Sending mixes to clients

SeanHurley

How many of y'all do a bit of DIY mastering (eg sticking an L1 on the mix) before sending it to a client for sign off?

I generally leave a fair amount of room for the mastering engineer to do his/her bit. There is no point giving the mastering engineer something squashed to sh1t. However, sending a mix to a client that doesn't keep up in the loudness wars can often result in a "it sounds a bit quiet/wimpy" response.

Any thoughts?

11811

Send them both? A non touched mix for the master engineer to work with and "demo" as such showing roughly what a master might achieve for the track?

DamagedTrax

SeanHurley wrote: »

How many of y'all do a bit of DIY mastering (eg sticking an L1 on the mix) before sending it to a client for sign off?

I generally leave a fair amount of room for the mastering engineer to do his/her bit. There is no point giving the mastering engineer something squashed to sh1t. However, sending a mix to a client that doesn't keep up in the loudness wars can often result in a "it sounds a bit quiet/wimpy" response.

Any thoughts?

i send a 44/24 mix for mastering.
a 44/16 listen master (final eq'ing, limiting, stereo width etc.. whatever it needs to sound as close to a finished master as possible).
an mp3 of the listen master and an instrumental if requested.

if you're mixes arent competing tell the band to turn up the volume

i look at sending listen masters as a good way to delve into mastering and learn about it. like everything else in engineering, the only real way of learning is by doing. as long as you let the client know that your master is only a test it should all be good.

DamagedTrax

heres an example of one of my listen masters from a session last week.

i tried to keep it dynamic but still loud. im no mastering engineer but trying the process gets me a little closer each time.

http://soundcloud.com/damaged_audio/the-shady-boys-bright-boy

PaulBrewer

Never give a Client anything that isn't banging !

I normally run my mix and bounce down one for 'The Master' and a second with a tasty smothering of Massey's Master Limiter which may or may not get MP3ed by Itunes for the client.

There's a good interview with Robert Orton who mixes Lady GaGa in Resolution Magazine a while ago where he say he learned early on that if you give a pun an unloudified mix and someone else gives them a Loud one , taking it they are both 'good' mixes ........ you loose every time !

The other thing to do is make sure it's not a complete mix, miss half the intro or loose the last beat ....... then the rough won't end up on itunes which I have seen happen !

madtheory

Even in the days of cassette we always added loadsa top and compressed it, just as a demo.

krd

SeanHurley wrote: »

How many of y'all do a bit of DIY mastering (eg sticking an L1 on the mix) before sending it to a client for sign off?

I generally leave a fair amount of room for the mastering engineer to do his/her bit. There is no point giving the mastering engineer something squashed to sh1t. However, sending a mix to a client that doesn't keep up in the loudness wars can often result in a "it sounds a bit quiet/wimpy" response.

Any thoughts?

What do you send to the mastering engineer? Multitrack stems or a mixdown?

My D.I.Y mastering is to create a mixdown that sounds fine to me (in Ableton). Then export the wave file, and load it into Goldwave, shareware audio editor. I can see the wave clearly. I then hard limit the spikes away and maximise the volume to 100%. If I go too far the mix sounds burnt. I can see on the original wave how far I can go - if I limit into the region where it starts to get busy, the mix starts to sound bad.

Once you get to the point your limiting is burning the mix - you're at the limit of how loud you can make something. You could use multi band compression and more limiting to fill out the spectrum.

Do you give your clients A) A mixdown Mixdown and master stems - If it's just a mixdown you may as well bang it up to the max. If a mastering eng were to do anything more special than limiting they'd probably need stems. Otherwise there's no real difference in what you could do - with say goldwave.

DamagedTrax

krd wrote: »

If a mastering eng were to do anything more special than limiting they'd probably need stems. Otherwise there's no real difference in what you could do - with say goldwave.

theres a lot more that can be done to a stereo mix than just limiting. you can get right into mid/side for a start...

madtheory

Stems are a myth. I've never used them nor has any mastering engineer ever asked. I never ask for them either. You do the best mix you can, then hand it over. Mastering with your eyes is not very effective. Neither is doing your own mastering- it's important to get an objective pair of ears to polish your mix.

Furthermore, one of the most important functions of mastering is to make a collection of different tracks into a cohesive body of work, like an album. Generally one doesn't master stuff in isolation.

M/S processing is not done very often either, unless there's something wrong with the mix and a recall is not possible.

krd

DamagedTrax wrote: »

theres a lot more that can be done to a stereo mix than just limiting. you can get right into mid/side for a start...

Stereo eq'ing?

Is the mastering engineer expected to redo your eq'ing?

There's lots that can be done with any master. Apply exciterizing effects to the mix, for example. I don't know if that's particularly any good.

krd

madtheory wrote: »

You do the best mix you can, then hand it over. Mastering with your eyes is not very effective.

No, mastering with your eyes is not effective by itself. But you can see some things that it's impossible to hear. In the old days you just had the VU metres - now you can see all the spikes. You see lots of things.

When I see a wave that looks like a solid block - I'm nearly always certain it's going to sound awful - and it usually will. Lots of commercial recordings have been mastered to solid blocks.

So, sometimes your eyes will not deceive you.

Anima

But you can see some things that it's impossible to hear

Hmm I'm struggling to think of anything that you can see but can't hear if I'm honest. Can be easier to spot clipping that might be difficult to hear but I guess it doesn't matter if you can't hear it.

Looking at a waveform in general only gives you the most basic of indications of what the signal might contain.

For instance, there are lots of different sawtooth waveforms that look dissimilar (relatively) but their frequency content is very much the same, just different weighting of the harmonics. So looks can be deceiving in the time domain (waveform shape).

DamagedTrax

krd wrote: »

Is the mastering engineer expected to redo your eq'ing?

mid/side isnt exactly re-doing your eq. the sides may need a touch of sparkle or the mid may need to be brought to the front a little more. a touch of m/s eq or comp can provide the results.

remember, we all have the tools to master, the reason we use a mastering engineer is for a fresh set of ears and experience. if an experienced mastering engineer suggests something, i'll listen. and by that stage the mix is burnt into my ears anyway, so im hardly in a position to make judgements on the final touches.

madtheory

Damaged is spot on. The point of mastering is not to "make something better". That assumes that there's always something wrong! It (should be) just a polish, a nip and tuck to get a bunch of tracks to work together so the listener doesn't have to adjust the volume and eq all the time. If a mix is really bad, one might use M/S processing if a recall is not possible.

Using your eyes is a mistake IMO. Fact is, almost all DAWs do not have an accurate waveform display. It's accurate enough for editing, that's all. For a start, it doesn't show what the reconstruction filter is doing.

The real problem with DAWs is that they allow us to rely on our eyes.

krd

madtheory wrote: »

Using your eyes is a mistake IMO. Fact is, almost all DAWs do not have an accurate waveform display. It's accurate enough for editing, that's all. For a start, it doesn't show what the reconstruction filter is doing.

The real problem with DAWs is that they allow us to rely on our eyes.

That's why I use Goldwave. I know the graphical wave representations in Ableton are not accurate. Their not perfectly accurate in Goldwave either. But for me they're better than in Ableton.

In Goldwave the hard limiter will sometimes miss spikes. Some times spikes will not be displayed.

Don't get the wrong idea thinking that I'm claiming you can see everything in the graphic. You can't but you can see quite a bit at a glance. At a glance that takes seconds - not spending minutes painstakingly listening to the track again and again. Where you will wear down your hearing with fatigue.

Both your eyes and your ears will deceive you.

Anima wrote: »

Hmm I'm struggling to think of anything that you can see but can't hear if I'm honest. Can be easier to spot clipping that might be difficult to hear but I guess it doesn't matter if you can't hear it.

Many spikes are just artefacts. They could be caused by anything. Like static from intergalactic cosmic rays (I didn't make that up) or Sun spots (I didn't make that up either). They can be only milliseconds long. They're impossible to hear.

But. How they're important. Something like soundcloud maximises the mp3 you sent it, to the highest peak in the file. If there's a tiny spike you've missed it will lower the volume of the entire track.

There's more to it the whole thing though. I'd like to see a 3d representation of the wave - to see the frequencies. I was listening to a recently remastered track of Fela Kuti's last night (Dead Nigerian guy) - the track was perceptual really loud and clear.

On a 3d representation of the spectrum I would probably be able to see spikes ( and just bits that are too loud) of frequencies that are just not needed. You would use a combination of your eyes and your ears. The best mastered stuff I've heard is stuff that is perceptually very loud but also light and airy at the same time.

But Beethoven beats everyone. When he went deaf he started using a wooden block between the piano and his chin - so he could hear the vibrations through his jaw.

madtheory

Where can we hear this mastered music?

krd

madtheory wrote: »

Where can we hear this mastered music?

Mine?

https://www.youtube.com/watch?v=3y1iV4kedmg

I am an amateur trying to produce my own music. I'm slowly getting there after a few hundred dead ends.


I haven't been using Ableton that long and I'm still struggling with lots of aspects of general production.

Anima

Like static from intergalactic cosmic rays (I didn't make that up) or Sun spots (I didn't make that up either).

That only applies to analog circuits not digital, which I assume is what you're using but maybe not.

Something like soundcloud maximises the mp3 you sent it, to the highest peak in the file. If there's a tiny spike you've missed it will lower the volume of the entire track.

But isn't that why you'd limit it yourself first to smooth over those peaks? Most decent normalisers will have a threshold so that a single sample like as you describe won't ruin the result.

Normalising isn't a great thing necessarily either. It will make everything louder in the track, even the noise.

Like all these kinds of tools, its best to use them in small amounts. Really you should fix most of the problems before reaching the master bus so that you only needed to apply subtle amounts of processing to get that final 5% volume.

krd

Anima wrote: »

That only applies to analog circuits not digital, which I assume is what you're using but maybe not.

I know. The point is you'll get artefacts and spikes from all kinds of places. Digital has artefacting too.

But isn't that why you'd limit it yourself first to smooth over those peaks? Most decent normalisers will have a threshold so that a single sample like as you describe won't ruin the result.

That's true. But you can't count on what will happen in the normalisation. From experience I can tell you soundcloud normalises to the peak sample in the sound file. Even a sample your sound editor ignores.

Normalising isn't a great thing necessarily either. It will make everything louder in the track, even the noise.

There's normalising the levels of the track so everything is at the same level. Then there is normalising to the peak. Where the volume of the track is raised to the point that the peak is at 0db. This is what sound cloud does.

Like all these kinds of tools, its best to use them in small amounts. Really you should fix most of the problems before reaching the master bus so that you only needed to apply subtle amounts of processing to get that final 5% volume.

I'm only figuring it out as I go about doing what I'm doing. I was looking at the sound academy glossary last night. There are hundreds of items in the glossary. There's possibly hundreds of things left out. That's hundreds of things you need to know and be able to do, to produce music.

madtheory

I think the kind of "mastering" you use is not really mastering at all. There's an art in getting the bass drum pumping with the track, and that depends on how the master compressor/ limiter is set. It's an artistic process, not really an engineering one.

The track you posted is a good example. It sounds well balanced and pumps well, but it would still need the polish of mastering by an independent and objective pair of ears, IMO. That said, it's hard to judge the sound quality with the youtube compression. There's a distracting pre echo on the bass drum.

Normalising is measuring the highest peak, raising that to 0dBFS and then raising everything else by the same amount. You seem to be differentiating between normalising the whole mix, and normalising each element of the multitrack.

Normalising only ever ended up in DAWs because Pro Tools evolved from Sound Designer, a program for editing samples on the EMulator II. It was useful for making a bunch of multisamples peak at the same level. It's use in mastering is questionable at best, because 90% of the time it does not affect perceived loudness.

Anima

That track sounds decent Krd. It sounds fairly well mixed as well.

I don't think you need to be focusing on normalising and indiviual samples like you are though. It's wasted effort and you should take a step back and try to improve your method instead of trying to correct mistakes after.

The main thing to remember with loudness is that it only partially has to do with dB levels. Thats the acoustic / engineering / mechanical part, which is important and what all of our tools are based on (limiters, EQs, compressors).

However there is another side to it, which is how our ears are going to perceive the sound. Think of when someone claps their hands beside your ears. If you're in a quiet room, this is a deafening sound. If you are in a football stadium, you'll barely notice it.

All of these different scenarios and reasons for these effects fall under the heading of psychoacoustics.

I did a module of it in my course last year and found it to be very beneficial.

krd

madtheory wrote: »

I think the kind of "mastering" you use is not really mastering at all.

I refer to my mastering technique as Ghetto mastering. I'm sure there is a lot more that can be done. But I'm not at the point of doing it - or sending it to someone else for mastering.

There's an art in getting the bass drum pumping with the track, and that depends on how the master compressor/ limiter is set. It's an artistic process, not really an engineering one.

There is an art to it. There's much more to it than the master c/l. There is a black art to it - than I know as in I can hear it in other recordings but can't pin down. The pain with pencilling in on a DAW - is having to know precisely what you're doing.

Like, if you get three people - a bass player, a guitarist, and a drummer. Give the bass player a bass line of a few notes, the guitarist a few chords, and the drummer the instructions to play a pulsing groove. After a few minutes of playing they'll start making subconscious little changes that will make the over all thing pump more. And that pump will persist through all kinds of degradation.

The track you posted is a good example. It sounds well balanced and pumps well, but it would still need the polish of mastering by an independent and objective pair of ears, IMO.

There's plenty wrong with it. Still sounds better than I would have been able to do a few months ago. I've set myself the exercise of trying to create tracks quickly and post them to Soundcloud as complete as possible pieces, instead of doodling. I'm learning from each one. I've also learned that a lot of the technique I'm slowly picking up is subconscious and I'm not sure how I did some things.

Before I posted that one, I tried to do some EQ'ing to bring the kick out stronger, but I kept losing the a major part of the bass I wanted to hang onto. I didn't want to side chain it - because I didn't want the bass sound to make the push/pull sound.

That said, it's hard to judge the sound quality with the youtube compression.

Youtube was killing me for ages. The Youtube processing can really destroy the sound of the original recording. I think it's to do with spikes again - though I'm not sure. I know now, that if I prepare the sound like I would for soundcloud. It comes out much much clearer than before - where what I was uploading was seriously degraded for reasons unknown to me. Some people - dance music producers and record companies can get really pristine sounding recordings on You tube. You tube don't really give a good guide to getting your sound right.

There's a distracting pre echo on the bass drum.

I'm experimenting with different kicks. I've started sampling them directly from other recordings. I'm not sure what's been done with a lot of the kicks - some of them have a strong presence that cuts through the mix better. I've been listening for kicks a lot recently. I'll figure it out.

madtheory

What I mean is, the compression is an integral part of the track, it's part of the composing process. It can totally change the feel as you say. Which is entirely unlike mastering.

The pre echo is a result of the youtube compression. The key to getting a track to sound good when there's lossless compression on it- is to have it mastered properly. This requires an experienced pair of ears, and accurate monitoring (good speakers and an acoustically neutral room). A good master will sound good anywhere, assuming there's nothing up with the mix. Funnily enough, it's very easy to get a good mix when you have good monitoring, including room acoustics.

krd

Anima wrote: »

All of these different scenarios and reasons for these effects fall under the heading of psychoacoustics.

Psychoacoustics are very important. It covers the whole perceptual hearing thing. The ear adjusts to sounds very quickly. Changes in sounds are perceived in particular ways.

I was listening to a percussion performance last night by a classical percussion group. There were massive dynamics in the recording. Most of it was at a very low level. So when the loud bits came in, they sounded very loud. But they peaked just at the same level that a contemporary piece of music would be more or less constantly. I was actually playing the music very low - the loud bits sounded really loud. Also there were quite a few very quite moments which also had a great dynamic to them.

Like it says in the wikipedia article - a clap in a library sounds very loud - the same clap on a busy street will be barely audible.

Your hearing turns up and down it's own volume control by itself. Like how your eyes adjust to the dark and sudden blast of light can be uncomfortable.


Contemporary music does suffer from the desire to max out the spectrum in production - if compressors and limiters were banned every where and engineers not allowed have graphic tools where they could see the wave or a VU meter - music might get more interesting.


I think the trick to making something sound really loud could actually be silence.

Very Zen Buddhist type idea, but there you go. I've been listening to a crunchy and loud sounding Electro track recently. I'm nearly sure a lot of it's loudness and dynamics have been achieved through short barely perceptible silences here and there. I know from listening I can hear bits where every instrument is killed and say a bass stab is let take all the space for a fraction of a second - it's not perceptible that rest of the music is not there unless you listen closely, then you can hear it very clearly.

Anima

Perfection is achieved, not when there is nothing more to add, but when there is nothing left to take away.

The whole mixing / mastering process always reminds me of that quote. You're right about silence being important. I think you have to be careful to have your sounds only as big as they need to be and remove / attenuate everything else.

I'm starting to believe more and more that it's really about getting the original sound right before you start EQ'ing and compressing and all that. I've often found that most of the problems with electronic sounds is by synthesising poor sounds that take up to much room and aren't focused enough in a specific region.

It's much better to make smaller sounds and layer them then to try and create one monolithic sound because its a nightmare to get it to fit with everything then.

krd

Anima wrote: »

I'm starting to believe more and more that it's really about getting the original sound right before you start EQ'ing and compressing and all that. I've often found that most of the problems with electronic sounds is by synthesising poor sounds that take up to much room and aren't focused enough in a specific region.

It's much better to make smaller sounds and layer them then to try and create one monolithic sound because its a nightmare to get it to fit with everything then.

This I've this heard too. It's one reason supposedly using presets on synths is a bad idea. They've been made too fat for the purpose of sounding big as a demo. Sounds need to be thinned down to give each other space.

If you listen to this carefully - you'll hear sounds being made to appear and disappear - it's not just normally masking it's pretty extreme sound removal (I'm not sure if it's side chaining - or if had to be done manually)- and short silences and drops before big crushing sounds.

https://www.youtube.com/watch?v=Dnxr9fWUaXU

madtheory

You're right about contrast- it's the key in any art form.

krd wrote: »

if compressors and limiters were banned every where and engineers not allowed have graphic tools where they could see the wave or a VU meter - music might get more interesting.

If you know what you're doing those are all useful tools. The idea of banning them is wrong-headed. The actual problem is that the tools get abused. If all sound engineers actually were engineers, as in properly trained, and good room acoustics and speaker systems was the norm rather than the exception in domestic environments, we wouldn't have a loudness war.

krd

But how long have the loudness wars really been going on? Musicians and engineers may have been trying to get one over on each other forever.

Take Tchaikovsky's 1812 Overture.


The 1812 Overture is scored for an orchestra that consists of the following:

* brass band1 (finale only)
* woodwind: piccolo, 2 flutes, 2 oboes, cor anglais, 2 clarinets in B♭, 2 bassoons
* brass: 4 horns in F, 2 cornets in B♭, 2 trumpets in E♭, 3 trombones, tuba
* percussion: timpani, bass drum, snare drum, cymbals, tambourine, triangle, bells2, cannon3
* strings: (violins I, II, violas, cellos, double basses)


The loudness war problems aren't just with the engineers and acoustics. The compression and limiting used by radio stations can be brutal. I keeping hearing this effect in broadcasts, that the sound wobbles - like a circuit trying to rebalance itself. Unusual swells and drops over dynamic bits of the music.

Sometimes the recordings - big commercial recordings sound burnt

Don't even get me started on television. There's been a TV loudness war with ads that has been intensifying over the last few years. You're watching a film - the ads come on - if you don't mute or lower the volume quick enough you're blown out of it.

It's impossible to ban compressors. But I wouldn't be surprised if broadcasting standards changed, as they do. I know there was talk of regulating the sound on ads in the UK. It's not capturing anyone's attention - it's harassing people.

madtheory

Loudness in this context refers to a lack of dynamic range. Dynamic range is contrast.

The 1812 is an extreme example of contrast, not loudness. The opening goes on for nearly 10 mins IIRC, starting at ppp and up to a low p (p= piano which means "soft"). The big climax calls for canons as you say. At that point we're at fff (f=forte meaning loud). That's as wide a dynamic range as one can find in western music! Totally unlike the problem you're describing.

If one has good monitoring, it's much easier to hear these things, which was my point about acoustics etc.

What's happening in the wider broadcast world is that the EBU (European Broadcasting Union) has recently settled on a standard for measuring loudness and is about to implement it. The driving force was complaints from consumers about the loudness of adverts. This is because films generally exploit dynamic range in the same way Tchaikovsky did, but advertisers tend to be completely lacking in subtlety.

Finally, musicians are complicit in this "war", they're the ones who sign off on the masters after all.

PaulBrewer

I find Canons a b-atch to mix, especially in 3/4 .... but the worst is Nuclear Bombs , very hard get the low end right ....

PaulBrewer

madtheory wrote: »

Finally, musicians are complicit in this "war", they're the ones who sign off on the masters after all.

Yup