Recent Videos
The Dark Arts | Air Mastering | Part 2
LoopTV present part 2 of 2 of ‘The Dark Arts’ Mastering series, our new extended interview and discussion with legendary mastering engineer Ray Staff. Recently filmed at AIR Mastering at the infamous AIR Studios in London – we explore the some of the questions surrounding the dark art of mastering.
AIR Mastering located within the confines of AIR Studios and is the newest division of the AIR Lyndhurst group. It builds upon over 40 years of experience in providing high end audio to world’s music industry. Established and headed up by MPG award winning mastering engineer Ray Staff and former Alchemy engineer Matt Colton.
Between them they have worked with artists including Led Zeppelin, The Rolling Stones, The Clash and Black Sabbath. Most recently Ray has mastered albums for Muse and Corinne Bailey Rae.
Here Ray talks about the modern Mastering engineers role and offers some tips on how to best prepare your audio for the mastering process. Along with some advice on why exporting mono Left and Right audio files over stereo interleaved can yield better audio fidelity.
Enjoy!
LoopTV – The Music Producers Channel
Related Videos
28 Responses to “The Dark Arts | Air Mastering | Part 2”
Leave a Reply
Subscribe To LoopTV
Sponsors
All NewsSocial
Search
About Us
Loopmasters music production video channel for producers, musicians and DJs. Watch producer interviews, music software tutorials, and music technology news & reviews, as we take you into the studios of some of the best music makers on the planet.
Contact
Email LoopTV
Credits
Framework by Graph Paper Press
Design by CueSounds
All content © 2012 by LoopTV
stereo interleaved : not really familiar with this term. could you point me in a direction to research more on this. and i’ve asked around and i keep hearing that a good sound card for a pc really helps with recording. i wonder if it has any affect on bouncing or output quality, presence after leaving your d.a.w??
any help would be nice.. big ups to the loop t.v crew and ray…thanks
Just landed on this place via Google research. I love it. This post change my perception and I am acquiring the RSS feeds. Cheers Up.
Might the reason be why non-interleaved stereo sounds better is because each channel get processed separately by the FX?
When using interleaved stereo, there are a lot of fx (both analog and digital) that process the the stereo image as ‘a wholed’, its still stereo, but you could say that the data used for some processes is mono.
For example a gated delay would give different results, if the treshold is different for each channel. Or a compressor. etc…
Good post, I can’t say that I agree with everything that was said, but very good information overall:)
Actually, I feel kind of sad that a professional mastering engineer can say something like that about stereo interleaved.
The term “interleaved” itself only refers to the way the sample data gets stored in the file, but that data remains EXACTLY THE SAME whether we decide to store them in the same stereo file or in separate mono files.
The format of a Wave file is actually simple: it just consist of “chunks” of information. The first chunk tells you the kind of RIFF file you´re dealing with. If it´s audio, then this should be “WAVE”. The following chunk tells you all info you need about the audio file: The number of channels (yes, you can accomodate 2, 3, 4, 5 or more channels in a single WAVE File), the bitrate, the samplerate, the bit resolution an so on.
The third chunk contains the actual audio data, which is going to be “interleaved” if we are talking about more than ONE channel. ¿What does “interleaved” mean?
It’s actually simple: as many of us know, PCM digital audio consists of a sucession of samples with a certain wordlenght (16 bit, 20 bit, 24 bit, etc) at a certain samplerate (44.1KHz, 48KHz, 96KHZ, etc). Each of this samples is just a binary number. Then follows the next, and then the other, and so on.
When you have two channels, you can do one of the following: either you write all the information of one entire channel first, and then the other, or you can write the first sample of channel #1 and then the first sample of channel #2. Then, the second sample of channel # 1 and then the second sample of channel # 2, and so on, to the end of the file.
That’s what the term “interleaved” means, and as you can see, it has nothing to do with the QUALITY of the samples stored.
Writing a simple program that can split a stereo Wave file in two mono’s is not complicated, nor is the contrary. The information in either case is EXACTLY THE SAME.
¿Why do some people perceive something else?
For the same reason that some people can get healed with salted water from their imaginary or even real disease: it’s called PLACEBO EFFECT.
Don´t take my word for it; please investigate, see for yourself.
If you want, you can get some deeper info about WAVE files in here:
https://ccrma.stanford.edu/courses/422/projects/WaveFormat/
From the UNIVERSITY OF STANFORD.
Greetings.
Bob Ludwig was talking along similar lines in a recent online video about how he always requests to work from the actual physical drive that the tracks were recorded onto.
Even though he is aware that a cloned drive will be bit for bit identical, he can ” perceive ” a difference in the sound between the 2.
2 of the most succesful mastering engineers the world has ever known clearly suffer terribly from the PLACEBO EFFECT.
If only they had ears as half as good as those legendary engineers at The University Of Stanford. Say HV Master, can you post up your CV so we can see all the classic recordings you have cut?:)
I must say that I, like most of the known world, respect Bob Ludwig enormously as mastering engineer and nobody, especially me, can show any work that may be equal or even comparable to his works.
But the subject of the FILES, the way data get stored in those files is not within his jurisdiction or any other sound engineer. It was defined by a conceal of ELECTRONIC ENGINEERS, an entirely different group of people to which I am honoured to belong.
Would you say that, because a famous chef said once that cows are not mammals that means that they are not?
Does this lack of knowledge about nature mean that what he cooks is NOT delicious?
We are talking about something that’s NOT subjective. When Data of multiple channels is stored within the same file gets interleaved whether you like it or not. You have no choice. You have nothing to do with it. Only the software.
If you don´t want to seriously dig into this, I can give you a simple example. Say, the continuous stream of samples of left channel is:
1,2,3,4,5,6
and the right channel stream is:
A,B,C,D,E,F
This is the way they would be stored in the file:
1,A,2,B,3,C,4,D,5,E,6,F
Absolutely no consequences in audio quality or usability.
Again, as much as I respect the ears of Bob Ludwig, Ted Jensen, Bob Katz and many others, this issues are not of their competence but that of the makers of the equipment they use.
Anyway, Puttinitonwax, if an electronic engineer tells you that R=V/I, please don’t insult them just because they don’t have any grammy award to their name. Ohm’s law is there for anyone that wants to learn it, including you and me. This is INDEPENDENT of your subjective perception, as DIGITAL ELECTRONICS are.
If you want to learn something useful, just use the link posted above. The University of Stanford may not be a Grammy Award winners but then it doesn’t need to, because ELECTRONICS ENGINEERING are not the competence of the GRAMMY academy.
I, as a member of the jury, know that very well.
But of course, that’s up to you. If you want to learn something, just read the post. If you don’t, well, don’t read anything.
But there’s no need to insult anybody just because you don’t know what are they really talking about.
Greetings.
hvmaster wrote:
this issues are not of their competence but that of the makers of the equipment they use.
Indeed.
I would add myself to the list of mastering engineers who hear differences with interleaved, and indeed many other areas of digital audio which should sound the same because the numbers are the same, but they don’t.
I’d agree that it is 99% probable that any issues are hardware related, rather than (in this case) an inherent problem with interleaving. My personal impression with this particular issue the problem lies with the exporting of files from the DAW that the majority of records are made on (I’ll avoid naming it, as a pro should never blame his tools, if you get my drift).
However, as audio professionals we are going to try for the best possible sound, and if there are issues we will try to avoid it. As it is already too late by the time we mastering engineers get it, it’s better to avoid the problem in the 1st place by advising our clients to give us 2 mono files.
So, everyone is right. Indeed an interleaved file is numerically identical to the monos it breaks down into, but we are hearing real differences due to the limitations of hardware.
To quote Bob Ludwig’s wisdom: “never turn your back on digital.”
As an adjunct, I’d defend myself and my fellow professionals against the assertion that it is purely placebo effect. We can all get fooled occasionally by the placebo effect in audio (as well as anything else). There is however a large number of the most respected audio engineers independently coming to the same conclusions about certain aspects of digital audio, and this is compelling.
You have somewhat shot yourself in the foot though already, in your 1st post you accuse Ray & the profession of being fooled by the placebo effect, and in your second the blame is put on the equipment. As a scientist surely this distinction is crucial, one requires equipment re-design the other doesn’t. Fundamental difference.
Andy Jackson
TubeMastering
hvmaster – your pedantic approach to this undermines the point you are trying to make. It doesn’t really matter whether and interleaved file is the same as L/R mono files – these are just the names that they are referred to by. The evidence of our ears is what counts and the (so-called) interleaved files sound different from the (so-called) L/R mono ones. What is more the interleaved files made using different DAWs also sound different from each other.
They shouldn’t and the data is identical but something is affecting the way they sound.
I am sad that someone from a respected establishment such as Stanford is not more interested in finding out why!
I offer an apology if I hurted any feelings; that was never my intention.
Andy: I think I understand why you say that thing of shooting my own foot. I tried to express an idea and probably didn’t succeed on that.
Let me try again, please.
About the PERCEPTION, yes, it´s totally subjective, and if you say that your steak is lacking some chilli, there’s nothing that I or anybody else can say about it. It´s YOUR perception what counts, and nothing else.
But if the chef says that he used 4 gr of chili on everyone´s dish, that´s something entirely different. It is easily verifiable, and after that there´s no argue. One thing is that you’d like some more and saying that the chef gave you less than anybody else is another.
The first one is a matter of perception, and the second, a lie.
When I said that hearing two things that are EXACTLY the same and saying that they´re different falls between the category of placebo, I wasn´t saying anything different.
What I said something later about “issues that are not of their competence”, I was talking about the design of the equipment.
This means that the way a file gets stored in a file matters to a systems designer, not a sound engineer. Double mono is the same as Stereo interleaved because of their DESIGN, not to their use.
Now, if you are listening to one using one equipment and the other using another, well, this is just common sense.
A friend of mine just suggested me to suggest you to perform the following experiment:
“Open”, or “Load” or “Import” three sound files to the same project. The command you use depends, of course, on the DAW you are using. This three files must be created one after another using the same project with the same DAW. The first must be STEREO INTERLEAVED and the other two, the L and R mono files.
Be careful with the placing, doh. The stereo file must be located sample-accurate with respect to the two channels of the mono-files. No volume change in anything. Just pan the “R” file fully to the right and the “L” to the left.
You have to pay attention to what you´re doing because it´s really easy to do this the wrong way.
Now, invert the polarity of the stereo file. You should do this by processing the file, instead of using a real time plugin, because the latency introduced may lead to errors. Of course, you can always use compensation, if you know how to do it.
Now, hit playback.
If you hear a single sound, even unmeasurable, this only means that you did something wrong.
Both signals are going to totally cancel each other because they are EXACTLY THE SAME.
What’s more: every time you hit “play” on your DAW, if you have more that ONE MONO TRACK, you are going to be hearing INTERLEAVED AUDIO, whether you want it or not. And this is not optional: is the very nature of using computers for audio.
Unlike other digital devices, which can use concurrent processes, computers use a single processor for everything, so every command and data passes through in a SEQUENTIAL way, meaning “one after another”, and audio is no exception.
Computers with multiple processors are no exception to this rule, because for all practical purposes, they are being “seen” by the operating system as one.
Hit play, and the program will search for first sample of track # 1, then first sample of track # 2, then first sample of track # 3 and so on, until it reaches the last track. All this data is then “mixed” or “summed” in software or hardware, depending on your system and finally sent to the interface. Then the program searches for second sample of track #1, second sample of track # 2, second sample of track # 3 and so on.
In playback, they are always “interleaved”. There´s no choice. The difference between two MONO files and one STEREO file is the way they get organized within the file. Once they are “loaded”, “opened” or “imported” into a project, there is no difference. They are always interleaved.
I know people with “golden ears” who can tell if you are using two different CD players or just one, but this is an entire different subject.
There´s simply no way of building two IDENTICAL devices, when it comes to the analog part of the matter, which is what they are hearing. We can be talking about the same model and brand, they´re still different.
Subtle differences in the components arise, even when they´re the same model of the same brand and were made of the same sillicon block the same day by the same operator in the same plant.
We, humans, cannot produce two entirely identical pieces of material yet. You use two different devices, you hear different things if your ears are sharp enough, and I´m shure Bob Ludwig´s are
Tony, I´m sorry if you feel this approach “pedantic”. It was not meant to be like that. I was just trying to make a simple point : perception is perception and facts are facts.
Look: If I killed the animal, if there are still some feathers in my kitchen, tell me you don´t like my way of preparing chicken, but don´t tell me it´s NOT CHICKEN!
Again, sorry for the inconveniences. It´s probably just a matter to learn how to keep my big mouth shut.
Greetings.
HV, NEVER stop expressing your opinions and beliefs on here fella.
We LOVE debate, and this has been one of the most interesting and involved yet.
We all just want to learn and get better at what we do.
Mel
Loop Studio
I haven’t read all through the responses above, so I hope I’m not repeating something someone has already said. But in response to hvmaster, I would say this; granted the data may be the same in interleaved and non-interleaved files on the drive, but the the way the data is reassembled (played-back) may not be identical.
The real time effects of a processor playing back two differently constructed files may alter the sound. I’m speculating here – I don’t know if that is or even could be the answer, but just looking at the data on the drive isn’t the be-all and end-all.
I would imagine something like this is what Ray (and other top-flight mastering engineers) are hearing. And I don’t doubt they’re hearing something. It’s something which needs investigating.
… when folks read news online, their eyes go for text first, particularly captions and summaries, and graphics only later. — Bryan & Jeff Eisenberg, usability consultants and authors
Hey. pleasant webpage, would you prefer to share links?
Any chance you know how to get rid of the nofollow tag on your blog comment? I just start with WP and I couldn’t find it … Sorry for being off topic. Thanks ! (Pardon my English i’m French)
I found your weblog website on google and test a few of your early posts. Proceed to keep up the excellent operate. I just further up your RSS feed to my MSN News Reader. Searching for ahead to reading more from you in a while!…
Your idea is ofutstanding; the issue is something that not enough people are speaking intelligently about. I am very happy that I stumbled across this in my search for something relating to this.
I enjoy you because of your entire efforts on this web page. My mom enjoys engaging in investigation and it’s really easy to understand why. Most people know all about the compelling medium you present efficient guidelines through your website and in addition attract participation from others on that article so my child has always been understanding so much. Enjoy the remaining portion of the new year. You are always performing a glorious job.
I enjoy you because of your own hard work on this website. My mother really loves going through internet research and it’s easy to see why. A lot of people know all relating to the compelling medium you give efficient steps via this website and in addition encourage response from some others on this point while our girl has been becoming educated a great deal. Enjoy the remaining portion of the year. You are always doing a useful job.
These are a selection of what you should know before you spend your dough with a web designer in north wales!
Just want to say your article is as astounding. The clearness in your post is just great and I can assume you’re an expert on this subject. Fine with your permission let me to grab your feed to keep up to date with forthcoming post. Thanks a million and please carry on the enjoyable work.
There is a secret system for making money. Time is short and the opportunity is limited. The software is super secret right now but if you get there fast, you can end up as one of the exclusive limited number of users. The software was created by a guy who became fed up with all of the usual money making methods that online marketers typically use. What this software actually does is something that you have never seen before. And I guarantee you’ll understand the genius of it within 30 seconds. The best part about this thing is that you will see that it works right away. There’s no waiting for anything to kick in. Results can happen in as little as 30 seconds. Go ahead and take a look at the video presentation and see what you think -> http://tinyurl.com/chdiscount
Hiya, I am really glad I have found this info. Today bloggers publish just about gossip and internet stuff and this is actually frustrating. A good web site with interesting content, that is what I need. Thank you for making this web site, and I’ll be visiting again. Do you do newsletters? I Can not find it.
I like your post and appreciate it and going to subscribe. I hope you will rock my world in future
When I originally commented I clicked the -Notify me when new comments are added- checkbox and now each time a comment is added I get four emails with the same comment. Is there any way you can remove me from that service? Thanks!…
Learned a lot from your post, write some more.
Hello.
It’s been a while since the last time I wrote here.
This is for AES members : please look for the latest article from Andres Mayo about a congress on mastering. Bob Ludwig, Bernie Grundman and many other big names attended, and shure enough, there it was : the myth of the stereo interleaved.
Some of them reported a few differences between one and the other … well, you get the idea.
Finally, an engineer from apple computer explained the process of reading and loading mono and stereo files, and EVERYBODY in the room understood. There is no difference whatsoever.
If anyone wants to read it, this is the paper : “JAES_V58_1_2_PG65.pdf” and the title, “Mastering in an ever expanding Universe”, hosted by Andres Mayo.
Just to illustrate the point, the following list of numbers correspond to the first bytes in a very short wave file:
82 73 70 70 246 10 0 0 87 65 86 69 102 109 116 32
The first four are to be read as ascii, the next four as little endian numbers, and the next eight as ascii again. Doing this, we get:
“RIFF, 2816 (bytes in size),WAVE fmt ”
So, it becomes apparent that there are no esoterics here. The information in a wave file is presented with no ambiguities in a very simple manner, so there is no room for confusion. It is just a list of numbers.
Dual mono and stereo interleaved are exactly the same thing.
Greetings.
It’s gonna be another additional information I may know. It is really good to explore and read over some websites like this which really is a good source of knowledge.