Deliberately introducing clipping?

[[email protected]]

I've always been under the impression that avoiding clipping is a basic tenet of digital audio. Looking at this mastering tutorial, if I understand what this guy is saying, he deliberately introduces a limited amount of clipping - 2:30 "I wanted to get a little bit of clipping on the A to D".

??

http://youtu.be/2V9d6zQzcrg

[Mike Rivers[_2_]]

On 3/8/2014 2:36 AM, wrote:
> I've always been under the impression that avoiding clipping is a
> basic tenet of digital audio. Looking at this mastering tutorial, if
> I understand what this guy is saying, he deliberately introduces a
> limited amount of clipping - 2:30 "I wanted to get a little bit of
> clipping on the A to D".

It's one of the techniques used in order to make an audio clip stand out
in the crowd by appearing to be louder. Everybody says they don't want
distortion, but everybody wants their audio to sound louder than
everybody else.

Use it or not as you see fit. Got something that needs to attract
people? Make it louder. Got people interested in your work already? Make
it sound better (not louder).



--
For a good time, visit http://mikeriversaudio.wordpress.com

[Frank Stearns]

writes:

>I've always been under the impression that avoiding clipping is a basic ten=
>et of digital audio. Looking at this mastering tutorial, if I understand wh=
>at this guy is saying, he deliberately introduces a limited amount of clipp=
>ing - 2:30 "I wanted to get a little bit of clipping on the A to D".

It appears to be part of the loudness wars BS... "I want to get it just a little
louder before I even start mastering." But a small bit of credit: "but not more than
10-15 samples of clipping as more that and it's audible."

Maybe, maybe not. In pop music, probably doesn't matter. In acoustic/classical music
15 samples of clipping could be very audible, and not in a flattering way.

Brassplyer is right, it's generally not a good idea to introduce digital clipping
(or analog clipping, for that matter). But that's where a big chunk of the industry
has been for quite a while now -- anything to get it just a little louder than the
next guy's.

To me, it's a sad commentary on the utter lack of genuine aesthetic; a prime example
of what I call "manufactured music." Everything, absolutely everything, is slapping
you in the face, relentlessly, with "pre-clipping" and serial compression to make
sure that happens. The tune becomes another "big and loud" cliche; the man has done
his job well.

I still watch how some younger people interact with "music" (more like modulated
white noise in the background to plug any holes of contemplative silence in their
lives). "Loud" in this context is all they've known. They've been so horribly
short-changed.

Frank
Mobile Audio
--

[geoff]

On 8/03/2014 8:36 p.m., wrote:
> I've always been under the impression that avoiding clipping is a basic tenet of digital audio. Looking at this mastering tutorial, if I understand what this guy is saying, he deliberately introduces a limited amount of clipping - 2:30 "I wanted to get a little bit of clipping on the A to D".
>
> ??
>
> http://youtu.be/2V9d6zQzcrg
>


That you have asked here indicates that yo do have some suspicion about
this guys advice.

Clipping = poor man's aural exciter. And even those are sus in the
first place ....!

geoff

[Peter Larsen[_3_]]

geoff wrote:

> On 8/03/2014 8:36 p.m., wrote:

>> I've always been under the impression that avoiding clipping is a
>> basic tenet of digital audio. Looking at this mastering tutorial, if
>> I understand what this guy is saying, he deliberately introduces a
>> limited amount of clipping - 2:30 "I wanted to get a little bit of
>> clipping on the A to D".

>> http://youtu.be/2V9d6zQzcrg

> That you have asked here indicates that yo do have some suspicion
> about this guys advice.

> Clipping = poor man's aural exciter. And even those are sus in the
> first place ....!

All kinds of deriding comments could be made, anyway the guys job was to
demonstrate how to use a whacking lot of waves plug-ins, and that he did,
and while it resulted in all kinds of abominable sonic aberrations - such as
a totally detached treble - it also is a demonstration of how to address
invidudual sonic problems should they occur. So yes, it is worth seeing and
listening to on reasonably good loudspeakers. You do not have to agree in
the production method and aims to respect the guys craftmanship.

> geoff

Kind regards

Peter Larsen

[[email protected]]

Peter Larsen wrote: "...listening to on reasonably good loudspeakers. You do not have to agree in the production method and aims to respect the guys craftmanship. "


Craftsmanship?? Since when is destroying the music craftsmenship?

As an engineer I can do no harm to my client's music - whether reproduced live or in recorded form - even if they tell me to do it or bribe me.

[Scott Dorsey]

> wrote:
>I've always been under the impression that avoiding clipping is a basic ten=
>et of digital audio. Looking at this mastering tutorial, if I understand wh=
>at this guy is saying, he deliberately introduces a limited amount of clipp=
>ing - 2:30 "I wanted to get a little bit of clipping on the A to D".

There are a lot of people out there today who are unfamiliar with actual
live acoustic music. They don't actually know what instruments really sound
like. And, many of these people associate loudness with the sound of clipping
because that's how they have listened to loud music much of their life.

Correspondingly, adding clipping can give these people a sense of music
being louder than it really is.

Now, why anyone would want to do something so horrible to distribution
medium, I don't know. We live in an age with a lot of "mastering engineers"
who really don't care about the long-term prospects of their product.
At some point, all these clipped recordings are going to be perceived as
kids as sounding like the kind of records their parents listened to.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[None]

> wrote in message
...
> As an engineer ...

LOL! You will never be an engineer. You're just too stupid.

[Les Cargill[_4_]]

wrote:
> Peter Larsen wrote: "...listening to on reasonably good loudspeakers.
> You do not have to agree in the production method and aims to respect
> the guys craftmanship. "
>
>
> Craftsmanship?? Since when is destroying the music craftsmenship?
>

Well.... distorted guitars, anybody?

> As an engineer I can do no harm to my client's music - whether
> reproduced live or in recorded form - even if they tell me to do it
> or bribe me.
>

--
Les Cargill

[John Williamson]

On 09/03/2014 10:45, wrote:
> Peter Larsen wrote: "...listening to on reasonably good loudspeakers. You do not have to agree in the production method and aims to respect the guys craftmanship. "
>
>
> Craftsmanship?? Since when is destroying the music craftsmenship?
>
> As an engineer I can do no harm to my client's music - whether reproduced live or in recorded form - even if they tell me to do it or bribe me.
>
As an engineer, it is easy to do harm to your client's music. It's often
easier than *not* doing harm to it.

I suspect that what you mean is that you will *try not to*...

--
Tciao for Now!

John.

[hank alrich]

Les Cargill > wrote:

> wrote:
> > Peter Larsen wrote: "...listening to on reasonably good loudspeakers.
> > You do not have to agree in the production method and aims to respect
> > the guys craftmanship. "
> >
> >
> > Craftsmanship?? Since when is destroying the music craftsmenship?
> >
>
> Well.... distorted guitars, anybody?

JUst horrible!!! LIke the guy told Zappa, that'll never sell.

> > As an engineer I can do no harm to my client's music - whether
> > reproduced live or in recorded form - even if they tell me to do it
> > or bribe me.
> >


--
shut up and play your guitar * HankAlrich.Com
HankandShaidriMusic.Com
YouTube.Com/WalkinayMusic

[Mike Rivers[_2_]]

On 3/9/2014 6:45 AM, wrote:

> Craftsmanship?? Since when is destroying the music craftsmenship?

To some, making it louder and drawing more attention to it is a craft.
There are many ways to do this, clipping is one. You wouldn't want to
clip a chamber music ensemble to make it louder, but it's effective on a
garage band where things are distorted on purpose as part of the sound.

> As an engineer I can do no harm to my client's music - whether
> reproduced live or in recorded form - even if they tell me to do it
> or bribe me.

If you're working on music for which clipping is clearly inappropriate,
then by all means, don't do it. You probably wouldn't even be asked to
do so. On the other hand, if you're working on music where distortion is
an integral part of the sound, you aren't likely to be harming it by
driving something in the mastering signal path into clipping. It's not
life-threatening.


--
For a good time, visit http://mikeriversaudio.wordpress.com

[Sean Conolly]

> wrote in message
...
I've always been under the impression that avoiding clipping is a basic
tenet of digital audio. Looking at this mastering tutorial, if I understand
what this guy is saying, he deliberately introduces a limited amount of
clipping - 2:30 "I wanted to get a little bit of clipping on the A to D".

??


It depends on what you're planning on doing to the music - if it's going to
be limited and compressed in post then there's probably no harm done by
letting it clip a little during tracking. The question is why would someone
want to?

I'll speculate that the concern is about the noise floor and gain staging:
there is always (as far as I know) an analog stage before the A/D converter,
and the noise floor of this stage is fixed relative to the converter: eg X
dB below 0 dBFS. Even though the dynamic range of the converter itself can
handle any peaks (by lowering the signal), the noise floor in the track
increases as the signal into the A/D chip is reduced. Depending on the
quality of the hardware it may be insignificant, but it's always non-zero
and some case becomes the real limit to the dynamic range of the interface.

So I can understand someone deciding to trade off some peaks that would be
lost eventually anyway in favor of reducing the noise floor - especially if
you're working with older or cheaper hardware. That said, I don't agree with
it - even with my old cheapo Emu interface the noise floor is low enough to
be unoticable in most cases, and is low enough to easily clean up if I'm
feeling pedantic about it.

For just about any decent modern interface I think it's a complete
non-issue - just keep it well below clipping and it'll be fine.


Sean
(Anticipating many corrections .... )

[Peter Larsen[_3_]]

Sean Conolly wrote:

> It depends on what you're planning on doing to the music - if it's
> going to be limited and compressed in post then there's probably no
> harm done by letting it clip a little during tracking. The question
> is why would someone want to?

Because it can be the cleanest limiting possible if you keep it shorter than
some 2 milliseconds, depending on context/type of signal(waveform).

> Sean

Kind regards

Peter Larsen

[Sean Conolly]

"Peter Larsen" > wrote in message
k...
> Sean Conolly wrote:
>
>> It depends on what you're planning on doing to the music - if it's
>> going to be limited and compressed in post then there's probably no
>> harm done by letting it clip a little during tracking. The question
>> is why would someone want to?
>
> Because it can be the cleanest limiting possible if you keep it shorter
> than some 2 milliseconds, depending on context/type of signal(waveform).
>

Well it's certainly going to be the least limiting you can get away with.
Being naturally pedantic I'd still want to look at the resulting waveform
before I'd adopt it as a normal practice for myself. I'm not going to assume
it'simple truncate even though it should be.

Sean

[geoff]

On 10/03/2014 8:21 a.m., Peter Larsen wrote:
> Sean Conolly wrote:
>
>> It depends on what you're planning on doing to the music - if it's
>> going to be limited and compressed in post then there's probably no
>> harm done by letting it clip a little during tracking. The question
>> is why would someone want to?
>
> Because it can be the cleanest limiting possible if you keep it shorter than
> some 2 milliseconds, depending on context/type of signal(waveform).

As long as it's nopt clipping to FS on a device that's going to go make
D-As go ape in unpredictable ways and spew out god knows what.

geoff

[Peter Larsen[_3_]]

geoff wrote:

> On 10/03/2014 8:21 a.m., Peter Larsen wrote:
>> Sean Conolly wrote:

>>> It depends on what you're planning on doing to the music - if it's
>>> going to be limited and compressed in post then there's probably no
>>> harm done by letting it clip a little during tracking. The question
>>> is why would someone want to?

>> Because it can be the cleanest limiting possible if you keep it
>> shorter than some 2 milliseconds, depending on context/type of
>> signal(waveform).

> As long as it's nopt clipping to FS on a device that's going to go
> make D-As go ape in unpredictable ways and spew out god knows what.

Some opamps do technically fascinating things to the waveform when they
clip, yes. I don't advocate clipping in tracking, but I also do not loose
sleep over it if it happens because I know what Auditions unclipper can
unclip to - at least for me - inaudibility.

All should try to make some deliberately clipped samples, easy when
processing 16 bit digital, align average level and listen to what is
audible. I did that with a listening panel once upon a time, interesting
results. You need to know what your equipment does when it runs out of bits
..... and you most definitely want to know in advance if it does not clip
cleanly.

> geoff

Kind regards

Peter Larsen

[Sean Conolly]

"Peter Larsen" > wrote in message
k...
> geoff wrote:
>
>> On 10/03/2014 8:21 a.m., Peter Larsen wrote:
>>> Sean Conolly wrote:
>
>>>> It depends on what you're planning on doing to the music - if it's
>>>> going to be limited and compressed in post then there's probably no
>>>> harm done by letting it clip a little during tracking. The question
>>>> is why would someone want to?
>
>>> Because it can be the cleanest limiting possible if you keep it
>>> shorter than some 2 milliseconds, depending on context/type of
>>> signal(waveform).
>
>> As long as it's nopt clipping to FS on a device that's going to go
>> make D-As go ape in unpredictable ways and spew out god knows what.
>
> Some opamps do technically fascinating things to the waveform when they
> clip, yes. I don't advocate clipping in tracking, but I also do not loose
> sleep over it if it happens because I know what Auditions unclipper can
> unclip to - at least for me - inaudibility.
>
> All should try to make some deliberately clipped samples, easy when
> processing 16 bit digital, align average level and listen to what is
> audible. I did that with a listening panel once upon a time, interesting
> results. You need to know what your equipment does when it runs out of
> bits .... and you most definitely want to know in advance if it does not
> clip cleanly.

Agreed - but I recommend looking as well as listening.

Sean

[[email protected]]

There is a fundamental technical difference clipping the A/D compared to clipping in the analog domain and that difference is due to ALIASiNG.
Consider clipping an 8 kHz tone in the analog domain. The harmonics are 16, 24 etc. Everything above 16 will be filtered off by the anti-alias filter.

[Sean Conolly]

> wrote in message
...
There is a fundamental technical difference clipping the A/D compared to
clipping in the analog domain and that difference is due to ALIASiNG.
Consider clipping an 8 kHz tone in the analog domain. The harmonics are 16,
24 etc. Everything above 16 will be filtered off by the anti-alias filter.
Now consider clipping the same 8 kHz tone in the A/D. The same harmonics
are created but since we are now after the anti alias filter, the higher
harmonics are not removed and they will be in fact aliased or folded back
down into the audible range. The aliased tones are not even harmonically
related after folding and will sound particularly nasty.

I'm not saying you should never clip the A/D, but be aware of the
technology. If the clipping is very brief or if you want grunge, then go
for it.

----------------------------------------

I believe most modern converters are running the A/D at a higher sample rate
and then filtering down to the specifed sample rate, instead of relying on
an analog filter only.

But even without that I can't see how clipping the wave at them moment it is
converted to digital is going to cause aliasing. The truncation is just data
until something converts it back to analog.

Sean

[[email protected]]

Scott Dorsey wrote: "We live in an age with a lot of "mastering engineers" who really don't care about the long-term prospects of their product. "


I used to think that way - blame loudness & clipping all on the engineers. Over at GearSlutz they have explained to me that recording/mixing/mastering are service businesses, and that if you wanted to have food on the table and keep a roof over your head, yeah, you could educate your clients(the musicians, their producers) all you want, but then you would just give the client a hypercompressed, brickwall-limited ultra-loud master, if that's what they requested.


So it boils down to educating the end-consumer as to what a good record sounds like, and what knob they should use if they want it louder.

[Scott Dorsey]

> wrote:
>Scott Dorsey wrote: "We live in an age with a lot of "mastering engineers" =
>who really don't care about the long-term prospects of their product. "
>
>
>I used to think that way - blame loudness & clipping all on the engineers. =
> Over at GearSlutz they have explained to me that recording/mixing/masterin=
>g are service businesses, and that if you wanted to have food on the table =
>and keep a roof over your head, yeah, you could educate your clients(the mu=
>sicians, their producers) all you want, but then you would just give the cl=
>ient a hypercompressed, brickwall-limited ultra-loud master, if that's what=
> they requested.

Well, that's true, and again it's mostly the result of the client not
caring about long-term prospects. Which I suppose is wise given the
market for music in general today.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."