -8LUFS, a mastering engineer will refuse to work on your mix because there is no place to work. It's like asking for a fill up at the gas station with an already full tank !

Maybe we could have a ''Comment my master'' section where we'd post the mix and the master together and get help, but in ''Fix this mix'', I considere the songs posted as if they were on the mixing stage. And the mixing stage IS NEVER about loudness. It's about interactive sonics and levels between the instruments.
When sending a mix to the mastering, it is suggested to send it with peaks below -9 to -6 db. In fact you can send a mix at -18db and it will be ok.

I usually put a limiter on the masterbus just to check how my mix will sound with heavy compression but I print the mix without it.
Some will use a compressor on the mix but with a 1 or 2 db of gain reduction only. . . Oh and lowering the master fader before posting or heavy compression at the track level isn't gonna change the result of overcompression...

Another common mistake is to record too hot. if you record 15tracks at -3db peaks, you will have a lot of trouble to not overload the masterbus.
Common recording level is -18db, with some peaks that can go higher but not much. Then the sum of your tracks will be decent and easier to mix.

Oh I forgot, our brain is wired to think louder is better. Everything we do in mixing should be checked at equal level. If you put an eq and for some reason the plugin makes the track louder you will be tricked to think it sounds better. unless we level match and A/B it and may discover it sounds worst than it was..
;)

pcrecord, post: 436923, member: 46460 wrote: Another common mistake is to record too hot. if you record 15tracks at -3db peaks, you will have a lot of trouble to not overload the masterbus.

Yup. Marco's right.

There's just no reason to record "hot" anymore (I'm talking about to the tracks of your DAW, not driving an external Tube or XFO preamp for harmonic distortion - that's a different beast) as the newer bit resolution technology ( 24, 32, 32 float) doesn't require you to track hot in order to gain full bit resolution. There was a time when resolution was dependent upon amplitude, but those days are long gone.

Unless you are intentionally going after bit clipping distortion - and yeah, some guys do like that sound - then keep your tracks at around -16 to -18db LUFS, and your master bus will thank you. ;)

pcrecord, post: 436923, member: 46460 wrote: When sending a mix to the mastering, it is suggested to send it with peaks below -9 to -6 db. In fact you can send a mix at -18db and it will be ok.

When I sent the last album that I engineered out to be mastered ( Thomas W. Bethel ), I sent him mixes with a LUFS average of around -20db (or so) with peaks at around -5db (or so). This enabled him to be able to do what I hired him to do. I gave him lots of room to work with. Some of those mixes had a DR of up to 15 db... and, I did use some limiting on some subgroup buses ( and even on the Master 2-bus as well), but it was light limiting, and used more for sonic "glue" than it was for upping the level of the mix. My job wasn't to worry about the ultimate level - that's what I expected Tom to do - my job was to make those mixes sound as good as possible, with clarity, punch, warmth, presence and silk... and none of those things had anything to do with whether or not I was tracking at -5db RMS on my DAW.

A lot of people newer to the craft think that they need to really get the limiter on a 2-bus hitting and smacking, in order for it to be effective...and nothing could be further from the truth; as most of the time, very light amounts of limiting ( -2db to -4db or so) will be quite effective and more than enough for getting a sonic glue on the stereo bus.

You don't need to slam the meters for a limiter to work. ;)

FWIW
-d.

pcrecord, post: 436923, member: 46460 wrote:
Another common mistake is to record too hot. if you record 15tracks at -3db peaks, you will have a lot of trouble to not overload the masterbus.
Common recording level is -18db, with some peaks that can go higher but not much. Then the sum of your tracks will be decent and easier to mix.

this is definitely the case, ive only just discovered how low i should be tracking, i habitually pushed inputs up to -0.1 db peaks.
this will not be the case on new recordings i can tell you that, i've heard to track between -12 and -18 from most people.

DonnyThompson, post: 436925, member: 46114 wrote: Yup. Marco's right.

There's just no reason to record "hot" anymore (I'm talking about to the tracks of your DAW, not driving an external Tube or XFO preamp for harmonic distortion - that's a different beast) as the newer bit resolution technology ( 24, 32, 32 float) doesn't require you to track hot in order to gain full bit resolution. There was a time when resolution was dependent upon amplitude, but those days are long gone.

yes i definitely had the mind set that maxing the inputs results in more information to work with in the track.

DonnyThompson, post: 436925, member: 46114 wrote: A lot of people newer to the craft think that they need to really get the limiter on a 2-bus hitting and smacking, in order for it to be effective...and nothing could be further from the truth; as most of the time, very light amounts of limiting ( -2db to -4db or so) will be quite effective and more than enough for getting a sonic glue on the stereo bus.

i also do have a limiter on the master, a Waves L2 at about -6db, which is definitely compounding the issue.
i've pulled back on all the tracks to at least give a bit more headroom going into the master bus, in lieu of rerecording everything

CrazyLuke, post: 436903, member: 48048 wrote: Good song. Bass and groove have a "Rio" by Duran Duran thing going. Song sounds a bit slammed (I've gotten hammered in this forum for the very same thing), so you might want to ease up on your levels going into your 2-bus.
Also, did you multi-take your vocals and blend the two or three tracks? Good job, as it makes the lead vox full and pro sounding.

yeah it is really slammed, hot recording, and a limiter on the master bus.
yes i have multi-tracked the vocals, 4 layers including an octave on the verse, and 6 + a second voice harmony in the chorus
thanks for the compliments, at least i'm getting a few things right ;)

This is the perfect thread for me as I'm moving forward on tracking a really big project (the scratch tracks are done) and the idea of proper gain staging is not only a must for getting the best finished project but also facilitates recording workflow much better. With my scratch tracks (all recorded in piecemeal at different db levels) I had to use the the volume trim envelope to make the parts work together.

So if I understand it correctly -- I've my master fader on the each track --- the zero db lvl is the unity-gain position. Until now I've generally recorded with the fader in this position and adjusted my input levels to not peak beyond zero. From what I've read this is already "too hot" in some opinions. There also seems to be some discussion about average db levels, and max peak levels..... from what I'm hearing from you all are you suggesting keeping the max peaks themselves down around -8 to -12? and the average in the -18 range ---- is it that specific? To accomplish this my track fader would probably need to be set well below 0db --- maybe in the -12 range?

Ok - so if I do this and I track acoustic guitars using two mics -- a dynamic and a condenser -- well the condenser is way louder. I could set the master fader on each track appropriately so that the two elements blend but there might not be enough room there to do that --- so I could also stagger the input levels on my interface --- question there is: is there a too hot level there too? Like I've always avoided going beyond 75% on the dial (say the 2:00 or 3:00 clock position on the dial). I keep an eye out for clipping on the interface (the red light) and make sure I have none. Should I be going as hot as I can on the interface without getting a red clip light? Does the input level here limit or enhance the headroom that is available in the DAW for mixing and mastering? As far as workflow goes -- do I basically adjust all track faders to a starting position then turn the input level up until my peaks are in the desired range -- and that's the input lvl?

What happens when I get to my 2 channel master fader? --- if that's low I literally am not going to be able to hear anything when tracking or mixing. I guess while tracking I can have that as hot as I want because that's only relevant when I'm mixing and rendering to a 2 track master. I know if Everything is too low input wise my master fader is going to be pushed pretty high (beyond the 0db position) just to be heard ----

And how do I identify my peaks and my averages? So far I've been using the meter bars which I think on the master bus show peaks, but averages? How is this best monitored or checked.

And if I understand the process -- I basically mix my track and render it at super-low db lvl --- providing my mixing environment has enough amplification for that ---- and then for mastering the two channel rendered mix is sent to someone who then adds a bunch of gain to it (amongst) other things and basically returns a mastered track that is pumped up after the fact?

This would suggest there are two pathways towards a master ----- basically a mix master where these treatments are being applied to the stems and the master bus during the mix process ------- and then a different process where the mix is preserved (and untouched) and only mastered after the fact with additional gain and other tricks by the masterer. It would appear to me that a mastering person may want (or need) to get deeper into individual tracks --- so I guess there's the possibility of providing the masterer with rendered stems ----- and a reference mix? It strikes me that if this is done then the mastering person is actually mix/mastering -- and that mixing before this is more about the rendering of a specific track and providing a suggestion to the next guy in the chain with a reference mix...

So am I getting this right? I guess there's something more macro happening in a second stage mastering process as done on a 2 track master of a mix that is itself separate from mixing.

DogsoverLava, post: 436965, member: 48175 wrote: And how do I identify my peaks and my averages? So far I've been using the meter bars which I think on the master bus show peaks, but averages? How is this best monitored or checked.

A metering plug-in such as Izotope Insight will have an array of different metering options to many different country standards, from memory there are about 12 options.
there is an average LUFS and db range meter on most. You can set it to the level of your choice as well. I use it and find it does a great job.

DogsoverLava, post: 436965, member: 48175 wrote: This would suggest there are two pathways towards a master ----- basically a mix master where these treatments are being applied to the stems and the master bus during the mix process ------- and then a different process where the mix is preserved (and untouched) and only mastered after the fact with additional gain and other tricks by the masterer. It would appear to me that a mastering person may want (or need) to get deeper into individual tracks --- so I guess there's the possibility of providing the masterer with rendered stems ----- and a reference mix? It strikes me that if this is done then the mastering person is actually mix/mastering -- and that mixing before this is more about the rendering of a specific track and providing a suggestion to the next guy in the chain with a reference mix...

There is two approaches, stereo mastering, which is the most common approach where the mastering engineer masters the 2 track stereo .wav file, then there is stem mastering, which is a good approach if you want the mastering engineer to have more control over the entire track, but this is more time consuming and hence more costly. With stem mastering the tracks are bounced down into groups (say, drum & bass, or guitars, or vocals grouped individually) as mixed by the mix engineer and once things are sounding as they should those stem groups are individually bounced down seperately.

Another option for the mastering engineer is to have a copy of the work session saved (if you both are using the same DAW program) as they then can make any adjustments to the levels should things need tweeking.

Its fair to say that it really is up to the mastering engineer once he has heard your track to then decide what path to take to acheive whats best for the final result.

Recording at -18db is a common practice that not only help to not overload the masterbus but most plugins are tuned to that kind of gain to maximise their quality of rendering. Also, many consumer and prosummer preamps won't sound good if pushed too much.. they mostly have a sweet spot much lower than we think.

-18db is not a rule .. not at all. It could be lower or a bit higher.
A nice exercice to do is to adjust all the gains so it sounds like a premix and when you playback the song, it doesn't overload the masterbus..
For exemple, if I record a 15 tracks band, it may be -15db but if I record a 32 tracks band, it could be lower like -24db.
Altought before you record an important project, you should make some tests an discover how your preamps and interface reacts at different volumes.
You don't want to be so low that when put at the right level you get a lot of noise.

Another aspect to investigate is the live monitoring levels. If like me the musicians play with headphones and they are powered by your audio interface, you can't put the gains very low because the headphones may not get enough levels in result...

OK -- so what I've done is go into a project I have been working on that requires I think pretty big ranges and adjust or retrack the scratch tracks so that with the faders set at the zero position all levels were peaking at around -18. Leaving my Master similarly set I've rendered 5 minutes of the song that covers the range of tracks/volumes. It would appear to me that the master peaks somewhere around -8? So with a basic mix (and with just my scratch tracks) can I ask you if what I have here is deliverable to a masterer (assuming that I wanted to do this ?) Have I done this right with respect to gain staging --- if yes - then what happens next? It gets punched up? Did I do anything wrong? If my Master is still a little hot can I just drop the master fader a bit?

Here's the track -- remember that it's still basically scratch tracks to test gain staging setup in my DAW so that I finish with something that is good and pro in terms of moving on to the next step (mastering). I'm trying to get my ducks in a row before tracking the whole project and finding out I screwed myself. Dropbox link the the wav file below (because I'm assuming I would deliver the wav to a mastering engineer). Any opinions on the gain structure/staging would be greatly appreciated. Did I do this right? (the mix itself is very quick and basic)

https://www.dropbox.com/s/syp6t3s2t3xll3t/Stairway-gain-stage-test.wav?dl=0

The level sounds good to me...plenty of headroom and good dynamic range IMO.

Overall it sounds pretty good to me on first listen...I'll have another listen tomorrow.

Thanks - my question is basically: if this was my project mix - is it within the guidelines for mastering? . I have another separate mix of completely different scratch tracks recorded without the gain staging I've done here - I may post that as well to confirm that it would not be appropriate for mastering so there's a compare contrast....

Without running through Izotope Insight to check the LUFS level and going by ear, it sounds OK to me.

Maybe we can ask one of our resident mastering engineers such as Thomas W. Bethel for his expert opinion.

Thomas has done some mastering work for DonnyThompson and Donny speaks very highly of his work.

Donny would also be a good ear (pardon the pun) to advise you on what levels he has used in the past prior to sending work for mastering.(y)

It seems like there's alot of upper mids going on.. Especially around 2 k.. What kind of speakers are you using to monitor? What type of reference mixes do you use? Try pulling out some 2k with a gentle Q.. It will take the edge off and give the illusion that you warmed it up.. The advice from others of not slamming your 2-bus will help alot as well.

Thanks Chris -- this is not a real mix (it's just an exercise in setting gain staging at the recording point). These are just scratch tracks as I prepare to record a N4N version of this - I'm just asking that we pretend the mix and tracks were great - does this fall into the acceptable parameters with respect to gain and recording levels for a mastering engineer to master.

I basically made sure all my pre fader input levels on the vsts were inputting at peaks of -18ish, and my recording levels from my mics were similarly staged (I actually had to boost the dynamic mic level about 8 db and the condenser by about 5 or 6) to fall in the -18 range with all faders in the zero position... I then did a quick mix and rendered the master bus.... So basically pretending that the tracks and mix is amazing - the question is is this result acceptable for mastering -- and should this be my workflow with respect to tracking this project for good. (I'm still several weeks away from completing that process).

I was responding to the OP and the song he posted.. as far as sending material to a mastering engineer or making sure you have dynamic range in your music. I like the Bob Katz idea of setting up your monitors when mixing to be 82 db or so running pink noise at -20 db. That way when you mix you have 20 db of headroom. You need to be careful though because when set up like that you can't play anything else though it like a mastered cd or stuff on the internet as you will blow your brains out with monitors set to have 20 db headroom. when mixing that way you guarantee dynamic headroom and a mastering engineer should have no issues with receiving at that volume.

Chris Perra, post: 436985, member: 48232 wrote: I was responding to the OP and the song he posted.. as far as sending material to a mastering engineer or making sure you have dynamic range in your music. I like the Bob Katz idea of setting up your monitors when mixing to be 82 db or so running pink noise at -20 db. That way when you mix you have 20 db of headroom. You need to be careful though because when set up like that you can't play anything else though it like a mastered cd or stuff on the internet as you will blow your brains out with monitors set to have 20 db headroom. when mixing that way you guarantee dynamic headroom and a mastering engineer should have no issues with receiving at that volume.

Right -- my bad -- and sorry to have hijacked the thread --- back to the OP --- his mix is considerably loud -- is it a mix, or a mastered mix -- or a mix/master... like is it done? How do you conceive of it?

My guess it's a limiter on the main bus and either the faders are driven up against the limiter or its a combination of bus compression and level boost that causes that mix level. Listening with 20 db headroom would guarantee that could never happen. You would go deaf. I would guess the OP was comparing his mix to a mastered cd at a much lower listening volume that 20 db headroom which makes you want to crank everything up.

Sean G, post: 436979, member: 49362 wrote: Without running through Izotope Insight to check the LUFS level and going by ear, it sounds OK to me.

Maybe we can ask one of our resident mastering engineers such as Thomas W. Bethel for his expert opinion.

Thomas has done some mastering work for DonnyThompson and Donny speaks very highly of his work.

Donny would also be a good ear (pardon the pun) to advise you on what levels he has used in the past prior to sending work for mastering.(y)

You'd have to ask Thomas ( Thomas W. Bethel ) what his preference is, perhaps depending on the style.
I can say that for the last project I sent to Tom, I mixed at a LUFS level of around -22db, with peaks at around -6 db or so, ( but never hotter than that).

DR is very important to me when I mix; Tom was happy to hear me say that it was.

For accurate LUFS metering, I use this: https://www.klangfreund.com/lufsmeter/

It's important to note that there are some meters in some DAW's and plugs that are not accurate ( currently) , although many manufacturers are now fixing them.

As of this writing, here is a list of those that are inaccurate, followed by those that are accurate...

These are the meters that are currently inaccurate:

Slate FG-X
Logic 9 and Logic X (Take care – the multimeter displays peak RMS by default, which muddies the water even further…)
Waves PAZ - 4dB off (?!)

These are meters that are accurate:

Wavelab
TT Meter
Ozone 5
Waves Dorrough
Voxengo SPAN
(Actually by default it’s wrong but this can be corrected using the “RMS +3dB” preference)
Reaper is about right, but the time constant needs to be correctly set to 300ms
T-RackS CS Meter (Latest version)
Ableton (Version 9.5)
Waves WLM

source: http://productionadvice.co.uk/lufs-dbfs-rms/

Ian Shepard ( who's site I linked to above as the source for the lists ), has a great deal of knowledge, especially when it comes to mixing and mastering for the preservation of the dynamic range. I've learned a lot from him. ;)

FWIW,
-d.

Chris Perra, post: 436987, member: 48232 wrote: My guess it's a limiter on the main bus and either the faders are driven up against the limiter or its a combination of bus compression and level boost that causes that mix level. Listening with 20 db headroom would guarantee that could never happen. You would go deaf. I would guess the OP was comparing his mix to a mastered cd at a much lower listening volume that 20 db headroom which makes you want to crank everything up.

Pretty much, limiter on the master bus tracks, and comparing to mastered commercial MP3's. I've also recorded the tracks way too hot. I've since brought all tracks down to reduce how hard they are driving the bus, and ive eased up on the limiter considerably.

Chris Perra, post: 436981, member: 48232 wrote: It seems like there's alot of upper mids going on.. Especially around 2 k.. What kind of speakers are you using to monitor? What type of reference mixes do you use? Try pulling out some 2k with a gentle Q.. It will take the edge off and give the illusion that you warmed it up.. The advice from others of not slamming your 2-bus will help alot as well.

thanks, i'll give that a go, should make it less harsh sounding? I use KRK monitors.