Bring Back Dynamics!

[rob sparx]

EDIT: Oops going off on 1 again!! I've got a bit of a difference of opinion with Jason on the whole loudness/EQ in mastering thing (not the overcooking bizness thats another story) but I shouldn't be airing my dirty laundry in public as its a bit out of order - sorry Jason! Like he says if u want your master to turn out right you should sit in on the session I just ain't got time for that right now I'm too busy sorting out my album.

Everyones got a different opinion on what a good mixdown is & some people like more loudness than others I could waffle for hours about this bollox but its a waste of time. I'll post a mix on this forum next week so anyone who wants to look at my mixdowns can check it out then........

[rob sparx]

xTheWiddler said: "but dont give me solid blocks and say thats how its supposed to be for dance music"

Its however you want it to be its what it sounds like thats important not how it looks! If I write a more chilled tune its obviously not going to be a solid block but if I'm writing a throbber/wobbler/reecey tune then I want as much impact/drive as possible cos those tunes are going to be played in the loudest part of the set.

I'm saying you should get music as loud as possible WITHOUT compromising the dynamics of the tune. If your good at mixing then that means you can get a solid block (at the louder drops not the whole tune!) without it sounding muddy and overcompressed - like you say you always can turn the volume down in the mix if you think its too loud. It takes a lot more skill to do a loud mix that doesn't distort/overcompress than it takes to do a quiet mix.

EDIT: Actually I had another look at my mixdowns and they're not quite solid blocks you can still make out the waveshapes they're just quite squashed so your right solid blocks are a step too far but I'd say a bit of waveform squashing is allowed if the mixdown is good!

[junglist]

but dont give me solid blocks and say thats how its supposed to be for dance music

Worst tune for that is Rusko & Caspa's; Bread Get Bun tune. It has a horrible bass line that a 3 year old could have made.

Here is the bass line Broken down from that tune:

Wob Wob Wob Wob Wob Wob Wob Wob Wob Wob Wob
......................................................Pitch + 1.....Pitch -0.2
00:26:00.......................................00:36:00....00:39:00 (time)

[d-code]

99% of modern day D&B mix downs = shit.

thats my contribution to this discussion

thank you, i agree, and its happening to some dubstep

i dont mean dynamics as in make it a jazz tune

but dont give me solid blocks and say thats how its supposed to be for dance music

i don't think dubstep is going down that route at all in all honesty.

a lot of the d&b guys are under the impression that it has to be really fucking loud and maxed out and to be fair most of them aren't experts in this field (not saying that i am) and the finished tune usually sounds pretty dam awful.

thats why i for one leave it to the proper experts - like jason and co down at transition.

[jade_monkey]

Amisane: "If one of my tunes looks like a solid block I always turn the master down until you can make out a waveform. Would you say that is bad practice?"

Yes that is bad practise - the volume you mix down at depends on whether you want to master the tune yourself or get someone to master it for you. If someone else is going to master it then usually they'll say get an average of 0db which might be a peak of +2/+3db on the master volume (a little bit squashed but not much) - too loud and the mastering engineer won't have any headroom to work with so won't be able to do much with your mix although cheaper engineers will overcook your mix just to prove they've done something 2 and you might actually end up paying to make your tune sound worse! If your doing your own mastering (not recommended if your mixdown aint perfect) then get it as loud as you can.

Even if your not going to master your own tune then I'd still say keep the volume in your tune as loud as you can get it WHILST WRITING (you can reduce the volume later on) - this is a good idea because at higher volumes any dodgy frequency overlaps become very obvious in the form of distortion then you can solo the offending channels and eq/process them until there is no distortion. Write a tune at a lower volume and you might not hear the distortion until later on when the mastering engineer attempts to limit your mix and can't get much boost on it because of dodgy frequencies present in your mix which you didn't even notice!

Thats like the opposite of what i was teached. I think there is a big difference in monitoring at a loud volume and masterting it loud. If your waveform is a block, then you definetely made some mistakes. And I also recognized the loudness war in electronic music, lately.
So if you are mixing digital, you should keep at least a headroom of -3db for mastering. I work around -6db usually.

But... Rob, your stuff sounds ace.

[rob sparx]

Thx 4 the compliment - as I've said I was exaggerating a bit when I said it was a block there's just a bit of squashing going on.

I'm no expert in the field of mastering and I agree with you its far more important to mix the tune down loud than to master it loud. I've only done my own mastering on the last few records and I've not even listened to them yet but my mates who have r saying good things about them (probably cos they're a few db louder than normal!) - got that idea from Noisia as they do all their own mastering - hasn't done them any harm so I thought I'd give it a try.

Most of the tunes I've released have been sent off stuff for mastering @ -1 but a lot of engineers disagree how loud the volume has to be and its a regular argument on some of the other forums I'm on. The 0db average as I understand is how Stu @ Metropolis likes his mixdowns (sorry if thats not right Stu its 2nd hand information - correct me if I'm wrong on that 1) which is good enough for me as his skills are legendary. What confuses me about the whole loudness thing is that some top RNB/Pop producers using expensive pro-tools setups are constantly pushing their volumes up yet many engineers are saying to do the opposite! When lots of people are giving out conflicting advice you start thinking its all a matter of opinion - I just like my tunes to sound how I mix them down not how someone else thinks they should sound. Sending off a tune to be mastered isn't like waving a magic wand to improve your mix - they can come out sounding (arguably) worse if your not present in the session and at the moment (later in the year might b different) I don't have the time to sit in on sessions so I'm just going down the DIY route for a bit.

[jade_monkey]

http://en.wikipedia.org/wiki/Loudness_war

Maybe you should give this a read. This explains why the R&B people do what you said they do. Me for my part, can't barely listen to most of this stuff because of that, it hurts my ears.

[rob sparx]

Read the article - at least the 4-6db of boosting I'm using (digital clipping or limiting) is less then the +9db of "contempary recordings" which I presume talking about RNB/Pop. I haven't found an software mastering analogue emulator that I'm happy with yet - what software would you recommend for this use?

Not sure if this is relevant but no-ones mentioned distribution of EQ is this discussion which I think is important because lots of treble sounds harsh and hurts your ears but I think lots of bass is nice and I like a bit of distortion/drive in the sub frequencies - although my mixes are +4 to +6 overall, volume above the upper bass range (about 300hz i think) is normally only an average of -10db. Its the sub region where I drive it 2 fuk peaking just under 0db (looks like a mountain peaking @ 60hz) - can't stand lots of treble like in RNB production I try 2 smooth things out so they don't hurt my ears but I love Bass!

[jtransition]

I am gonna go out on a limb here ;
I think there is a lot of confusion here around the level that a mix is recorded at inside a DAW(cubase logic reason etc)and the physical loudness of that mix.
The two things are different, The mix level DBFS is a digital scale and relates to how much level is inside the DAW ,If you mix down too hot then you introduce digital clipping similar to limiting the audio, This can restrict what can be done at the mastering stage.
DBVU,relates to the dynamic range of the audio which is the difference between the peaks and the troughs in the audio waveform.
This is not just my opinion it is fact.
Mastering is about translation, My job is to make sure that the audio translates onto the end format in a manner that the client wishes, I can give my advise but it is the customer who chooses what they want and as such we have the best tools and ability to make that translation as unobtrusive or intrusive as is appropriate.
There is plenty of accurate info on the internet as well as the normal misinformation. Anyone who is interested here is a link a for a good book about audio and some interesting info about loudness etc.
Regards
Jason


http://www.amazon.com/Mastering-Audio-S ... 0240805453
http://www.digido.com/index.php?option=com_kb&Itemid=67
http://www.myspace.com/transitionmastering (look at the video)

[vendetta]

@ xwiddler - my 2p's worth

I reckon you're right to point out that music in general is becoming 'louder' sometimes by 'over compression' in comparison to live recordings/ other types of music etc. But i reckon you need to be more say more first about how you are going to say and measure the loss of dynamic range and why that's a bad thing? Sort of what dubstep/dance 'should' like if you see what i mean?

Isn't it really high frequency/ too much top end distortion that's bad for hearing not just loudness/compression general? Its more when people boost the mid/top end to increase the perceived loudness of a track (the frequency bands which are most sensitive in most people's hearing) or when its badly engineered in a club that it badgers your ears. - I.e. when a venue's sound engineer doesn't understand dance music 'natural sound'? and engineers it like it was a rock band etc, and its then made worse when djs boost the levels to try and compensate giving everyone tinnitus and dead spots in your hearing response?

I mean dance music in general sounds strange to your ears without a bit of 'over compression' i reckon as we've got used to it. Like when you compare you own tracks to mastered tracks - loads of other factors here obviously ... think of eq presets 'dance' 'jazz' etc. Its a bit like playing a live jazz band through a reggae or dance sound system, it'll sound weird or weak if you see what i mean, as your ears are used to certain eq/compression ratios for certain types of music.


Like if you play a track that isn't as 'overcompressed' as the others in the middle of your set (perceived as a drop in volume) the change to different dynamics can make the track seem out of place/weaker, draining a bit of energy from the mix (isn't that part of the reason dj's like playing as loud no matter how good the mix is it always seems to loose something if its quieter no?). So maybe in that way, if your sets about a progression of bigger and bigger tracks, if more compression lets you squeeze a bit more juice out of the soundsystem people will want to keep compressing more and more. (Sort of like how the bpm in dnb/jungle has increased more an more as a way of hyping up the crowd as times gone on.)

Is that sort of you mean Rob Sparx in this bit you said above?

Not sure if this is relevant but no-ones mentioned distribution of EQ is this discussion which I think is important because lots of treble sounds harsh and hurts your ears but I think lots of bass is nice and I like a bit of distortion/drive in the sub frequencies - although my mixes are +4 to +6 overall, volume above the upper bass range (about 300hz i think) is normally only an average of -10db. Its the sub region where I drive it 2 fuk peaking just under 0db (looks like a mountain peaking @ 60hz) - can't stand lots of treble like in RNB production I try 2 smooth things out so they don't hurt my ears but I love Bass!

@ Jason/ Rob

if its not a trade secret like (i respect its your living etc), can you give any tips on how you go about eq/compressing dubstep? Do you have an idea of the types of sound systems it will be played on and try and account for that as well as home listening etc? Or just test it on as many systems as possible. Seems like its well hard with dubstep as its about feeling the bass as well as hearing it, as in its proper club music if you see what i mean?

[jade_monkey]

I am gonna go out on a limb here ;
I think there is a lot of confusion here around the level that a mix is recorded at inside a DAW(cubase logic reason etc)and the physical loudness of that mix.
The two things are different, The mix level DBFS is a digital scale and relates to how much level is inside the DAW ,If you mix down too hot then you introduce digital clipping similar to limiting the audio, This can restrict what can be done at the mastering stage.
DBVU,relates to the dynamic range of the audio which is the difference between the peaks and the troughs in the audio waveform.
This is not just my opinion it is fact.
Mastering is about translation, My job is to make sure that the audio translates onto the end format in a manner that the client wishes, I can give my advise but it is the customer who chooses what they want and as such we have the best tools and ability to make that translation as unobtrusive or intrusive as is appropriate.
There is plenty of accurate info on the internet as well as the normal misinformation. Anyone who is interested here is a link a for a good book about audio and some interesting info about loudness etc.
Regards
Jason


http://www.amazon.com/Mastering-Audio-S ... 0240805453
http://www.digido.com/index.php?option=com_kb&Itemid=67
http://www.myspace.com/transitionmastering (look at the video)

^ That.

Jason, could you explain a little bit more of the relationship of the dynamic range and the so called loudness war?

The Katz book you mentioned is really worth to be read. I only managed to read some chapters but i will definetely get back to it someday.

[rob sparx]

To Jason:

I'm definately not confused - every time I've said say DB in this thread I mean DBFS not DBVU (ie the monitor level on the master volume in my DAW) I'm talking about digital not analogue measurements.

As I've already mentioned I completely agree with you sending louder mixdowns for mastering is a big no-no as the engineer would have no headroom to work and you'd be wasting your money BUT if I'm just sending the tune straight to cutting without mastering and not using a master limiter then I will mixdown up to +6 in the red as the tune requires no further processing - I would never send a mixdown that loud for mastering (excluding past mistakes when I really didn't know squat) that would be very foolish!

To Vendetta:

That is exactly what I was getting at with the EQ thing! The part of your ear that picks up bass is much much more resilient than the part that picks up treble anyone who's been to a DMZ night knows that the bass is probably gonna hurt your stomache more than your ears but treble - ouch! I completely agree with the other stuff you've said as well - if you don't boost your music other people still will which will cause problems in the mix - also as u say limiting allows you achieve a higher percieved volume so at full blast a rig will sound significantly louder with a tune that has more limiting/drive on it.

To be honest I'm not one for doing things by the book I couldn't give a fuk if every mastering guide in the world told you you should do this and shouldn't do that - rules are made to be broken! If I can't hear my tunes distorting then why should I turn the volume down - its only the sub that I'm driving and I like the sound of a driven sub. I don't ever use a master compressor and the tunes I do use a master limiter on tend to be mellower anyway - no individual channel in my tracks are ever louder than -2db and I'm happy with way my mixdowns sound without mastering so I don't see why I should be turning the volume down when it sounds exactly how I want it to (and still relatively quiet compared to RNB music)!

People who really benefit from mastering are those who can't get a very good mixdown especially those using reason - which is a LOT of producers in dubstep. For those people its pretty essential to get your stuff mastered otherwise it will sound amateur.

[spencertron]

People who really benefit from mastering are those who can't get a very good mixdown.

i think i agree with this, Every professional studio i've been to, the producer involved has often said something similar...i.e. "lets just concentrate on a decent mixdown", whenever mastering was mentioned.

If you intend on having something mastered, work at lower volumes i.e. leave plenty of headroom. A good mixdown will not require a great deal of mastering.

[r]

a bad mix is a bad mastering dears

wicked sounding mix makes a wicked sounding mastering

[jade_monkey]

People who really benefit from mastering are those who can't get a very good mixdown.

i think i agree with this, Every professional studio i've been to, the producer involved has often said something similar...i.e. "lets just concentrate on a decent mixdown", whenever mastering was mentioned.

If you intend on having something mastered, work at lower volumes i.e. leave plenty of headroom. A good mixdown will not require a great deal of mastering.

Well, my impression was that the main task of mastering is to make sure that the sound quality does not differ, either you hear it on the headphones, your stereo or in the club. Also the mastering process focusses on the overall sound quality of an album. If you write a tune one is louder, the other one not... so the mastering engineer makes sure that the whole album has the same sound quality and stuff.

[rob sparx]

Yea your right - I was just talking about 12"s not albums - doesn't really matter if a 12" has one side louder/different EQ than the other but on an album any differences are gonna stand out big time. That reminds me - Bigup 2 Jason on those 2 Burial albums they sounded sick.

[james fox]

BUT if I'm just sending the tune straight to cutting without mastering and not using a master limiter then I will mixdown up to +6 in the red as the tune requires no further processing .

you have tunes cut to vinyl with no mastering at all, and these tunes are clipping your DAW master bus at +6dB?

blimey

[rob sparx]

Got the idea after reading a computer music interview with Noisia - was wondering how the hell they got their mixes sounding so phat and it turns out in most of tunes they were sending the stuff straight 2 cutting with no hardware involved in the production side of things the logic being that they get their tunes sounding how they want them to sound so why get anyone else involved when they were already happy with their mix! Seeing as they've got 1,379,343 myspace profile views and counting I don't think its done their careers any harm and it probably saved them a quid or two in the early days.

I'm not recommending this approach to everyone by any means but it works for me - if I hadn't mentioned it here I doubt anyone would have even realised the tunes had not been mastered. Same goes for Neros latest batch which have also not been mastered although everyone who hears them thinks they have!

[james fox]

interesting. will defo have to listen to your tunes and see for myself

[rob sparx]

Please let me know what you think - the tunes I've sent straight to cutting so far are DUBTING007 and ZAUDIO007 but there's a few more in the pipeline that will come soon.