The No.1 Website for Pro Audio
 Search This Thread  Search This Forum  Search Reviews  Search Gear Database  Search Gear for sale  Search Gearslutz Go Advanced
Biggest mastering Screwup?!!! Sia "Some people have real problems"
Old 19th February 2008
  #1
Lives for gear
 
nlc201's Avatar
 

Biggest mastering Screwup?!!! Sia "Some people have real problems"

Unbelievable. Apparently some people do have real problems.

First of all, great record. Love it. Great tunes and awesome performances. Recording and mixing sound awesome.


Second of all, the mastering, WTF???????????!!!!!!!!!!!!!!!!!!!!



I put it on and immediately noticed that it was a lot quieter than most modern records. Now, I mean ALOT quieter. Interesting, I thought, perhaps people are going back to a more sane level of final output levels for CDs. But, after listening further, I noticed many of the same loudness artifacts I hear on most modern "loud" records. The limiter becomes quite audible at many places as the track slips into some fairly serious distortion. As I sit, I'm listening to it on a pair of PMC AML1s driven by a Cranesong Avocet so my chain is fairly accurate. So, it's got the symptoms of the modern "loudness" wars yet it's significantly quieter than most records. A quick glance at the digital meters on the avocet helped to shed light on the unbelievable answer. The record seems to be peak limited at somewhere around -8 dbfs. WTF?!!!!!

Yep, you heard me. The record was highly peak limited at around -8 dbfs. So that begs the next question, why?!!!!!!!! The reasoning behind the loudness wars is to have your record louder than the next guys', even at the expense of fidelity. But this record gives away a lot of its fidelity and doesen't even compare to other records as it's 8 dB quieter. It loses in both cases. So, this leads me to believe that this was actually a giant screwup somewhere between the pre-mastering and manufacturing stages that nobody happened to catch. I guess that's what happens when you have a coffee company in charge of putting out your record.


Ironically, if they kept the overall RMS levels of the record the same and took off the peak limiting, it would probably sound quite stellar.......


Hooray for 14.5 bit records!!!!
Old 19th February 2008
  #2
Gear Addict
 
johnlink's Avatar
 

Have you told the mastering engineer that you are not satisfied?

John Link
Old 19th February 2008
  #3
Deleted User
Guest
I guess it can safely be encoded to mp3 without any clipping tendencies.

"Always look on the bright side of life".


Best Regards
Patrik
Old 19th February 2008
  #4
Gear Nut
 

One day a band came to me with a master they were not satisfied with... It was exactly the same problem. After hearing how it was awfully limited and distorted but with a low level, I analyzed it with the RME Digicheck and there was nearly no headroom left... except the 6 dB range from the top that was never reached by the signal ! I called the mastering studio to understand, and after a visit to them, the reason was :

To compare the song with/without the limiter and not being influenced by the gain, they dropped "level out" of the limiter by about the same amount the signal was limited (I do that too)... But they forgot to bring the "level out" of the limiter back to zero before exporting the whole album ! And I guess (even if they did not told me so) that forgetting the drop of volume, when comparing with commercial Cd's, they tried to get a high level with the -6dB, resulting this awfully limited master. Maybe a bad day, but for sure a client loss (at least for them).
Old 19th February 2008
  #5
radio friendly ???

Looks like someone misinterpreted a broadcast audio norm for 'radio friendly' ?

Btw, I love your avatar !

Perhaps Beaker went into mastering nowadays

Peter van't Riet
FineTune Mastering
Old 19th February 2008
  #6
Gear Addict
 
streaky's Avatar
 

Quote:
Originally Posted by nlc201 View Post

Yep, you heard me. The record was highly peak limited at around -8 dbfs. So that begs the next question, why?!!!!!!!! The reasoning behind the loudness wars is to have your record louder than the next guys', even at the expense of fidelity
I wasn't aware there was any reasoning behind the loudness wars!

Quote:
But this record gives away a lot of its fidelity and doesn't even compare to other records as it's 8 dB quieter. It loses in both cases. So, this leads me to believe that this was actually a giant screwup somewhere between the pre-mastering and manufacturing stages that nobody happened to catch. I guess that's what happens when you have a coffee company in charge of putting out your record.

Interesting thoughts...I haven't heard the album but I have worked on some of her stuff with Zero7 before and if it's anything like this new album, it's not the sort of music that needs to be crushed (if any 'needs' to be...)

I do know where this was mastered and I would be surprised that the engineer in question would create such a bad result, he's got a great pedigree. It's not that I've not heard bad sounding masters from their studios and multiple engineers, because I have! But what mistakes could be made between mastering and manufacture? It's not going to be the quality of the CDs or some post-mastering processing.

For me it basically lies with the approval of a reference CD. If the record company like the sound, they approve it, simple as that. So if any mistake has been made, it was approving that record. But then I haven't heard it so I might disagree with you on the sound!
Old 19th February 2008
  #7
Lives for gear
 
Paul Gold's Avatar
 

Verified Member
I'd put my money on post processing in the office by the intern.
Old 19th February 2008
  #8
Deleted User
Guest
To check that the peak levels settled as planned, onto the very master disc, is in my world a normal part of the normal QC procedure.

Just a few snips here and there, from the Plextor driver in playback mode and a fast eye on any suitable peakmeter worthy to say "-0.3 dB".


Best Regards
Patrik
Old 19th February 2008
  #9
Lives for gear
 
Jerry Tubb's Avatar
 

Verified Member
Quote:
Originally Posted by Paul Gold View Post
I'd put my money on post processing in the office by the intern.
Possibly intentional... as revenge for all those months of cleaning toilets for free!

On the bright side -8dBFS peaks ought to leave plenty of room for iZotope RX DeClipper when they're ready to reprint!

JT
Old 19th February 2008
  #10
Lives for gear
 
Paul Gold's Avatar
 

Verified Member
I meant the record company intern who was told to resequence it. Hopefully a studio intern wouldn't do that.
Old 20th February 2008
  #11
Quote:
Originally Posted by Paul Gold View Post
I meant the record company intern who was told to resequence it.
... and did so in iTunes with SoundCheck turned on.

- J.
Old 20th February 2008
  #12
Gear Guru
 
Thomas W. Bethe's Avatar
 

Verified Member
I have a hip hop client that comes in with his mixes exactly as described. The mixes do not go above -4 dBFS but he limits and compresses them so much that they are always distorted.

I have told him about his "problem" more than once but he still keeps doing it. His major problem is that he does not understand what he is doing in terms of recording /gain staging/ mixing so it is hard for him to fix what he does not understand. He is a GREAT hip hop artist and his beats are the best I have ever heard BUT he is a lousy recording engineer.

I have gotten mixes in from well respected mixers with the same problem. The are good mixes but some one strapped a 2 channel limiter/peak clipper across the mix bus and told the software or what ever to not let the peaks exceed -4 dBFS. So we have a mix that is "overdone" but never goes above -4 dBFS.

Maybe the original recording was this way. It is hard to blame the mastering engineer all the time when you don't know what the original material sounded like OR what the client wanted.

I do agree in this case that it was probably a screw up somewhere along the line to the final glass master but it would be hard to place the blame without hearing all the intervening steps.
Old 21st February 2008
  #13
Lives for gear
 
allencollins's Avatar
 

Could this be the first phase of

THE LOUDNESS WAR PEACE TREATY ?


This will be as significant as the Battle of Waterloo Treaty and the Versailles Treaty!!


Thanks GOD someone has seen the light
Old 21st February 2008
  #14
Gear Addict
 
Spookym15's Avatar
 

This is interesting. From all the engineers I have talked to they always throw a limiter on the master buss, they usually leave 4 dB of head room for the mastering engineer. I usually check with the person who is doing the master to see how much head room they need. My experience is generally about 3db of head room and the mastering engineer will usually go up to -1db, because of the crappy converters in most stereos. I mix so that the loudest I ever get it at about -3db, and I do use a peak limiter to just to make sure that it does not go over. The most my mixes ever get to is about -2db at most, which is why the limiter is is on it.

Is this wrong? There should be something agreed upon between engineers and mastering engineers.
Old 21st February 2008
  #15
Lives for gear
 
jamsmith's Avatar
 

I just don't get it. I used compression on vocals because most singers between poor mic and poor vocal technique present to much dynamic range that needs to be tamed. I put compression on bass player who plays with uneven dynamics. I put limiters on snares and toms if the drummer lacks dynamics. That is it. On a final mix, I may use some limiting to tame a few unruly peaks. I like my music to breath. I love dynamics! Who are all these people who want their music squashed?

"He's a great engineer but he uses to much buss compression"

How is someone who kills dynamics a great engineer?
Old 21st February 2008
  #16
Gear Maniac
 

Quote:
Originally Posted by Paul Gold View Post
I'd put my money on post processing in the office by the intern.
I think you're right! And possibly he/she did it on Windows Media or iTunes.
Old 21st February 2008
  #17
Lives for gear
 

Verified Member
Quote:
Originally Posted by Spookym15 View Post
This is interesting. From all the engineers I have talked to they always throw a limiter on the master buss, they usually leave 4 dB of head room for the mastering engineer. I usually check with the person who is doing the master to see how much head room they need. My experience is generally about 3db of head room and the mastering engineer will usually go up to -1db, because of the crappy converters in most stereos. I mix so that the loudest I ever get it at about -3db, and I do use a peak limiter to just to make sure that it does not go over. The most my mixes ever get to is about -2db at most, which is why the limiter is is on it.

Is this wrong? There should be something agreed upon between engineers and mastering engineers.
Yes, you are wrong.

A lot of mix engineers don't use a limiter on the master buss because they are careful not to have the sum of their tracks go over 0 dB, and lower is even better.

Some of the pro mixers I work with pump up their "final" mix for the client, limiters/vol. maximizers and/or clipping, but also print a mix that has no limiters/volume maximizers on the buss.

They send me both versions. The uncompressed, for ease of mastering, and the hyped up "final" so I can hear what the clients approved.

This is a great way to work with the mixer and I wish everyone would do it.

And no, there are no standards. I was working on some mixes today that had to be lowered by -12 dB to travel through my analog chain without distorting! These mixes were NOT from my preferred mixers but are examples of the mixing engineers who haven't given a lot of thought to what is going to happen to their work after it leaves their studio.

I think the minimum standard should be the first scenario. Do the mix without buss add-ons that are used exclusively to make the track loud. By all means use plug-ins or outboard gear on the stereo bus, if it sounds good to you, but don't do it for volume purposes. Then cut a mix with the mastering plugs, or whatever you think you have to use, to make your clients not ask the question "why isn't it as loud as my favourite CD". This way, when the client becomes aware of mastering and decides to go a dedicated professional, the mastering engineer is not hamstrung by your attempt at mastering.

If you don't provide this option, you are not giving your client options and in my books, this is not completing the job properly.
Old 21st February 2008
  #18
Gear Addict
 
Spookym15's Avatar
 

Quote:
Originally Posted by Andy Krehm View Post
Yes, you are wrong.

A lot of mix engineers don't use a limiter on the master buss because they are careful not to have the sum of their tracks go over 0 dB, and lower is even better.

Some of the pro mixers I work with pump up their "final" mix for the client, limiters/vol. maximizers and/or clipping, but also print a mix that has no limiters/volume maximizers on the buss.

They send me both versions. The uncompressed, for ease of mastering, and the hyped up "final" so I can hear what the clients approved.

This is a great way to work with the mixer and I wish everyone would do it.

And no, there are no standards. I was working on some mixes today that had to be lowered by -12 dB to travel through my analog chain without distorting! These mixes were NOT from my preferred mixers but are examples of the mixing engineers who haven't given a lot of thought to what is going to happen to their work after it leaves their studio.

I think the minimum standard should be the first scenario. Do the mix without buss add-ons that are used exclusively to make the track loud. By all means use plug-ins or outboard gear on the stereo bus, if it sounds good to you, but don't do it for volume purposes. Then cut a mix with the mastering plugs, or whatever you think you have to use, to make your clients not ask the question "why isn't it as loud as my favourite CD". This way, when the client becomes aware of mastering and decides to go a dedicated professional, the mastering engineer is not hamstrung by your attempt at mastering.

If you don't provide this option, you are not giving your client options and in my books, this is not completing the job properly.
Thanks for that, it helps. I usually try to just work with the mastering engineer. But sometimes it is a well, we will figure it out later. So I am stuck wondering. It is also scary how many rap/hip hop guys think that Mastering is mixing and editing too.
Old 21st February 2008
  #19
Gear Addict
 

I just loaded in the Sia album. It sounds very good (suggesting that the mastering was done well) but -7.8 below 0 is totally unacceptable.

As a consumer of this CD I would like an explanation. I consider it defective.

Tony Cousins did the mastering (it's credited on the CD). I would consider him one of the top MEs in the world and always does amazing work. Does anyone know him? Would love to hear his thoughts on this.
Old 21st February 2008
  #20
Gear Maniac
 

It has happened to me twice! I mastered 2 separate full length CDs and gave the clients reference and masters. When the replicated CD's came back, all the levels dropped, quality sounded crappy. Upon investigation, I found out someone at the distribution company (who was paying for their manufacturing) took the liberty to make a copy of the master on his PC using Windows Media and sending in the copy for replication and keeping the master as a backup.
Old 21st February 2008
  #21
Gear Addict
 

Quote:
Originally Posted by MT Groove View Post
It has happened to me twice! I mastered 2 separate full length CDs and gave the clients reference and masters. When the replicated CD's came back, all the levels dropped, quality sounded crappy. Upon investigation, I found out someone at the distribution company (who was paying for their manufacturing) took the liberty to make a copy of the master on his PC using Windows Media and sending in the copy for replication and keeping the master as a backup.
That's absolutely inexcusable. What was the end result? Did the distribution company repress?
Old 22nd February 2008
  #22
Gear Maniac
 

Quote:
Originally Posted by grandmasters View Post
That's absolutely inexcusable. What was the end result? Did the distribution company repress?
The album never got officially released. They didn't pressed up that many units. He's redoing some songs so next go around hopefully they won't screw it up.
Old 23rd February 2008
  #23
JPF
Gear Nut
 
JPF's Avatar
 

thats too bad ,i've been A-B ing "Butterflies" in my PT sesh for this singer song writer i'm working with because i thought that cut just shined!was at a really decent level ,not just slammed up the ying yang...Sia's really talented WTF?
Old 23rd February 2008
  #24
Lives for gear
 
Don S's Avatar
 

I've heard from some venerated engineers to leave the out of limiter at -0.7 to keep consumer CD players from distorting. Is this correct or a safe way to go?
Old 23rd February 2008
  #25
Mastering
 

Quote:
Originally Posted by Don S View Post
I've heard from some venerated engineers to leave the out of limiter at -0.7 to keep consumer CD players from distorting. Is this correct or a safe way to go?
When? During the mixing or mastering? It's an entirely different point of view.

"Maximizing engineers" will need to use a protective brickwall peak limiter during mastering. There are now at least three of them that interpolate to calculate intersample peaks which could cause consumer players to distort and overload. These three limiters are the PSP Xenon, TC 6000 and powercore Brickwall, and I believe the Sonnox, but I'm not 100% sure. With reasonable confidence (underline REASONABLE) if not pushing the crap out of it, you can use these limiters without too many artifacts (underline TOO MANY) at their ceiling setting of 0.0.

However, when using an ordinary, non-oversampling limiter (like the majority out there) could very well require that you drop the output level (ceiling) to -0.3 to -0.7 dB because that would keep the true (interpolated) peaks below the 0.0 dBFS on the analog output. What a way to go... in the next few years you will find other limiters following that example as I do find using the true brickwall types does seem to be producing a bit purer sound from those consumer players. Or, when not, dialing in that lower output level.

BK
Old 23rd February 2008
  #26
JPF
Gear Nut
 
JPF's Avatar
 

Little Black Sandals

i don't know...just listened to Little Black Sandals on her myspace and it sounded good! GREAT! i mean all there,dynamic, ballsy enough, she's really throwing licks dwn ala Aretha on 10tnth power lol...what the heck do i know i'm just a bass player trying to learn how to make records lol cheers
MySpace.com - JP Fitting - Burbank, California - Alternative - www.myspace.com/jpfitting
Old 23rd February 2008
  #27
Lives for gear
 
Don S's Avatar
 

Quote:
Originally Posted by bob katz View Post
When? During the mixing or mastering? It's an entirely different point of view.

BK
Hi Bob,

I produce mostly classical recordings and try to use minimal processing. I've been using the stock limiter in Pyramix on the master as "insurance" against overs.
Thanks for the education on intersample peaks. Great book by the way! I learned quite a bit from it, however I don't remember reading about intersample peaks, was it in there? I'm sure what they were referring to was the majority of non-compensating limiters out there. (L2, L3, ect.) Which catagory does the PMX limiter fall into?
Old 23rd February 2008
  #28
Moderator
 
psycho_monkey's Avatar
 

Verified Member
Quote:
Originally Posted by Spookym15 View Post
This is interesting. From all the engineers I have talked to they always throw a limiter on the master buss, they usually leave 4 dB of head room for the mastering engineer. I usually check with the person who is doing the master to see how much head room they need. My experience is generally about 3db of head room and the mastering engineer will usually go up to -1db, because of the crappy converters in most stereos. I mix so that the loudest I ever get it at about -3db, and I do use a peak limiter to just to make sure that it does not go over. The most my mixes ever get to is about -2db at most, which is why the limiter is is on it.

Is this wrong? There should be something agreed upon between engineers and mastering engineers.
If you're going to use a limiter at all, set it at -0.5 (ie to make sure there's no intersample peaks). any lower and you're not leaving 3dB of headroom - you're re-defining -3dB as your new "0" and throwing away the extra resolution. When people talk about leaving headroom, it's room to catch the odd peak here and there - not to be dead space.

That said, in 6 years of studio work I've never worked for an engineer who limits the master mix bus. Compression yes - but never limiting.
Old 23rd February 2008
  #29
Mastering
 

Quote:
Originally Posted by Don S View Post
Hi Bob,

I produce mostly classical recordings and try to use minimal processing. I've been using the stock limiter in Pyramix on the master as "insurance" against overs.
Thanks for the education on intersample peaks. Great book by the way! I learned quite a bit from it, however I don't remember reading about intersample peaks, was it in there? I'm sure what they were referring to was the majority of non-compensating limiters out there. (L2, L3, ect.) Which catagory does the PMX limiter fall into?

Thanks for the nice comments on my book! I go into intersample peaks and intersample peak limiters in more detail in the second edition, but I believe a brief touch on it in the first edition.

I suspect the PMX is a simple, non oversampled limiter, or I probably would have heard otherwise. Anyone who is aware of a new oversampled (intersample) peak limiter, please let me know as I'm keeping a list for the 3rd edition.

BK
Old 25th February 2008
  #30
Lives for gear
 
mixerguy's Avatar
I hear all sorts of weird distortion (especially in the bass gtr) on that SIA cd.

YUCK

it sounds BAD
Post Reply

Welcome to the Gearslutz Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Similar Threads
Thread
Thread Starter / Forum
Replies
quietdrive / The Moan Zone
50
RainbowStorm / So Much Gear, So Little Time
25
Neve Sucks! / So Much Gear, So Little Time
78
chessparov / So Much Gear, So Little Time
13

Forum Jump
Forum Jump