0dBFS+ Levels in Digital Mastering

[Gray_Wolf]

0dBFS+ Levels in Digital Mastering

http://service.tcgroup.tc/media/Level_paper_AES109(1).pdf

Interesting. I have noticed that in my better recorded stuff
that the maximum levels tend to be at least -3dB and some as low as -6dB.

There are a few songs I have that the audio level, in my poor acoustic
environment, ranges from too low to too loud in my quieter moods.
(kman eat your heart out)

IIRC they were recorded at least 20 years ago.

GW

[[email protected]]

gray_wolf wrote: "kman, eat your heart out"


I never said ALL engineers were guilty of bad practices, and
besides, I came to realize two years ago where the "hot CD"
edict came from.

Semi-related, look at the levels on the DR Database entry
for the 1987 CD release of Billy Joel's "The Stranger"

http://dr.loudness-war.info/album/view/88060

Only a few of the tracks peak to within 1dB of full scale,
with one track peaking 8dB below it(!)

I OWN this edition, and it sounds just great on my
players.

[[email protected]]

(Joel, continued):

OTOH, all of the tracks on this this 2004 "remaster" of The
Stranger appear to have been dynamically compressed and
or peak-limited:
http://dr.loudness-war.info/album/view/94338

The sound is strained and edgy, and lacks
the emotional cues of the original CD in my previous post.

[[email protected]]

Whoever decreed that ALL audio tracks MUST peak within
1/1,000,00dB of full scale must be on something strong!!

[geoff]

On 10/10/2015 2:45 p.m., wrote:
> (Joel, continued):
>
> OTOH, all of the tracks on this this 2004 "remaster" of The
> Stranger appear to have been dynamically compressed and
> or peak-limited:
> http://dr.loudness-war.info/album/view/94338
>
> The sound is strained and edgy, and lacks
> the emotional cues of the original CD in my previous post.
>

Peaks touching 0dBFS (for one sample) has nothing to do with 'loudness',
dynamic range, or "strained harsh sound", but you've finally grasped
that, haven't you ?u.

Actually 0dBFS peaks can cause problems on many DAs, but backing off
half a dB should fix that. But not those symptoms you meantion.

geoff

[Nil[_2_]]

On 10 Oct 2015, geoff > wrote in
rec.audio.pro:

> Peaks touching 0dBFS (for one sample) has nothing to do with
> 'loudness', dynamic range, or "strained harsh sound", but you've
> finally grasped that, haven't you ?u.

No, it will take at least another 2 years before he gets it.

> Actually 0dBFS peaks can cause problems on many DAs, but backing
> off half a dB should fix that. But not those symptoms you
> meantion.

[Ty Ford]

On Saturday, October 10, 2015 at 2:07:04 PM UTC-4, Nil wrote:
> On 10 Oct 2015, geoff > wrote in
> rec.audio.pro:
>
> > Peaks touching 0dBFS (for one sample) has nothing to do with
> > 'loudness', dynamic range, or "strained harsh sound", but you've
> > finally grasped that, haven't you ?u.
>
> No, it will take at least another 2 years before he gets it.
>
> > Actually 0dBFS peaks can cause problems on many DAs, but backing
> > off half a dB should fix that. But not those symptoms you
> > meantion.

SIDEBAR

I was mixing and mastering a project on Logic Express a year or so go. The tracks had been recorded by someone else. I was called in to mix and master.. I was having a particularly difficult time mixing one track. The vocal would sound pretty grungy during short sections, but no level indicators were showing anything alarming.

Finally I put a compressor in the two mix and without any adjustment, it was showing 8 dB gain reduction! I did a global reduction of 8 dB and the vocal grunge stopped and the whole mix opened up. Somewhere in Logic Express a digital buss was being overloaded, but there was no place to see it.

This doesn't happen in Pro Tools.

Regards,

Ty Ford

[Scott Dorsey]

Ty Ford > wrote:
>
>Finally I put a compressor in the two mix and without any adjustment, it wa=
>s showing 8 dB gain reduction! I did a global reduction of 8 dB and the voc=
>al grunge stopped and the whole mix opened up. Somewhere in Logic Express a=
> digital buss was being overloaded, but there was no place to see it.
>
>This doesn't happen in Pro Tools.

Oh, it sure used to! Pro Tools 4 did that kind of thing all the time. These
days all the Pro Tools busses are 32-bit floats, though, so they are next to
impossible to overload.
--scott

--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Mike Rivers[_2_]]

Ty Ford wrote:
>> Somewhere in Logic Express a=
>> >digital buss was being overloaded, but there was no place to see it.

>> >This doesn't happen in Pro Tools.

On 10/11/2015 10:25 AM, Scott Dorsey wrote:

> Oh, it sure used to! Pro Tools 4 did that kind of thing all the time. These
> days all the Pro Tools busses are 32-bit floats, though, so they are next to
> impossible to overload.

Most DAWs from at least the last five years are like that. Even on the
PreSonus StudioLive console, you can send 16 full scale channels to the
main stereo bus, scrape the output meters off the pin with the main
fader, and all the distortion goes away.



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

[JackA]

On Friday, October 9, 2015 at 9:40:06 PM UTC-4, wrote:
> gray_wolf wrote: "kman, eat your heart out"
>
>
> I never said ALL engineers were guilty of bad practices, and
> besides, I came to realize two years ago where the "hot CD"
> edict came from.
>
> Semi-related, look at the levels on the DR Database entry
> for the 1987 CD release of Billy Joel's "The Stranger"
>
> http://dr.loudness-war.info/album/view/88060
>
> Only a few of the tracks peak to within 1dB of full scale,
> with one track peaking 8dB below it(!)

Sounds like a darn fine reason to buy it!!!
I'll cherish the multi-tracks, don't need a Sony Boy to remaster. Thanks.

Jack

>
> I OWN this edition, and it sounds just great on my
> players.

[JackA]

On Friday, October 9, 2015 at 9:47:34 PM UTC-4, wrote:
> Whoever decreed that ALL audio tracks MUST peak within
> 1/1,000,00dB of full scale must be on something strong!!

I was told, no clipping! Otherwise, Loudness Wars would have never existed.

Jack

[geoff]

On 12/10/2015 4:03 p.m., JackA wrote:
> On Friday, October 9, 2015 at 9:47:34 PM UTC-4, wrote:
>> Whoever decreed that ALL audio tracks MUST peak within
>> 1/1,000,00dB of full scale must be on something strong!!
>
> I was told, no clipping! Otherwise, Loudness Wars would have never existed.
>
> Jack
>


Loudness and loudness wars have nothing to do with clipping.

geoff

[Trevor]

On 12/10/2015 4:27 PM, geoff wrote:
> Loudness and loudness wars have nothing to do with clipping.

Rubbish. To gain maximum "loudness", moderate to severe clipping is
commonly used. Sure you could compress without clipping, but just look
at the waveform of nearly every pop CD released in the last decade if
you want to see clipping *IS* part of the loudness wars.

Trevor.

[geoff]

On 12/10/2015 7:32 p.m., Trevor wrote:
> On 12/10/2015 4:27 PM, geoff wrote:
>> Loudness and loudness wars have nothing to do with clipping.
>
> Rubbish. To gain maximum "loudness", moderate to severe clipping is
> commonly used. Sure you could compress without clipping, but just look
> at the waveform of nearly every pop CD released in the last decade if
> you want to see clipping *IS* part of the loudness wars.
>
> Trevor.


That's where generalisations are unhelpful when one is trying to be concise.

No CD that I've purchased in the last decade exhibit clipping that I've
noticed, though a few are over-compressed to hell (not many
fortunately). And some I've been prompted to actually check ! Not much
'current' pop though I concede.

Loudness is achieved by extreme compression and/or limiting. If digital
clipping occurs, that is a *technical error* - not inherently part of
the hyper-compression process. And the same 'loudness' could be achieved
without any clipping.

That it may be a common error, or a deliberate misuse, is a different story.

You can also clip to hell, and *not* have hyper-compression or loudness.
Lots of distortion though. But you know that.

geoff

[Trevor]

On 12/10/2015 6:43 PM, geoff wrote:
> On 12/10/2015 7:32 p.m., Trevor wrote:
>> On 12/10/2015 4:27 PM, geoff wrote:
>>> Loudness and loudness wars have nothing to do with clipping.
>>
>> Rubbish. To gain maximum "loudness", moderate to severe clipping is
>> commonly used. Sure you could compress without clipping, but just look
>> at the waveform of nearly every pop CD released in the last decade if
>> you want to see clipping *IS* part of the loudness wars.
>>
>
> That's where generalisations are unhelpful when one is trying to be
> concise.
>
> No CD that I've purchased in the last decade exhibit clipping that I've
> noticed,

Have you actually looked? Or perhaps you don't buy any pop music?


> though a few are over-compressed to hell (not many
> fortunately). And some I've been prompted to actually check ! Not much
> 'current' pop though I concede.

So not relevant to what I said in that case. While I don't buy much pop
myself, I have looked at very many of those disks to see why so many big
artists (who can afford the best people) sound so bad. :-(


> Loudness is achieved by extreme compression and/or limiting. If digital
> clipping occurs, that is a *technical error*

Yes, but I am NOT talking about theory, I am talking about what ANYONE
with a DAW can see for themselves!


> - not inherently part of the hyper-compression process. And the same 'loudness' could be achieved
> without any clipping.

No, it's already hyper-compressed before the gain is adjusted into
clipping for that little bit more.


> That it may be a common error, or a deliberate misuse, is a different
> story.

No, that it is deliberate *IS* what I was saying.


> You can also clip to hell, and *not* have hyper-compression or loudness.
> Lots of distortion though. But you know that.

Right, I certainly know what CAN be done, and unfortunately the common
practice these days is to do it BOTH ways to gain even more "loudness".
ie. first compress the hell out of it, *AND* let it clip as much as they
think they can get away with.

Trevor.

[geoff]

On 12/10/2015 9:48 p.m., Trevor wrote:

>
> Right, I certainly know what CAN be done, and unfortunately the common
> practice these days is to do it BOTH ways to gain even more "loudness".
> ie. first compress the hell out of it, *AND* let it clip as much as they
> think they can get away with.

I suggest that the clipping does more with harshness than loudness, and
adds comparatively little in comparison to compression and limiting.

..... which makes it even more pointless and vile.

geoff

[[email protected]]

On Monday, October 12, 2015 at 1:27:32 AM UTC-4, geoff wrote:
> On 12/10/2015 4:03 p.m., JackA wrote:
> > On Friday, October 9, 2015 at 9:47:34 PM UTC-4, wrote:
> >> Whoever decreed that ALL audio tracks MUST peak within
> >> 1/1,000,00dB of full scale must be on something strong!!
> >
> > I was told, no clipping! Otherwise, Loudness Wars would have never existed.
> >
> > Jack
> >
>
>
> Loudness and loudness wars have nothing to do with clipping.
>
> geoff
__________

Actually, I was criticizing the need to have
all tracks on a CD peak within 1/10 of full scale,
for WHAT EVER reason.

[JackA]

On Monday, October 12, 2015 at 1:27:32 AM UTC-4, geoff wrote:
> On 12/10/2015 4:03 p.m., JackA wrote:
> > On Friday, October 9, 2015 at 9:47:34 PM UTC-4, wrote:
> >> Whoever decreed that ALL audio tracks MUST peak within
> >> 1/1,000,00dB of full scale must be on something strong!!
> >
> > I was told, no clipping! Otherwise, Loudness Wars would have never existed.
> >
> > Jack
> >
>
>
> Loudness and loudness wars have nothing to do with clipping.

Neither does 3kHz!!!

Jack
>
> geoff

[JackA]

On Friday, October 9, 2015 at 9:40:06 PM UTC-4, wrote:
> gray_wolf wrote: "kman, eat your heart out"
>
>
> I never said ALL engineers were guilty of bad practices, and
> besides, I came to realize two years ago where the "hot CD"
> edict came from.
>
> Semi-related, look at the levels on the DR Database entry
> for the 1987 CD release of Billy Joel's "The Stranger"
>
> http://dr.loudness-war.info/album/view/88060
>
> Only a few of the tracks peak to within 1dB of full scale,
> with one track peaking 8dB below it(!)
>
> I OWN this edition, and it sounds just great on my
> players.

Like dull sound, meet Keith!...

http://www.keithhirsch.com/target-cds

Jack :)

[[email protected]]

> > >
> > > I was told, no clipping! Otherwise, Loudness Wars would have never existed.
> > >
> > > Jack
> > >
> >
> >
> > Loudness and loudness wars have nothing to do with clipping.
> >

just like in a cascade of analog stages, there are various forms of clipping.

In analog, you can clip the output stage or you can clip in the pre-amp.

On a CD, you can clip by having several consecutive samples on the CD reach full scale or you can clip the signal at an earlier stage and have that clipped signal accurately recorded on the CD.

In other words, you don't have to be at full scale to have clipping.

It's still clipping either way.

Mark

[Scott Dorsey]

geoff > wrote:
>
>No CD that I've purchased in the last decade exhibit clipping that I've
>noticed, though a few are over-compressed to hell (not many
>fortunately). And some I've been prompted to actually check ! Not much
>'current' pop though I concede.

The problem is that in the digital world, clipping is whatever you define
it as. I tend to set metering so three consecutive FS samples light the
over light, and so that is clipping.

Aggressive limiting that flat-tops the signal isn't necessarily clipping,
it's just aggressive limiting. But at what point does limiting turning
into clipping? The point at which three consecutive FS samples appear.

Much of the k man's confusion has to do with the fact that he can't get
the difference between reference levels and loudness.. and once you start
adding limiting, it doesn't matter _what_ your reference level is because
you can go infinitely high over it and still not light that red light.

Now... the truth is that I have seen some pop CDs that have as many as
eight consecutive FS samples... and I would call that clipping. But,
someone else who decides to calibrate their over light differently
might not, and that is the problem when you start using the word 'clipping'
in the digital world.

>Loudness is achieved by extreme compression and/or limiting. If digital
>clipping occurs, that is a *technical error* - not inherently part of
>the hyper-compression process. And the same 'loudness' could be achieved
>without any clipping.

The question is where limiting ends and clipping begins, and where that
exact line is actually is a philosophical question and not a technical one.
--scott

--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Scott Dorsey]

> wrote:
>
>Actually, I was criticizing the need to have
>all tracks on a CD peak within 1/10 of full scale,
>for WHAT EVER reason.

Why? There's nothing wrong with that. If the D/A converter is properly
designed, it's going to reproduce what went into it exactly.

Peak levels have nothing to do with actual loudness. I can have a very
quiet CD with one triangle hit that goes up to 0dBFS without even
sounding very loud because it's so brief.

So you _want_ your peak levels to be at 0dBFS. The question is where
you want your _average_ levels to be, because that's where loudness comes
from.

Now, there IS a reason to leave plenty of headroom when you're going to
be doing processing afterward, or on an original recording when you can
never be sure something loud and unexpected won't happen. But that has
absolutely no application to a final release CD.

I know we have all explained this to you many many times and you don't get
it, but I am going to try again.
--scott

--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[[email protected]]

On Monday, October 12, 2015 at 9:26:03 AM UTC-4, Scott Dorsey wrote:

>
> Much of the k man's confusion has to do with the fact that he can't get
> the difference between reference levels and loudness.. and once you start
> adding limiting, it doesn't matter _what_ your reference level is because
> you can go infinitely high over it and still not light that red light.

> --scott
>
> --
> "C'est un Nagra. C'est suisse, et tres, tres precis."


Yes I can, if explained properly. By YOU Scott - and
"N___" - keep it shut!! Let Dorsey have the floor.

[[email protected]]

On Monday, October 12, 2015 at 9:31:32 AM UTC-4, Scott Dorsey wrote:
>> wrote:
> >
> >Actually, I was criticizing the need to have
> >all tracks on a CD peak within 1/10 of full scale,
> >for WHAT EVER reason.
>
> Why? There's nothing wrong with that. If the D/A converter is properly
> designed, it's going to reproduce what went into it exactly.
>
> Peak levels have nothing to do with actual loudness. I can have a very
> quiet CD with one triangle hit that goes up to 0dBFS without even
> sounding very loud because it's so brief.
>
> So you _want_ your peak levels to be at 0dBFS. The question is where
> you want your _average_ levels to be, because that's where loudness comes
> from.
>
> Now, there IS a reason to leave plenty of headroom when you're going to
> be doing processing afterward, or on an original recording when you can
> never be sure something loud and unexpected won't happen. But that has
> absolutely no application to a final release CD.
>
> I know we have all explained this to you many many times and you don't get
> it, but I am going to try again.
> --scott
>
> --
> "C'est un Nagra. C'est suisse, et tres, tres precis."


Thanks for that explanation Scott.


My issue with peak normalization(which is what
you describe above is called), is that a song
with a lower peak-to-average ratio, peak normalized
alongside a song with a higher peak-to-average,
WILL SOUND LOUDER than the one with the higher
peak to average.


OTOH: If I use my ears to loudness-normalize 3-4
songs of different genres or production eras
(stuff from the 1970s, 1990s, and last month),
the final result is that they will all sound
equally loud to my ears, but some of them
my not peak at 0dB full scale - when I look
at the meters during playback. And in my
mind there's NOTHING WRONG with that; as long
as they all sound about as loud as each other
and sound GOOD.

[Trevor]

On 13/10/2015 12:25 AM, Scott Dorsey wrote:
> geoff > wrote:
>> No CD that I've purchased in the last decade exhibit clipping that I've
>> noticed, though a few are over-compressed to hell (not many
>> fortunately). And some I've been prompted to actually check ! Not much
>> 'current' pop though I concede.
>
> The problem is that in the digital world, clipping is whatever you define
> it as. I tend to set metering so three consecutive FS samples light the
> over light, and so that is clipping.

No, you first have to normalise the gain back to 0dBFS for that to work
since many CD's are first *severely clipped* then normalised to about
-0.3dBFS, so your clip lights will never come on. BUT the flat tops
remain regardless!

>
> Aggressive limiting that flat-tops the signal isn't necessarily clipping,
> it's just aggressive limiting.

Rubbish, peak limiting that causes flat tops IS clipping. YOU simply
don't understand what they have done, or the difference between
compression and limiting it seems.


> But at what point does limiting turning
> into clipping? The point at which three consecutive FS samples appear.

I'd agree with that, BUT it doesn't have to be FS, simply whatever point
they have renormalised to after clipping. That is the bit you seem not
to grasp.



> Much of the k man's confusion has to do with the fact that he can't get
> the difference between reference levels and loudness..

It seems you don't get the difference between 0dBFS and the chosen
maximum normalised level yourself.


> and once you start
> adding limiting, it doesn't matter _what_ your reference level is because
> you can go infinitely high over it and still not light that red light.

Or if you create square waves from sine waves and renormalise to
anything slightly less than 0dBFS. No red light, and NO "limiter" is
required to do that!


> Now... the truth is that I have seen some pop CDs that have as many as
> eight consecutive FS samples...

Hell, you haven't looked much, I have seen HUNDREDS of consecutive
samples at maximum NORMALISED level, ie FLAT TOPS!


> and I would call that clipping.

Me too.

> But,
> someone else who decides to calibrate their over light differently
> might not, and that is the problem when you start using the word 'clipping'
> in the digital world.

NOPE, clipping is clipping, whatever you choose as YOUR final peak level.


>> Loudness is achieved by extreme compression and/or limiting. If digital
>> clipping occurs, that is a *technical error* - not inherently part of
>> the hyper-compression process. And the same 'loudness' could be achieved
>> without any clipping.
>
> The question is where limiting ends and clipping begins, and where that
> exact line is actually is a philosophical question and not a technical one.

NOPE, clipping is the same regardless of the chosen peak level. When a
sine wave looks like a square wave, (ie completely flat tops) it is
clipped, even if the max level is normalised to -60dBFS !!!!!!!!!!!
I can EASILY do that, as can you if you want, and all the BS in the
world about whether it is clipped or peak limited does NOT change the sound.

Trevor.

[Mike Rivers[_2_]]

On 10/12/2015 7:29 AM, wrote:
> Actually, I was criticizing the need to have
> all tracks on a CD peak within 1/10 of full scale,
> for WHAT EVER reason.

For certain types of music played in certain situations, it's helpful of
every track on the CD is approximately the same _loudness_. If it's all
the same kind of music, and the kind of music that's usually fairly
heavily compressed anyway, then by making all the tracks peak at 0 dBFS
is an easy way to achieve this. You could just as easily make them all
peak at -6 dBFS, or -20 dBFS, but only wimps do that. ;)

The idea is that you don't have to adjust the playback volume when
switching from track to track. If you're driving in your car, working
out in the gym, or if you're DJ-ing in a dance club, it's one more
control you don't have to fiddle with.

If you're sitting in your easy chair in the living room, it's nice to
give your ears a little break now and then.

The real solution, and it's sort of there but nobody implements it, is
to encode some information about the loudness of the track and have the
playback system adjust the level using that information and your
listening preferences.


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

[Randy Yates[_2_]]

Mike Rivers > writes:
> [...]
> The real solution, and it's sort of there but nobody implements it, is
> to encode some information about the loudness of the track and have
> the playback system adjust the level using that information and your
> listening preferences.

Didn't (or doesn't) Dolby have a system / standard that does just that
for the movie/pro audio market?
--
Randy Yates
Digital Signal Labs
http://www.digitalsignallabs.com

[Scott Dorsey]

> wrote:
>
>My issue with peak normalization(which is what
>you describe above is called), is that a song
>with a lower peak-to-average ratio, peak normalized
>alongside a song with a higher peak-to-average,
>WILL SOUND LOUDER than the one with the higher
>peak to average.

Maybe. Maybe not. It's a case of apples and oranges. The question
is what the average level is.

>OTOH: If I use my ears to loudness-normalize 3-4
>songs of different genres or production eras
>(stuff from the 1970s, 1990s, and last month),
>the final result is that they will all sound
>equally loud to my ears, but some of them
>my not peak at 0dB full scale - when I look
>at the meters during playback. And in my
>mind there's NOTHING WRONG with that; as long
>as they all sound about as loud as each other
>and sound GOOD.

This because the peak level has little to do with perceived loudness.
--scott

--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Scott Dorsey]

Randy Yates > wrote:
>Mike Rivers > writes:
>> [...]
>> The real solution, and it's sort of there but nobody implements it, is
>> to encode some information about the loudness of the track and have
>> the playback system adjust the level using that information and your
>> listening preferences.
>
>Didn't (or doesn't) Dolby have a system / standard that does just that
>for the movie/pro audio market?

Yes, that would be the DIALNORM field in the Dolby AC-3 bitstream.

The MP3 files have a similar thing, but of course it's heavily abused
and invariably set to maximum.
--scott

--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Randy Yates[_2_]]

(Scott Dorsey) writes:

> Randy Yates > wrote:
>>Mike Rivers > writes:
>>> [...]
>>> The real solution, and it's sort of there but nobody implements it, is
>>> to encode some information about the loudness of the track and have
>>> the playback system adjust the level using that information and your
>>> listening preferences.
>>
>>Didn't (or doesn't) Dolby have a system / standard that does just that
>>for the movie/pro audio market?
>
> Yes, that would be the DIALNORM field in the Dolby AC-3 bitstream.
>
> The MP3 files have a similar thing, but of course it's heavily abused
> and invariably set to maximum.

No, it was something much more extensive than a field in AC-3. I think
it was/is "Dolby Volume," but I can't even remember where I found out
about this:

http://www.dolby.com/us/en/technologies/dolby-volume.html

--
Randy Yates
Digital Signal Labs
http://www.digitalsignallabs.com

[Scott Dorsey]

In article >, Trevor > wrote:
>Rubbish, peak limiting that causes flat tops IS clipping. YOU simply
>don't understand what they have done, or the difference between
>compression and limiting it seems.

How much of a flat top is needed for it to be clipping? One sample?
Many samples? At what point does limiting end and clipping begin?

On the other hand, let's say I have a drum hit that is only three samples
long and it's the only loud thing on the track. I run it through a
limiter... maybe it's just a pair of back to back diodes, and I chop it
down 20 dB. It sounds fine. Did I clip it, or did I limit it?
--scott

--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[geoff]

On 13/10/2015 2:25 a.m., Scott Dorsey wrote:
> geoff > wrote:
>> No CD that I've purchased in the last decade exhibit clipping that I've
>> noticed, though a few are over-compressed to hell (not many
>> fortunately). And some I've been prompted to actually check ! Not much
>> 'current' pop though I concede.
> The problem is that in the digital world, clipping is whatever you define
> it as. I tend to set metering so three consecutive FS samples light the
> over light, and so that is clipping.


I draw the line at 2, though 3 seems to be the commonly accepted criteria.

geoff

[geoff]

On 13/10/2015 3:23 a.m., Trevor wrote:
> On 13/10/2015 12:25 AM, Scott Dorsey wrote:
>> geoff > wrote:
>>> No CD that I've purchased in the last decade exhibit clipping that I've
>>> noticed, though a few are over-compressed to hell (not many
>>> fortunately). And some I've been prompted to actually check ! Not much
>>> 'current' pop though I concede.
>>
>> The problem is that in the digital world, clipping is whatever you
>> define
>> it as. I tend to set metering so three consecutive FS samples light the
>> over light, and so that is clipping.
>
> No, you first have to normalise the gain back to 0dBFS for that to
> work since many CD's are first *severely clipped* then normalised to
> about -0.3dBFS, so your clip lights will never come on. BUT the flat
> tops remain regardless!
>
>>
>> Aggressive limiting that flat-tops the signal isn't necessarily
>> clipping,
> > it's just aggressive limiting.
>
> Rubbish, peak limiting that causes flat tops IS clipping. YOU simply
> don't understand what they have done, or the difference between
> compression and limiting it seems.

I think this is the only thing where we really disagree. The rest of
your debate seems hair-splitting about definitions of terms.

I say that 'clipping' is only the result of a digital or analogue
*overload*, where the actual mathematical or electrical constraints of
the process are exceeded and nothing can exist above. Fuzz box !

'Limiting' is the *controlled* result of a digital or analogue process.
And yes, that can be extreme to the point of resembling clipping, but is
not the same thing.

Achieving, or adding to, limiting by clipping (sadly) is done,
presumably by those who are incompetent, lack understanding of the
implications, or have cynical intent.

geoff

[geoff]

On 13/10/2015 3:52 a.m., Mike Rivers wrote:
>
>
> The idea is that you don't have to adjust the playback volume when
> switching from track to track. If you're driving in your car, working
> out in the gym, or if you're DJ-ing in a dance club, it's one more
> control you don't have to fiddle with.
>
> If you're sitting in your easy chair in the living room, it's nice to
> give your ears a little break now and then.
>
>
>

The automotive scenario being the 'driving factor' behind the crushing
that radio stations apply to the music. Kind of kills it for other
situations though.

geoff

[geoff]

On 13/10/2015 2:05 a.m., JackA wrote:
> http://www.keithhirsch.com/target-cds Jack :)

A truly bizarre website.

geoff

[[email protected]]

a good way to differentiate clipping from limiting from compression is by
considering the attack and decay times relative to the features of the waveform

clipping is instantaneous gain changes
as soon as the waveform exceeds a threshold, the gain reduces
as soon as the waveform is below threshold, the gain increases
the attack and decay time are ZERO
back to back diodes do this
you can various ratios.. ie hard or soft clipping

limiting is a slower process
there is a finite attack and decay time that is longer than the time of a waveform cycle but is still very fast compared to the envelope usuall 1 to a few msec
again you can have various ratios


compression is even slower attack and decay time whose duration is
large compared to individual waveform cycles and can be several sylabbles or several seconds even

AGC is even slower typically operates over many seconds

those are the definitions I use, anyway...
Mark

[Mike Rivers[_2_]]

On 10/12/2015 11:22 AM, Randy Yates wrote:
>> The real solution, and it's sort of there but nobody implements it, is
>> >to encode some information about the loudness of the track and have
>> >the playback system adjust the level using that information and your
>> >listening preferences.
> Didn't (or doesn't) Dolby have a system / standard that does just that
> for the movie/pro audio market?

I was thinking about Dialnorm, but it's a movie thing that nobody
calibrates the decoder correctly. I'm not aware of any home/consumer
system that implements it. I think that there may be some reluctance on
the part of consumers to have their playback system told to turn the
volume up on this song and turn it down on that one.

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

[geoff]

On 13/10/2015 10:48 a.m., Mike Rivers wrote:
> On 10/12/2015 11:22 AM, Randy Yates wrote:
>>> The real solution, and it's sort of there but nobody implements it, is
>>> >to encode some information about the loudness of the track and have
>>> >the playback system adjust the level using that information and your
>>> >listening preferences.
>> Didn't (or doesn't) Dolby have a system / standard that does just that
>> for the movie/pro audio market?
>
> I was thinking about Dialnorm, but it's a movie thing that nobody
> calibrates the decoder correctly. I'm not aware of any home/consumer
> system that implements it. I think that there may be some reluctance
> on the part of consumers to have their playback system told to turn
> the volume up on this song and turn it down on that one.
>

My car iPod Touch has a automatic replay level function based on (I
think) a peak-level parameter encoded into the file.

When it works on a particularly 'different level' track there seems to
be a second-long level adjustment at the commencement of a song, which
is a pain. But not as much of a pain as having to reach for the knob
(car stereo) nearly *every* song while driving when playing a random
mix of material.

geoff

[[email protected]]

geoff:

Where is this leveling feature in the Touch's settings?

[geoff]

On 13/10/2015 1:52 p.m., wrote:
> geoff:
>
> Where is this leveling feature in the Touch's settings?


Called 'Sound Check' . Just a simple normalisation thing, I think.

iTunes end : https://support.apple.com/en-us/HT201724

iPod end :
http://ipod.about.com/od/iPhoneMusicApp/ss/Iphone-Ipod-App-Settings-Soundcheck-Eq-And-Volume-Limit.htm

geoff

[[email protected]]

geoff wrote: "Called 'Sound Check' . Just a simple normalisation thing, I think. "


Ahh yes - Now you're talking! Although I seriously doubt
it peak normalizes. I could expreimient with it by applying
it in iTunes to a couple of test tracks: one very dynamic(high
peak to avg ratio), the other, highly compressed(low peak
to avg). If they both sound as loud as each other, then it is not
peak-normalizing. The average is closer to what we hear, and
if you align the tracks by peak, the average of the less dynamic
track should sound louder.

[Trevor]

On 13/10/2015 5:08 AM, Scott Dorsey wrote:
> In article >, Trevor > wrote:
>> Rubbish, peak limiting that causes flat tops IS clipping. YOU simply
>> don't understand what they have done, or the difference between
>> compression and limiting it seems.
>
> How much of a flat top is needed for it to be clipping? One sample?
> Many samples?

At least 3 or more. (genuine square waves excluded) And as I have said,
some CD's have HUNDREDS of consecutive samples at maximum level, and
often many similar groups in one song. IF there are only 3 in fact I'm
not worried in the slightest, when there are hundreds I am. That is far
more common with pop CD's these days than many people seem to think,
simply because they never look.


> At what point does limiting end and clipping begin?

As soon as the knee ends and maximum level is reached for at least 3 or
more samples. I'm surprised you don't understand this already It's not
rocket science!


> On the other hand, let's say I have a drum hit that is only three samples
> long and it's the only loud thing on the track. I run it through a
> limiter... maybe it's just a pair of back to back diodes, and I chop it
> down 20 dB. It sounds fine. Did I clip it, or did I limit it?

IF the drum hit is now flat topped for those 3 samples you did BOTH,
regardless of what level you "chop it down to". (of course there is NO
such thing as a drum hit for 3/44,000's of a second!)

I'm totally amazed that someone with your experience still doesn't have
a vague notion of what clipping actually is. It can occur anywhere in
the chain, and is independent of final level. Just like you can clip a
Mackie mixers mix bus, even if the output fader is well below maximum.
In fact you can clip a single channel on ANY mixer, without clipping the
mixer output, or amplifier input. THAT channel is STILL clipped!!!

Trevor.

[Trevor]

On 13/10/2015 7:40 AM, geoff wrote:
> On 13/10/2015 2:25 a.m., Scott Dorsey wrote:
>> geoff > wrote:
>>> No CD that I've purchased in the last decade exhibit clipping that I've
>>> noticed, though a few are over-compressed to hell (not many
>>> fortunately). And some I've been prompted to actually check ! Not much
>>> 'current' pop though I concede.
>> The problem is that in the digital world, clipping is whatever you define
>> it as. I tend to set metering so three consecutive FS samples light the
>> over light, and so that is clipping.
>
>
> I draw the line at 2, though 3 seems to be the commonly accepted criteria.

Yep, and yet MANY Cd's have HUNDREDS of consecutive samples at maximum
level :-(
No-one can argue THAT is *NOT* clipping simply because the CD is
normalised to -0.3dBFS, CAN THEY? Or are some people who claim to be
"pro's" really that stupid?

Trevor.

[[email protected]]

Trevor, Scott:

And whether its clipping as in overloading a bus or
as in clipping aesthetically(a waveform), both
can sound like ****e!

[Trevor]

On 13/10/2015 7:50 AM, geoff wrote:
> On 13/10/2015 3:23 a.m., Trevor wrote:
>> On 13/10/2015 12:25 AM, Scott Dorsey wrote:
>>> geoff > wrote:
>>>> No CD that I've purchased in the last decade exhibit clipping that I've
>>>> noticed, though a few are over-compressed to hell (not many
>>>> fortunately). And some I've been prompted to actually check ! Not much
>>>> 'current' pop though I concede.
>>>
>>> The problem is that in the digital world, clipping is whatever you
>>> define
>>> it as. I tend to set metering so three consecutive FS samples light the
>>> over light, and so that is clipping.
>>
>> No, you first have to normalise the gain back to 0dBFS for that to
>> work since many CD's are first *severely clipped* then normalised to
>> about -0.3dBFS, so your clip lights will never come on. BUT the flat
>> tops remain regardless!
>>
>>>
>>> Aggressive limiting that flat-tops the signal isn't necessarily
>>> clipping,
>> > it's just aggressive limiting.
>>
>> Rubbish, peak limiting that causes flat tops IS clipping. YOU simply
>> don't understand what they have done, or the difference between
>> compression and limiting it seems.
>
> I think this is the only thing where we really disagree. The rest of
> your debate seems hair-splitting about definitions of terms.

Rubbish, When someone claims clipping is not clipping simply because
it's level has been post reduced, or that clipping is not clipping
simply because it was done by a "limiter" that is FARRRRR from hair
splitting about definitions, that is about that person not understanding
the basic fundamentals of recording!!!!!



>
> I say that 'clipping' is only the result of a digital or analogue
> *overload*, where the actual mathematical or electrical constraints of
> the process are exceeded and nothing can exist above.

Right, as occurs when a limiter exceeds it's absolute peak level, a
mixer or amplifier exceeds it's maximum rail voltage, or a digital
device exceeds it's digital full scale level, whether OR NOT, that level
is subsequently reduced AFTER clipping!!

>
> 'Limiting' is the *controlled* result of a digital or analogue process.
> And yes, that can be extreme to the point of resembling clipping, but is
> not the same thing.

Bull****!! A limiter by it's very nature will produce CLIPPING when it
exceeds it's knee and reaches it's maximum level. That's the whole point
of the device after all. One may choose to use it so it never exceeds
the knee (and hard limiting may not even have a knee!) but IF it does
exceed absolute maximum, then CLIPPING will occur! That you choose to
believe it is somehow not clipping just because YOU prefer to call it
limiting only shows you ignorance!!!

>
> Achieving, or adding to, limiting by clipping (sadly) is done,
> presumably by those who are incompetent, lack understanding of the
> implications, or have cynical intent.

Yep, but limiting is NOT necessarily the same as compression. You and
many others still seem to be totally confused about that. And of course
one can drive a compressor into clipping as well!
Clipping is clipping no matter how it is achieved. Arguing that it is
not is simply TOTAL BS!

Trevor.

[Mike Rivers[_2_]]

On 10/12/2015 6:24 PM, geoff wrote:
> My car iPod Touch has a automatic replay level function based on (I
> think) a peak-level parameter encoded into the file.
>
> When it works on a particularly 'different level' track there seems to
> be a second-long level adjustment at the commencement of a song

I would like to have something like that in my car. It might not work as
well for me as it does for you, however, because for me, a "song" (one
file) is a two hour radio program of music that often has song-to-song
level variations.

I've been trying to score a cheap couple of generations old iPod Touch
so I could use it for some audio apps that aren't available for my
Androids, but so far no luck. They seem to hold their value nearly as
well as microphones. ;)

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

[Mike Rivers[_2_]]

On 10/12/2015 10:32 PM, geoff wrote:
> Called 'Sound Check' . Just a simple normalisation thing, I think.

My TASCAM DR-40 has something like that. I leave it turned off. In the
DR-44, it's on all the time. TASCAM seems to be sensitized to rules
about hearing protection and this is one of the things that they do (in
addition to popping a reminder up on the screen when you turn the
playback volume up) to comply with the CE mark testing.

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

[Mike Rivers[_2_]]

On 10/13/2015 2:54 AM, wrote:
> geoff wrote: "Called 'Sound Check' . Just a simple normalisation thing, I think. "

> Ahh yes - Now you're talking! Although I seriously doubt
> it peak normalizes. I could expreimient with it by applying
> it in iTunes to a couple of test tracks: one very dynamic(high
> peak to avg ratio), the other, highly compressed(low peak
> to avg).

Why don't you test it by generating some sine wave test signals, both
continuous and with burst level changes, and learn what it's really
doing? That way you'd know (and can explain to us) why something sounds
louder or not.


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

[[email protected]]

On Tuesday, October 13, 2015 at 9:14:46 AM UTC-4, Mike Rivers wrote:
> On 10/13/2015 2:54 AM, wrote:
> > geoff wrote: "Called 'Sound Check' . Just a simple normalisation thing, I think. "
>
> > Ahh yes - Now you're talking! Although I seriously doubt
> > it peak normalizes. I could expreimient with it by applying
> > it in iTunes to a couple of test tracks: one very dynamic(high
> > peak to avg ratio), the other, highly compressed(low peak
> > to avg).
>
>
I think there are several setting in iTUNEs (or other programs) that you can use to change the way the normalization is performed. The key point however, is that none of the setting change the dynamics within a song. The ENTIRE song is adjusted by the same constant value. It is no different from setting the volumne control during playback. This does not really change the shape of the envelope or change the dynamics.

Mark

[[email protected]]

Mike Rivers wrote: "On 10/13/2015 2:54 AM, wrote:
> geoff wrote: "Called 'Sound Check' . Just a simple normalisation thing, I think. "


Why don't you test it by generating some sine wave test signals, both
continuous and with burst level changes, and learn what it's really
doing? That way you'd know (and can explain to us) why something sounds
louder or not.
- show quoted text -"


I already KNOW why there's a loudness difference between a
more dynamic song and a less dynamic song when both are peak-
normalized, I.E. to 0dBfs, given the same playback volume setting.


But terminology gets in the way of our understanding each other,
Mike. I understand CONCEPTS Mike - but fail like our previous
president at verbalizing them.


And I seriously DOUBT iTunes' Sound Check is peak-
based, or else it wouldn't serve any purpose beyond what has
already been done on thousands of CD titles. I know MP3Gain
is not peak-based, because I have NEVER had to adjust playback
volume ONCE during playing hundreds of mp3 versions of my
collection with mp3Gain applied - except when the phone rings,
or the wife comes home(!)


In fact, one of the biggest complaints I've read about Apple
Sound Check is about how much stuff it turns DOWN. Users
thus must turn their volumes up higher when listening to a
playlist that might span several(pre- and loudness war era)
decades. MP3Gain does too, depending upon where one
sets its threshold. It uses a scale: 87-105dB, with a
recommended setting of 89dB.


I normalize at 91, which triggers the clipping warning on
some songs, mainly stuff before 1980, but is relatively
inaudible to the ear. This level helps the small amps in
my mobile devices where the MP3s are played. Of course,
for nearly EVERYTHING released since 2000, it applies at
least 6dB of negative gain.


If I rip two mp3s of the same song off a hot modern
compressed CD, mp3 gain ONE of them, and place them
both in my DAW, it does show the original kissing full scale,
and the re-gained as pathetically small, with between 6-10dB
of headroom between the flat-tops and full scale.


So I DO understand what's going on Mike, Scott, geoff, and
that ASSHOLE thread-crasher who shall remain nameless,
it's putting things into words where I stumble.

[[email protected]]

wrote: "I think there are several setting in iTUNEs (or other programs) that you can use to change
the way the normalization is performed. The key point however, is that none of the setting change the dynamics within a
song. The ENTIRE song is adjusted by the same constant value. It is no different from setting the volumne control during
playback. This does not really change the shape of the envelope or change the dynamics.

-Mark"

^^ CORRECT! ^^

[Scott Dorsey]

gray_wolf > wrote:
>
>0dBFS+ Levels in Digital Mastering
>
>http://service.tcgroup.tc/media/Level_paper_AES109(1).pdf
>
>Interesting. I have noticed that in my better recorded stuff
>that the maximum levels tend to be at least -3dB and some as low as -6dB.

This paper is fifteen years old, and back then this was a new problem that
people hadn't really seen before.

This is a problem with converter design, and it's most often a problem with
trying to run a ladder dac with an I/V stage off the same rails as the ladder.

In the modern sigma-delta world the problem is different, but it still is
something converter designers need to be aware of. This, though, is not a
problem with the sound processing, it's a problem with the converter design.

Benchmark has a good white paper on how these problems were solved, I believe.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Scott Dorsey]

geoff > wrote:
>On 13/10/2015 2:05 a.m., JackA wrote:
>> http://www.keithhirsch.com/target-cds Jack :)
>
>A truly bizarre website.

People collect strange things. I suppose this is no worse than the people
who go ga-ga over Columbia six-eyes LP pressings.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Scott Dorsey]

> wrote:
>
>a good way to differentiate clipping from limiting from compression is by
>considering the attack and decay times relative to the features of the waveform

That's half of it... and the other half is the shape of the knee. But how
fast is fast? When does soft clipping become limiting and vice-versa?
--scott


--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Scott Dorsey]

In article >, Trevor > wrote:
>> Many samples?
>
>At least 3 or more. (genuine square waves excluded) And as I have said,
>some CD's have HUNDREDS of consecutive samples at maximum level, and
>often many similar groups in one song. IF there are only 3 in fact I'm
>not worried in the slightest, when there are hundreds I am. That is far
>more common with pop CD's these days than many people seem to think,
>simply because they never look.

Well, that's the degenerate case. Those CDs are clearly clipped. But I'm
talking about the borderline cases, because that's where it gets interesting.

>> At what point does limiting end and clipping begin?
>
>As soon as the knee ends and maximum level is reached for at least 3 or
>more samples. I'm surprised you don't understand this already It's not
>rocket science!

That's a good definition. Although... I might decide it is 2 samples or
8 samples and be able to make a good argument for those too.

>I'm totally amazed that someone with your experience still doesn't have
>a vague notion of what clipping actually is. It can occur anywhere in
>the chain, and is independent of final level. Just like you can clip a
>Mackie mixers mix bus, even if the output fader is well below maximum.
>In fact you can clip a single channel on ANY mixer, without clipping the
>mixer output, or amplifier input. THAT channel is STILL clipped!!!

Oh, I have a vague notion of what clipping actually is.... and a vague notion
of what limiting actually is.... but I can think of a LOT of examples that
are sitting directly on the border. I don't want a vague notion, I want a
precise mathematical description.

My line might be "if you can hear it, it's clipping, if you can't, it's
limiting." That's no less vague, though.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Randy Yates[_2_]]

(Scott Dorsey) writes:

> gray_wolf > wrote:
>>
>>0dBFS+ Levels in Digital Mastering
>>
>>http://service.tcgroup.tc/media/Level_paper_AES109(1).pdf
>>
>>Interesting. I have noticed that in my better recorded stuff
>>that the maximum levels tend to be at least -3dB and some as low as -6dB.
>
> This paper is fifteen years old, and back then this was a new problem that
> people hadn't really seen before.
>
> This is a problem with converter design, and it's most often a problem with
> trying to run a ladder dac with an I/V stage off the same rails as the ladder.
>
> In the modern sigma-delta world the problem is different, but it still is
> something converter designers need to be aware of. This, though, is not a
> problem with the sound processing, it's a problem with the converter
> design.

You keep saying this. Exactly what is the problem? Are you referring to
the 5 items on p.4 of that reference?

I have a couple of remarks regarding those items (if anyone wants to
hear them...):

1. If the digital filter is distorting/clipping, it's implemented wrong.
Any diligent DSP designer would look out for that sort of thing.

2. I think most of the rest of these problems arise from the basic
sampling theory explanation given on p.2. Several years ago I proved
that, with the theoretical brick-wall lowpass interpolation filter, it
is possible that a limited-range digital signal can give rise to an
infinite analog output. This was an analytical result, but it can have
practical implications, namely, that, given there are no infinite-length
filters in the real-world, the DAC designer should be able to specify
the absolute maximum analog output level the DAC will produce,
--
Randy Yates
Digital Signal Labs
http://www.digitalsignallabs.com

[[email protected]]

Scott Dorsey wrote: "My line might be "if you can hear it, it's clipping, if you can't, it's
limiting." That's no less vague, though. "



And that brings up a valid point: Clipping, as I've understood
it for the last 20 years, is when the electronics of a device
become momentarily overloaded. And how audible it sounds
is dictated largely by duration.

Limiting is more an aesthetic property, but again, its duration
determines its audibility.

A SIDE NOTE:
Via google groups, this thread, for about 5 minutes, around
11AM Eastern time, "disappeared" from the R.A.P. header
list. It is now back, as I am replying to it.

[Randy Yates[_2_]]

Randy Yates > writes:

> (Scott Dorsey) writes:
>
>> gray_wolf > wrote:
>>>
>>>0dBFS+ Levels in Digital Mastering
>>>
>>>http://service.tcgroup.tc/media/Level_paper_AES109(1).pdf
>>>
>>>Interesting. I have noticed that in my better recorded stuff
>>>that the maximum levels tend to be at least -3dB and some as low as -6dB.
>>
>> This paper is fifteen years old, and back then this was a new problem that
>> people hadn't really seen before.
>>
>> This is a problem with converter design, and it's most often a problem with
>> trying to run a ladder dac with an I/V stage off the same rails as the ladder.
>>
>> In the modern sigma-delta world the problem is different, but it still is
>> something converter designers need to be aware of. This, though, is not a
>> problem with the sound processing, it's a problem with the converter
>> design.
>
> You keep saying this. Exactly what is the problem? Are you referring to
> the 5 items on p.4 of that reference?
>
> I have a couple of remarks regarding those items (if anyone wants to
> hear them...):
>
> 1. If the digital filter is distorting/clipping, it's implemented wrong.
> Any diligent DSP designer would look out for that sort of thing.
>
> 2. I think most of the rest of these problems arise from the basic
> sampling theory explanation given on p.2. Several years ago I proved
> that, with the theoretical brick-wall lowpass interpolation filter, it
> is possible that a limited-range digital signal can give rise to an
> infinite analog output. This was an analytical result, but it can have
> practical implications, namely, that, given there are no infinite-length
> filters in the real-world, the DAC designer should be able to specify
> the absolute maximum analog output level the DAC will produce,

OK, a little crow I'll eat...

With a delta-sigma DAC, there is non-linearity (requantization)
involved. This can make it impossible or much more difficult to analyze
the maximum output level.
--
Randy Yates
Digital Signal Labs
http://www.digitalsignallabs.com

[Mike Rivers[_2_]]

On 10/13/2015 10:10 AM, wrote:
> But terminology gets in the way of our understanding each other,
> Mike. I understand CONCEPTS Mike - but fail like our previous
> president at verbalizing them.

This is why I suggested testing something that you could quantify rather
than depending on what you think you know and hear. It's much easier to
see what a level control is doing when looking at a steady or periodic
signal than when looking at a music waveform.

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

[[email protected]]

Mike Rivers wrote: "On 10/13/2015 10:10 AM, wrote:
> But terminology gets in the way of our understanding each other,
> Mike. I understand CONCEPTS Mike - but fail like our previous
> president at verbalizing them.

This is why I suggested testing something that you could quantify rather
than depending on what you think you know and hear. It's much easier to
see what a level control is doing when looking at a steady or periodic
signal than when looking at a music waveform.
- show quoted text -"

Or for a more real-life test: LISTEN to a song without level control
and then with it it, to HEAR what is going on. As others have said
in this group, loudness is personal and hard to quantify, I.E. to
express in mathematical or tech terms, such as "avg" "RMS",
and "peak".

[Scott Dorsey]

Randy Yates > wrote:
>
>You keep saying this. Exactly what is the problem? Are you referring to
>the 5 items on p.4 of that reference?

The whole point of the problem described in that paper is that you can have
a waveform whose peak is greater than 0dBFS even though the highest sample
in the waveform is only at 0dBFS... there can be an intersample peak which
is higher than any of the samples.

And... some converters were not designed with that in mind, and they clip
the peak.

>I have a couple of remarks regarding those items (if anyone wants to
>hear them...):
>
>1. If the digital filter is distorting/clipping, it's implemented wrong.
>Any diligent DSP designer would look out for that sort of thing.

Yes.

>2. I think most of the rest of these problems arise from the basic
>sampling theory explanation given on p.2. Several years ago I proved
>that, with the theoretical brick-wall lowpass interpolation filter, it
>is possible that a limited-range digital signal can give rise to an
>infinite analog output. This was an analytical result, but it can have
>practical implications, namely, that, given there are no infinite-length
>filters in the real-world, the DAC designer should be able to specify
>the absolute maximum analog output level the DAC will produce,

Yes.
Fifteen years ago, though, people weren't thinking so much about this.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Scott Dorsey]

Randy Yates > wrote:
>OK, a little crow I'll eat...
>
>With a delta-sigma DAC, there is non-linearity (requantization)
>involved. This can make it impossible or much more difficult to analyze
>the maximum output level.

This is why we have breadboards and scopes and computer simulation, so
we can actually test things before burning silicon!
--scott

--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[JackA]

On Tuesday, October 13, 2015 at 3:24:41 AM UTC-4, Trevor wrote:
> On 13/10/2015 7:40 AM, geoff wrote:
> > On 13/10/2015 2:25 a.m., Scott Dorsey wrote:
> >> geoff > wrote:
> >>> No CD that I've purchased in the last decade exhibit clipping that I've
> >>> noticed, though a few are over-compressed to hell (not many
> >>> fortunately). And some I've been prompted to actually check ! Not much
> >>> 'current' pop though I concede.
> >> The problem is that in the digital world, clipping is whatever you define
> >> it as. I tend to set metering so three consecutive FS samples light the
> >> over light, and so that is clipping.
> >
> >
> > I draw the line at 2, though 3 seems to be the commonly accepted criteria.
>
> Yep, and yet MANY Cd's have HUNDREDS of consecutive samples at maximum
> level :-(
> No-one can argue THAT is *NOT* clipping simply because the CD is
> normalised to -0.3dBFS, CAN THEY? Or are some people who claim to be
> "pro's" really that stupid?
>
> Trevor.

My concern is the use of a graphic equalizer by the listener, if you allow no headroom.

But, let's all consider... When I mixed that Joan Jett song, one stereo channel has a decent size peak in it that limited the rest of the audio. Was I to leave it be or suppress it. It did nothing for the audio.

Jack

[[email protected]]

>
> >> At what point does limiting end and clipping begin?
> >

there __is__ a definitive answer to that question

with clipping , the gain changes fast enough to follow the cycle by cycle waveform ____and therefore creates harmonic and intermodulation distortion___.
If you clip a sine wave, you can hear the harmonics.
Mathematically, this is a non-linar process.

Anything slower than that, is limiting, compression or AGC.
These are mmathematically linear process and do not create harmonics or intermod.
If you limit or compress or AGC a sine wave, you will not hear harmonics.


(I'm not saying these are good, they can still ruin the aesthtics by squashing the dynamic range, but they do not cause harmonics or intermod.

Clipping casues harmonics and intermod.

Matematically, clipping alone creates new frequencies, The others do not.



Mark

[[email protected]]

:


Hmmm.. When I limit(flat top) a pure sine wave, even by
just .5dB, it sounds different to me

[Mike Rivers[_2_]]

On 10/13/2015 2:34 PM, wrote:
> Hmmm.. When I limit(flat top) a pure sine wave, even by
> just .5dB, it sounds different to me

That's good, because when you clip off the tops it has more frequencies
in it than without. But this isn't always a bad thing. Adding harmonics
can make a boring timbre sound more musically interesting. It's done all
the time, sometimes even naturally. But whether it's beneficial or
harmful, it's still distortion.

Try looking at the before and after with a spectrum analyzer and you'll
have a better understanding of _why_ they sound different. If you're
using a DAW that will host VST plug-ins (most can) SPAN from Voxengo is
an excellent and free spectrum analyzer plug-in.



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

[[email protected]]

Mike Rivers wrote: "On 10/13/2015 2:34 PM, wrote:
> Hmmm.. When I limit(flat top) a pure sine wave, even by
> just .5dB, it sounds different to me

That's good, because when you clip off the tops it has more frequencies
in it than without. But this isn't always a bad thing. Adding harmonics
can make a boring timbre sound more musically interesting. It's done all
the time, sometimes even naturally. But whether it's beneficial or
harmful, it's still distortion.

Try looking at the before and after with a spectrum analyzer and you'll
have a better understanding of _why_ they sound different. If you're
using a DAW that will host VST plug-ins (most can) SPAN from Voxengo is
an excellent and free spectrum analyzer plug-in.
- show quoted text -"


One area it's not good is in many reissues labeled as "remastered",
where a lot of the highest peaks are hard-limited, on top of the whole
being compressed. It definitely sounds different, not just louder.

[JackA]

On Tuesday, October 13, 2015 at 2:21:21 PM UTC-4, wrote:
> >
> > >> At what point does limiting end and clipping begin?
> > >
>
> there __is__ a definitive answer to that question
>
> with clipping , the gain changes fast enough to follow the cycle by cycle waveform ____and therefore creates harmonic and intermodulation distortion___.
> If you clip a sine wave, you can hear the harmonics.
> Mathematically, this is a non-linar process.
>
> Anything slower than that, is limiting, compression or AGC.
> These are mmathematically linear process and do not create harmonics or intermod.
> If you limit or compress or AGC a sine wave, you will not hear harmonics.
>
>
> (I'm not saying these are good, they can still ruin the aesthtics by squashing the dynamic range, but they do not cause harmonics or intermod.
>
> Clipping casues harmonics and intermod.
>
> Matematically, clipping alone creates new frequencies, The others do not.
>
>
>
> Mark

I tend to think, if I could actually clip a sine-wave, I would not hear anything for a brief moment. You couldn't clip with magnetic tape, since it just keeps saturating more. However, you may be able to do it with digital, but maybe not, where it ends up as DC, then I'd think you'd hear harmonics.

Great subject.

Jack

[[email protected]]

On Tuesday, October 13, 2015 at 2:34:38 PM UTC-4, wrote:
> :
>
>
> Hmmm.. When I limit(flat top) a pure sine wave, even by
> just .5dB, it sounds different to me

if you flat top it, then it is not limiting, it is clipping

that is the essesnce of the difference

Mark

[[email protected]]

On Tuesday, October 13, 2015 at 2:55:31 PM UTC-4, wrote:
> Mike Rivers wrote: "On 10/13/2015 2:34 PM, wrote:
> > Hmmm.. When I limit(flat top) a pure sine wave, even by
> > just .5dB, it sounds different to me
>
> That's good, because when you clip off the tops it has more frequencies
> in it than without. But this isn't always a bad thing. Adding harmonics
> can make a boring timbre sound more musically interesting. It's done all
> the time, sometimes even naturally. But whether it's beneficial or
> harmful, it's still distortion.
>
> Try looking at the before and after with a spectrum analyzer and you'll
> have a better understanding of _why_ they sound different. If you're
> using a DAW that will host VST plug-ins (most can) SPAN from Voxengo is
> an excellent and free spectrum analyzer plug-in.
> - show quoted text -"
>
>
> One area it's not good is in many reissues labeled as "remastered",
> where a lot of the highest peaks are hard-limited, on top of the whole
> being compressed. It definitely sounds different, not just louder.

of course music, when limited sounds different because the dynamic range is changed... not becasue there are harmincs added....

I was talking about a lone sine wave

If a lone sine wave sounds (and looks) different, then it is clipping.

Mark

[geoff]

On 13/10/2015 7:54 p.m., wrote:
> geoff wrote: "Called 'Sound Check' . Just a simple normalisation thing, I think. "
>
>
> Ahh yes - Now you're talking! Although I seriously doubt
> it peak normalizes. I could expreimient with it by applying
> it in iTunes to a couple of test tracks: one very dynamic(high
> peak to avg ratio), the other, highly compressed(low peak
> to avg). If they both sound as loud as each other, then it is not
> peak-normalizing. The average is closer to what we hear, and
> if you align the tracks by peak, the average of the less dynamic
> track should sound louder.

I assume is is simple normalisation as it generally fails in making
things sound similar loudness. For it to be anything else would require
far more info that a simple 'peak-level' parameter in a header.

geoff

[geoff]

On 14/10/2015 7:34 a.m., wrote:
> :
>
>
> Hmmm.. When I limit(flat top) a pure sine wave, even by
> just .5dB, it sounds different to me


Um, you are not limiting it, you are clipping it.

Glad you can hear the difference, even if not understand it.


geoff

[geoff]

On 14/10/2015 7:55 a.m., wrote:
> Mike Rivers wrote: "On 10/13/2015 2:34 PM, wrote:
>> Hmmm.. When I limit(flat top) a pure sine wave, even by
>> just .5dB, it sounds different to me
> That's good, because when you clip off the tops it has more frequencies
> in it than without. But this isn't always a bad thing. Adding harmonics
> can make a boring timbre sound more musically interesting. It's done all
> the time, sometimes even naturally. But whether it's beneficial or
> harmful, it's still distortion.
>
> Try looking at the before and after with a spectrum analyzer and you'll
> have a better understanding of _why_ they sound different. If you're
> using a DAW that will host VST plug-ins (most can) SPAN from Voxengo is
> an excellent and free spectrum analyzer plug-in.
> - show quoted text -"
>
>
> One area it's not good is in many reissues labeled as "remastered",
> where a lot of the highest peaks are hard-limited, on top of the whole
> being compressed. It definitely sounds different, not just louder.

That is (in non-clipped examples) often because they have been EQed to
sound brighter and louder.

Not all remasters are done with that objective.

Actually-clipped music will sound brighter, harsh, and glarey. Just
what you want when listening to MP3s (etc, that add their own ****e) on
ear-buds that are themselves harsh and glarey....

geoff

[geoff]

On 14/10/2015 8:19 a.m., JackA wrote:
> O
> I tend to think, if I could actually clip a sine-wave, I would not hear anything for a brief moment. You couldn't clip with magnetic tape, since it just keeps saturating more. However, you may be able to do it with digital, but maybe not, where it ends up as DC, then I'd think you'd hear harmonics.
>
> Great subject.
>
> Jack

" I tend to think " - yeah right ....

Easy to clip program with the electronics prior to the actual tape.

geoff

[[email protected]]

geoff:

STOP telling me what I don't understand! You don't
know me at all.

That said, here is a good diagram illustrating
clipping: http://www.gmarts.org/pix/fx/fx_clip1.jpg

When I utilize "hard-limiter" in a DAW, the result
zoomed in looks exactly like the red lines in that
diagramSo does anyone know what the F- limiting
looks like in a DAW?!

[geoff]

On 13/10/2015 8:42 p.m., Trevor wrote:
>
> Bull****!! A limiter by it's very nature will produce CLIPPING when it
> exceeds it's knee and reaches it's maximum level. That's the whole
> point of the device after all. One may choose to use it so it never
> exceeds the knee (and hard limiting may not even have a knee!) but IF
> it does exceed absolute maximum, then CLIPPING will occur! That you
> choose to believe it is somehow not clipping just because YOU prefer
> to call it limiting only shows you ignorance!!!

Bull**** you ! A crude limiter might - a sophisticated limiter will
produce a 'knee' region that should not resemble clipping if you look
closer.

Glad my ignorance keeps such good company.

>
> Yep, but limiting is NOT necessarily the same as compression. You and
> many others still seem to be totally confused about that. And of
> course one can drive a compressor into clipping as well!
> Clipping is clipping no matter how it is achieved. Arguing that it is
> not is simply TOTAL BS!
>

Limiting should be a process closer to extreme compression that clipping.

geoff

[geoff]

On 13/10/2015 8:24 p.m., Trevor wrote:
>
> Yep, and yet MANY Cd's have HUNDREDS of consecutive samples at maximum
> level :-(
> No-one can argue THAT is *NOT* clipping simply because the CD is
> normalised to -0.3dBFS, CAN THEY? Or are some people who claim to be
> "pro's" really that stupid?
>
> Trevor.
>
>


Was I ever suggesting that some (even many) CDs or other digital media
does not exhibit clipping ?

Certainly not many of the sort that I purchase.

geoff

[[email protected]]

On Tuesday, October 13, 2015 at 4:20:19 PM UTC-4, wrote:
> On Tuesday, October 13, 2015 at 2:34:38 PM UTC-4, wrote:
> > :
> >
> >
> > Hmmm.. When I limit(flat top) a pure sine wave, even by
> > just .5dB, it sounds different to me
>
> if you flat top it, then it is not limiting, it is clipping
>
> that is the essesnce of the difference
>
> Mark

i shoudld clarfiy this...

if you flat-top the WAVEFORM, it is clipping..
if you see a sine wave waveform, and it is flat topped, that is clipping...


as discussed in another thread, the WAVEFORM is not the same as the ENVELOPE.

Limiting can and will flat top an envelope but not the waveform.

The difference between clipping and limiting, is closely related to the difference beween a WAVEFORM and an ENVELOPE.

Mark

[geoff]

On 14/10/2015 2:07 a.m., Mike Rivers wrote:
> On 10/12/2015 6:24 PM, geoff wrote:
>> My car iPod Touch has a automatic replay level function based on (I
>> think) a peak-level parameter encoded into the file.
>>
>> When it works on a particularly 'different level' track there seems to
>> be a second-long level adjustment at the commencement of a song
>
> I would like to have something like that in my car. It might not work
> as well for me as it does for you, however, because for me, a "song"
> (one file) is a two hour radio program of music that often has
> song-to-song level variations.
>
> I've been trying to score a cheap couple of generations old iPod Touch
> so I could use it for some audio apps that aren't available for my
> Androids, but so far no luck. They seem to hold their value nearly as
> well as microphones. ;)
>

I bought an iPod Touch (4G) 4+ years ago primarily for Signal Suite ,
Signal Scope Pro, and Guitar Toolkit. Pretty sure there will be Android
equivalents by now.

..... with the bonus of direct USB connection/control to my car stereo.
All music exclusively ripped from CD to ALAC.

geoff

[[email protected]]

..
>
> That said, here is a good diagram illustrating
> clipping: http://www.gmarts.org/pix/fx/fx_clip1.jpg
>
> When I utilize "hard-limiter" in a DAW, the result
> zoomed in looks exactly like the red lines in that
> diagramSo does anyone know what the F- limiting
> looks like in a DAW?!

Yes, you must Zoom in to see the WAVEFORM rather than the ENVELOPE.
And if the WAVEFORM appears as in the .jpg, then i would call that clipping (as the link does) even if the box that did it is called a limiter.

Mark

[geoff]

On 13/10/2015 8:42 p.m., Trevor wrote:
>
> Rubbish, When someone claims clipping is not clipping simply because
> it's level has been post reduced, or that clipping is not clipping
> simply because it was done by a "limiter" that is FARRRRR from hair
> splitting about definitions, that is about that person not
> understanding the basic fundamentals of recording!!!!!

Dunno where how you managed to extrapolate that out of anything I've said.

But you do appear to be claiming that all limiting is 'clipping', which
is crap.

geoff

[[email protected]]

On Tuesday, October 13, 2015 at 4:43:41 PM UTC-4, wrote:
> geoff:
>
> STOP telling me what I don't understand! You don't
> know me at all.
>
> That said, here is a good diagram illustrating
> clipping: http://www.gmarts.org/pix/fx/fx_clip1.jpg
>
> When I utilize "hard-limiter" in a DAW, the result
> zoomed in looks exactly like the red lines in that
> diagramSo does anyone know what the F- limiting
> looks like in a DAW?!

Limiting will maintain the sine wave shape of the WAVEFORM.

Limiting can squash the ENVELOPE (what you see when you don't zoom in)

Mark

[[email protected]]

wrote: "Limiting can and will flat top an envelope but not the waveform. "

Ok, flatten the loudest parts of a whole song envelope. But when I
zoom down in my daw to individual waves what do I see? FLAT
TOPS on the tallest waves!

Explain THAT, professor.

[geoff]

On 14/10/2015 10:02 a.m., wrote:
> wrote: "Limiting can and will flat top an envelope but not the waveform. "
>
> Ok, flatten the loudest parts of a whole song envelope. But when I
> zoom down in my daw to individual waves what do I see? FLAT
> TOPS on the tallest waves!
>
> Explain THAT, professor.


Easy. That is clipping, if the flat top exceeds (popularly accepted) 3
samples.

geoff

[[email protected]]

I'm going to try to distill this down as simple as i can.

Clipping will alter the WAVEFORM and the ENVELOPE. Since the waveform is altered, harmonics are created.


Pure limiting will NOT alter the WAVEFORM (and hence not create harmonics) but limiting will alter the ENVELOPE of course.

over and out....

Mark

[[email protected]]

On Tuesday, October 13, 2015 at 5:02:53 PM UTC-4, wrote:
> wrote: "Limiting can and will flat top an envelope but not the waveform. "
>
> Ok, flatten the loudest parts of a whole song envelope. But when I
> zoom down in my daw to individual waves what do I see? FLAT
> TOPS on the tallest waves!
>
> Explain THAT, professor.

Then you also have clipping

Mark

[geoff]

On 14/10/2015 9:43 a.m., wrote:
> geoff:
>
> STOP telling me what I don't understand! You don't
> know me at all.
>
> That said, here is a good diagram illustrating
> clipping: http://www.gmarts.org/pix/fx/fx_clip1.jpg
>
> When I utilize "hard-limiter" in a DAW, the result
> zoomed in looks exactly like the red lines in that
> diagramSo does anyone know what the F- limiting
> looks like in a DAW?!

Depends on the DAW and what it is doing when you implement it's 'hard
limiting'.

If it clips (actually clips) and that's not what you want, tweak the
controls, use a different plugin, or buy a different DAW (or editor).

geoff

[Scott Dorsey]

In article >,
> wrote:
wrote: "Limiting can and will flat top an envelope but not the waveform. "
>
>Ok, flatten the loudest parts of a whole song envelope. But when I
>zoom down in my daw to individual waves what do I see? FLAT
>TOPS on the tallest waves!
>
>Explain THAT, professor.

You might want to go back to the discussion a year or so ago where a dozen
people attempted futilely to explain to you the difference between envelopes
and waveforms.
--scott

--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[JackA]

On Tuesday, October 13, 2015 at 4:20:19 PM UTC-4, wrote:
> On Tuesday, October 13, 2015 at 2:34:38 PM UTC-4, wrote:
> > :
> >
> >
> > Hmmm.. When I limit(flat top) a pure sine wave, even by
> > just .5dB, it sounds different to me
>
> if you flat top it, then it is not limiting, it is clipping

-- I just want to know HOW to get a sinewave to clip. No crest, no sound. Flat-topping would create the harmonics.

Thanks.

Jack

>
> that is the essesnce of the difference
>
> Mark

[JackA]

On Tuesday, October 13, 2015 at 4:43:29 PM UTC-4, geoff wrote:
> On 14/10/2015 8:19 a.m., JackA wrote:
> > O
> > I tend to think, if I could actually clip a sine-wave, I would not hear anything for a brief moment. You couldn't clip with magnetic tape, since it just keeps saturating more. However, you may be able to do it with digital, but maybe not, where it ends up as DC, then I'd think you'd hear harmonics.
> >
> > Great subject.
> >
> > Jack
>
> " I tend to think " - yeah right ....
>
> Easy to clip program with the electronics prior to the actual tape.

I know, I know, show me a clipped sinewave on an oscilloscope, smart guy!! :-)

Jack
>
> geoff

[JackA]

On Tuesday, October 13, 2015 at 3:42:38 AM UTC-4, Trevor wrote:
> On 13/10/2015 7:50 AM, geoff wrote:
> > On 13/10/2015 3:23 a.m., Trevor wrote:
> >> On 13/10/2015 12:25 AM, Scott Dorsey wrote:
> >>> geoff > wrote:
> >>>> No CD that I've purchased in the last decade exhibit clipping that I've
> >>>> noticed, though a few are over-compressed to hell (not many
> >>>> fortunately). And some I've been prompted to actually check ! Not much
> >>>> 'current' pop though I concede.
> >>>
> >>> The problem is that in the digital world, clipping is whatever you
> >>> define
> >>> it as. I tend to set metering so three consecutive FS samples light the
> >>> over light, and so that is clipping.
> >>
> >> No, you first have to normalise the gain back to 0dBFS for that to
> >> work since many CD's are first *severely clipped* then normalised to
> >> about -0.3dBFS, so your clip lights will never come on. BUT the flat
> >> tops remain regardless!
> >>
> >>>
> >>> Aggressive limiting that flat-tops the signal isn't necessarily
> >>> clipping,
> >> > it's just aggressive limiting.
> >>
> >> Rubbish, peak limiting that causes flat tops IS clipping. YOU simply
> >> don't understand what they have done, or the difference between
> >> compression and limiting it seems.
> >
> > I think this is the only thing where we really disagree. The rest of
> > your debate seems hair-splitting about definitions of terms.
>
> Rubbish, When someone claims clipping is not clipping simply because
> it's level has been post reduced, or that clipping is not clipping
> simply because it was done by a "limiter" that is FARRRRR from hair
> splitting about definitions, that is about that person not understanding
> the basic fundamentals of recording!!!!!
>
>
>
> >
> > I say that 'clipping' is only the result of a digital or analogue
> > *overload*, where the actual mathematical or electrical constraints of
> > the process are exceeded and nothing can exist above.
>
> Right, as occurs when a limiter exceeds it's absolute peak level, a
> mixer or amplifier exceeds it's maximum rail voltage, or a digital
> device exceeds it's digital full scale level, whether OR NOT, that level
> is subsequently reduced AFTER clipping!!
>
> >
> > 'Limiting' is the *controlled* result of a digital or analogue process.
> > And yes, that can be extreme to the point of resembling clipping, but is
> > not the same thing.
>
> Bull****!! A limiter by it's very nature will produce CLIPPING when it
> exceeds it's knee and reaches it's maximum level. That's the whole point
> of the device after all. One may choose to use it so it never exceeds
> the knee (and hard limiting may not even have a knee!) but IF it does
> exceed absolute maximum, then CLIPPING will occur! That you choose to
> believe it is somehow not clipping just because YOU prefer to call it
> limiting only shows you ignorance!!!
>
> >
> > Achieving, or adding to, limiting by clipping (sadly) is done,
> > presumably by those who are incompetent, lack understanding of the
> > implications, or have cynical intent.
>
> Yep, but limiting is NOT necessarily the same as compression. You and
> many others still seem to be totally confused about that. And of course
> one can drive a compressor into clipping as well!
> Clipping is clipping no matter how it is achieved. Arguing that it is
> not is simply TOTAL BS!
>
> Trevor.

I guess you could say clipping is when a steady state occurs in audio, nothing is changing. Nothing changes, no sound. Sort of like listening to a battery connected to a speaker - power is used, but no sound.

Jack

[None]

< whineybitch @ gmail.com > wrote in message
...
> geoff:
>
> STOP telling me what I don't understand! You don't know me at all.

You've been making a huge stinking public display of what you don't
understand, for some years now. What makes you think nobody would
notice?

[None]

< thekma @ shortbus.edu > wrote in message
...
> wrote: "Limiting can and will flat top an envelope
> but not the waveform. "
>
> Ok, flatten the loudest parts of a whole song envelope. But when I
> zoom down in my daw to individual waves what do I see? FLAT
> TOPS on the tallest waves!
>
> Explain THAT, professor.

It's been explained to you, probably hundreds of times now. Explaining
it is a waste of time. You're not capable of understanding.

[JackA]

On Tuesday, October 13, 2015 at 5:09:12 PM UTC-4, geoff wrote:
> On 14/10/2015 9:43 a.m., wrote:
> > geoff:
> >
> > STOP telling me what I don't understand! You don't
> > know me at all.
> >
> > That said, here is a good diagram illustrating
> > clipping: http://www.gmarts.org/pix/fx/fx_clip1.jpg
> >
> > When I utilize "hard-limiter" in a DAW, the result
> > zoomed in looks exactly like the red lines in that
> > diagramSo does anyone know what the F- limiting
> > looks like in a DAW?!
>
> Depends on the DAW and what it is doing when you implement it's 'hard
> limiting'.
>
> If it clips (actually clips) and that's not what you want, tweak the
> controls, use a different plugin, or buy a different DAW (or editor).

And may I say, most people who hang-out in usenet have pirated copies of Pro Tools; same with Photoshop. That's why, after several years, I decided to purchase Goldwave.

Goldwave can even remove varying DC Offset, even though Scott said there is no such thing, but since he now ignores me, he must have found out there is such a thing!!

Welcome to usenet.

Jack
Jack
>
> geoff

[None]

< thkema @ gurgle.dum****sRus.com > wrote in message
...
> A SIDE NOTE:
> Via google groups, this thread, blah blah blah blah blah blah blah
> blah blah blah blah blah blah blah blah blah blah blah blah

Usenet does not give a **** about gurgle groups. Get a real newsreader
or stop your whining.

[None]

< ignorant-ignoramus @ shortbus.edu > wrote in message
...
> So I DO understand what's going on Mike, Scott, geoff, and

Still utterly failing to ignore me, I see. You understand how to
ignore me about as much as you understand audio. Dumb****.

> who shall remain nameless,

Heh. You just posted about me, again. You're as sharp as a bowling
ball!

> it's putting things into words where I stumble.

[geoff]

On 14/10/2015 12:15 p.m., JackA wrote:
> And may I say, most people who hang-out in usenet have pirated copies
> of Pro Tools; same with Photoshop. That's why, after several years, I
> decided to purchase Goldwave.

Maybe the other odd places that you hang out, with amateurs dabbling.

Here most people are professionals who pay for their software

geoff

[JackA]

On Tuesday, October 13, 2015 at 7:27:24 PM UTC-4, geoff wrote:
> On 14/10/2015 12:15 p.m., JackA wrote:
> > And may I say, most people who hang-out in usenet have pirated copies
> > of Pro Tools; same with Photoshop. That's why, after several years, I
> > decided to purchase Goldwave.
>
> Maybe the other odd places that you hang out, with amateurs dabbling.
>
> Here most people are professionals who pay for their software

I hope you don't mind me, an amateur, hanging out with the professionals!! I feel so honored!

Jack

>
> geoff

[Trevor]

On 14/10/2015 1:58 AM, Scott Dorsey wrote:
> > wrote:
>>
>> a good way to differentiate clipping from limiting from compression is by
>> considering the attack and decay times relative to the features of the waveform
>
> That's half of it... and the other half is the shape of the knee. But how
> fast is fast? When does soft clipping become limiting and vice-versa?

There is NO such thing as "soft clipping". It's either clipped or it's
not. There is soft limiting however which you probably mean, and it's
very simple, when the knee ends and the signal flat tops for a few
samples (usually accepted as at least 3 or more) then it has gone into
clipping.
I'm surprised you still don't grasp this?

Trevor.

[[email protected]]

"You've been making a huge stinking public display of what you don't
understand, for some years now. What makes you think nobody would notice"


YOU HAVEN'T CONTRIBUTED ANYTHING TO
THIS CONVERSATION - SO BUTT OUT, YOU
OVERFLOWING BED PAN!

I DON'T KNOW WHAT I DID TO OFFEND YOU,
AND YOU NEVER TOLD ME. EASY TO BERATE
SOMEONE ANONYMOUSLY, OVER THE INTERNET.

Dorsey, geoff, Mark, & Rivers, do not let this
used sanitary napkin deter you from mature
adult conversation. And do not offer it ANY help
when it is having trouble or asks a question.
Too many threads have died because of it.
Even if I am the only one it attacks, that
doesn't mean anyone should be strengthening
alliances with it.

[Trevor]

On 14/10/2015 2:02 AM, Scott Dorsey wrote:
> In article >, Trevor > wrote:
>>> Many samples?
>>
>> At least 3 or more. (genuine square waves excluded) And as I have said,
>> some CD's have HUNDREDS of consecutive samples at maximum level, and
>> often many similar groups in one song. IF there are only 3 in fact I'm
>> not worried in the slightest, when there are hundreds I am. That is far
>> more common with pop CD's these days than many people seem to think,
>> simply because they never look.
>
> Well, that's the degenerate case. Those CDs are clearly clipped. But I'm
> talking about the borderline cases, because that's where it gets interesting.

Few borderline cases in the pop world any more.

>>> At what point does limiting end and clipping begin?
>>
>> As soon as the knee ends and maximum level is reached for at least 3 or
>> more samples. I'm surprised you don't understand this already It's not
>> rocket science!
>
> That's a good definition. Although... I might decide it is 2 samples or
> 8 samples and be able to make a good argument for those too.

Go ahead. As I said 3 doesn't bother me, it's just a minimum. 8 would
mean I've done something wrong, but 1000 doesn't bother some mastering
engineers! :-(


>> I'm totally amazed that someone with your experience still doesn't have
>> a vague notion of what clipping actually is. It can occur anywhere in
>> the chain, and is independent of final level. Just like you can clip a
>> Mackie mixers mix bus, even if the output fader is well below maximum.
>> In fact you can clip a single channel on ANY mixer, without clipping the
>> mixer output, or amplifier input. THAT channel is STILL clipped!!!
>
> Oh, I have a vague notion of what clipping actually is.... and a vague notion
> of what limiting actually is.... but I can think of a LOT of examples that
> are sitting directly on the border. I don't want a vague notion, I want a
> precise mathematical description.

I gave you one above, and it's widely accepted by lots of recording
people, other than yourself.


> My line might be "if you can hear it, it's clipping, if you can't, it's
> limiting." That's no less vague, though.

Right, that is simply subjective, the accepted (by everyone else)
definition I gave is NOT.

Trevor.

[Randy Yates[_2_]]

writes:

>>
>> >> At what point does limiting end and clipping begin?
>> >
>
> there __is__ a definitive answer to that question
>
> with clipping , the gain changes fast enough to follow the cycle by cycle waveform ____and therefore creates harmonic and intermodulation distortion___.
> If you clip a sine wave, you can hear the harmonics.
> Mathematically, this is a non-linar process.
>
> Anything slower than that, is limiting, compression or AGC.
> These are mmathematically linear process

They are? I don't think so, Mark.

Let's go back to basics. A transformation T is linear iff

T(a*x1 + b*x2) = a*T(x1) + b*T(x2)

So let's consider a very simple limiter that does this:

L(x) = x, |x| < 1
L(x) = sgn(x) * 1, |x| >= 1.

Is this linear by the definition above? Nope. Here's
a simple counterexample.

Let x1 = 0.75, x2 = 0.75, a = 1, and b = 1. Then


a*T(x1) + b*T(x2) = 1 * 0.75 + 1 * 0.75
= 1.5

but

T(a*x1 + b*x2) = T(1.5) = 1.

Not linear.

I think you knew this. I'm not sure why you think it's linear.

--Randy


and do not create harmonics or intermod.
> If you limit or compress or AGC a sine wave, you will not hear
> harmonics.
>
>
> (I'm not saying these are good, they can still ruin the aesthtics by squashing the dynamic range, but they do not cause harmonics or intermod.
>
> Clipping casues harmonics and intermod.
>
> Matematically, clipping alone creates new frequencies, The others do
> not.


>
>
>
> Mark
>

--
Randy Yates
Digital Signal Labs
http://www.digitalsignallabs.com

[Trevor]

On 14/10/2015 5:21 AM, wrote:
>>>> At what point does limiting end and clipping begin?
>
> there __is__ a definitive answer to that question
>
> with clipping , the gain changes fast enough to follow the cycle by
> cycle waveform ____and therefore creates harmonic and intermodulation
> distortion___. If you clip a sine wave, you can hear the harmonics.
> Mathematically, this is a non-linar process.
>
> Anything slower than that, is limiting, compression or AGC. These are
> mmathematically linear process and do not create harmonics or
> intermod. If you limit or compress or AGC a sine wave, you will not
> hear harmonics.
>
>
> (I'm not saying these are good, they can still ruin the aesthtics by
> squashing the dynamic range, but they do not cause harmonics or
> intermod.
>
> Clipping casues harmonics and intermod.
>
> Matematically, clipping alone creates new frequencies, The others do
> not.

Fourier would disagree with that analysis. Take an easy example, hard
limit a sine wave, it is no longer a perfect sine wave and therefore
must contain harmonics. The others will all do the same to a lesser degree.

Trevor.

[Trevor]

On 14/10/2015 7:20 AM, wrote:
> On Tuesday, October 13, 2015 at 2:34:38 PM UTC-4, wrote:
>> :
>> Hmmm.. When I limit(flat top) a pure sine wave, even by
>> just .5dB, it sounds different to me
>
> if you flat top it, then it is not limiting, it is clipping
> that is the essesnce of the difference

So you think a limiter cannot cause clipping? I guess you have no idea
what one does then!

Trevor.

[Trevor]

On 14/10/2015 7:23 AM, wrote:
> If a lone sine wave sounds (and looks) different, then it is clipping.

Not necessarily. *Squash* a sine wave without clipping, ie it has
slightly flattened but still rounded tops, and it will sound different
due to the extra harmonics added. Only when it has flat tops can it
accurately be said to be clipped, by definition. Until then it is just
compressed or soft limited.

Trevor.

[Trevor]

On 14/10/2015 7:43 AM, geoff wrote:
> On 14/10/2015 8:19 a.m., JackA wrote:
>> I tend to think, if I could actually clip a sine-wave, I would not
>> hear anything for a brief moment. You couldn't clip with magnetic
>> tape, since it just keeps saturating more. However, you may be able to
>> do it with digital, but maybe not, where it ends up as DC, then I'd
>> think you'd hear harmonics.
>>
>> Great subject.
>>
>
> " I tend to think " - yeah right ....

:-)

> Easy to clip program with the electronics prior to the actual tape.

And the tape itself does not have infinite flux capability so it MUST
clip eventually. The only difference is it soft limits before that point.

Trevor.

[Trevor]

On 14/10/2015 7:49 AM, geoff wrote:
> On 13/10/2015 8:42 p.m., Trevor wrote:
>>
>> Bull****!! A limiter by it's very nature will produce CLIPPING when it
>> exceeds it's knee and reaches it's maximum level. That's the whole
>> point of the device after all. One may choose to use it so it never
>> exceeds the knee (and hard limiting may not even have a knee!) but IF
>> it does exceed absolute maximum, then CLIPPING will occur! That you
>> choose to believe it is somehow not clipping just because YOU prefer
>> to call it limiting only shows you ignorance!!!
>
> Bull**** you ! A crude limiter might - a sophisticated limiter will
> produce a 'knee' region that should not resemble clipping if you look
> closer.

As I have been saying ALL along, BUT will still clip past the knee.
Do you NOT get this? Or didn't bother to read what you are replying to?


> Glad my ignorance keeps such good company.

Yes, you are in fine company here. Judging by others posts many have a
problem with both the technicalities AND reading comprehension! :-(


>> Yep, but limiting is NOT necessarily the same as compression. You and
>> many others still seem to be totally confused about that. And of
>> course one can drive a compressor into clipping as well!
>> Clipping is clipping no matter how it is achieved. Arguing that it is
>> not is simply TOTAL BS!
>>
>
> Limiting should be a process closer to extreme compression that clipping.

And it IS, until the maximum level is reached, at which point it clips,
How many times do I have to say the same thing?

Trevor.

[Trevor]

On 14/10/2015 7:50 AM, geoff wrote:
> On 13/10/2015 8:24 p.m., Trevor wrote:
>> Yep, and yet MANY Cd's have HUNDREDS of consecutive samples at maximum
>> level :-(
>> No-one can argue THAT is *NOT* clipping simply because the CD is
>> normalised to -0.3dBFS, CAN THEY? Or are some people who claim to be
>> "pro's" really that stupid?
>
>
> Was I ever suggesting that some (even many) CDs or other digital media
> does not exhibit clipping ?

Where did I suggest you did?

> Certainly not many of the sort that I purchase.

Good for you.

Trevor.

[Trevor]

On 14/10/2015 7:50 AM, wrote:
> On Tuesday, October 13, 2015 at 4:20:19 PM UTC-4,
> wrote:
>> On Tuesday, October 13, 2015 at 2:34:38 PM UTC-4,
>> wrote:
>>> :
>>>
>>>
>>> Hmmm.. When I limit(flat top) a pure sine wave, even by just
>>> .5dB, it sounds different to me
>>
>> if you flat top it, then it is not limiting, it is clipping
>>
>> that is the essesnce of the difference
>>
>> Mark
>
> i shoudld clarfiy this...
>
> if you flat-top the WAVEFORM, it is clipping.. if you see a sine wave
> waveform, and it is flat topped, that is clipping...
>
>
> as discussed in another thread, the WAVEFORM is not the same as the
> ENVELOPE.
>
> Limiting can and will flat top an envelope but not the waveform.
>
> The difference between clipping and limiting, is closely related to
> the difference beween a WAVEFORM and an ENVELOPE.


As previously discussed, the envelope is simply an imaginary concept.
The waveform is the only reality whether you choose to zoom in or out.

Trevor.

[Trevor]

On 14/10/2015 7:59 AM, wrote:
> .
>>
>> That said, here is a good diagram illustrating clipping:
>> http://www.gmarts.org/pix/fx/fx_clip1.jpg
>>
>> When I utilize "hard-limiter" in a DAW, the result zoomed in looks
>> exactly like the red lines in that diagramSo does anyone know what
>> the F- limiting looks like in a DAW?!
>
> Yes, you must Zoom in to see the WAVEFORM rather than the ENVELOPE.

Nope it is all waveform (the envelope is simply imaginary) but yes you
must zoom in to see the waveform clearly, or in any detail.

> And if the WAVEFORM appears as in the .jpg, then i would call that
> clipping (as the link does) even if the box that did it is called a
> limiter.

Dead right!
However what is labelled "soft clipping" is technically incorrect, that
is compressed, but not yet clipped.

Trevor.

[Trevor]

On 14/10/2015 7:59 AM, geoff wrote:
> On 13/10/2015 8:42 p.m., Trevor wrote:
>>
>> Rubbish, When someone claims clipping is not clipping simply
>> because it's level has been post reduced, or that clipping is not
>> clipping simply because it was done by a "limiter" that is FARRRRR
>> from hair splitting about definitions, that is about that person
>> not understanding the basic fundamentals of recording!!!!!
>
> Dunno where how you managed to extrapolate that out of anything I've
> said.

Easy, just put back the bit I responded to and you deliberately snipped,
and it will all become clear. Of course I think it might have been
someone else I replied to anyway and you are being even more silly with
that reply.


> But you do appear to be claiming that all limiting is 'clipping',
> which is crap.

Dunno where how you managed to extrapolate that out of anything I've
said! In fact I have been VERY clear to spell it out even if you can't
read or understand. Let me try once more for the slow learners, Limiting
CAN become clipping when the knee (if there is one) is past.
Not all limiting causes clipping, but it certainly CAN!

Trevor.

[Trevor]

On 14/10/2015 8:02 AM, wrote:
> Limiting will maintain the sine wave shape of the WAVEFORM.

MAY, not will.

> Limiting can squash the ENVELOPE (what you see when you don't zoom in)

So if you think that is what a limiter does, what do you think a
compressor does? Or don't you get the difference?

Trevor.

[Trevor]

On 14/10/2015 8:07 AM, wrote:
> I'm going to try to distill this down as simple as i can.
>
> Clipping will alter the WAVEFORM and the ENVELOPE. Since the
> waveform is altered, harmonics are created.
>
>
> Pure limiting will NOT alter the WAVEFORM (and hence not create
> harmonics) but limiting will alter the ENVELOPE of course.

Gee I would have thought the word "Limit" might give you some clue as to
the difference between a compressor and a limiter!
Soft limiting may compress of course before it eventually "limits", and
thus clips.

Trevor.

[Trevor]

On 14/10/2015 8:09 AM, geoff wrote:
> On 14/10/2015 9:43 a.m., wrote:
>> geoff:
>>
>> STOP telling me what I don't understand! You don't
>> know me at all.
>>
>> That said, here is a good diagram illustrating
>> clipping: http://www.gmarts.org/pix/fx/fx_clip1.jpg
>>
>> When I utilize "hard-limiter" in a DAW, the result
>> zoomed in looks exactly like the red lines in that
>> diagramSo does anyone know what the F- limiting
>> looks like in a DAW?!
>
> Depends on the DAW and what it is doing when you implement it's 'hard
> limiting'.
>
> If it clips (actually clips) and that's not what you want, tweak the
> controls, use a different plugin, or buy a different DAW (or editor).


Love to hear YOUR definition of HARD limiting? And the difference
between HARD limiting and clipping? :-)

Trevor.

[geoff]

On 14/10/2015 4:05 p.m., Trevor wrote:
> On 14/10/2015 8:09 AM, geoff wrote:
>> On 14/10/2015 9:43 a.m., wrote:
>>> geoff:
>>>
>>> STOP telling me what I don't understand! You don't
>>> know me at all.
>>>
>>> That said, here is a good diagram illustrating
>>> clipping: http://www.gmarts.org/pix/fx/fx_clip1.jpg
>>>
>>> When I utilize "hard-limiter" in a DAW, the result
>>> zoomed in looks exactly like the red lines in that
>>> diagramSo does anyone know what the F- limiting
>>> looks like in a DAW?!
>>
>> Depends on the DAW and what it is doing when you implement it's 'hard
>> limiting'.
>>
>> If it clips (actually clips) and that's not what you want, tweak the
>> controls, use a different plugin, or buy a different DAW (or editor).
>
>
> Love to hear YOUR definition of HARD limiting? And the difference
> between HARD limiting and clipping? :-)
>
> Trevor.
>
>
Clipping = individual cycles of waveforms clipped level, 3 or more
cycles (yes, even if subsequently reduced in level).

Limiting = reduced variation of waveform envelope, with the threshold
usually higher which differentiates it from compression.

Hard limiting = very small variation in the amplitude of the waveform
envelope. Arguably including a degree of individual cycle waveform
distortion resembling the previously-linked 'soft-clipping' picture (as
long as no more than 3 samples over 0dBFS, even in a prior processing
stage and subsequently reduced).

geoff

[geoff]

On 14/10/2015 3:55 p.m., Trevor wrote:
>
> Not all limiting causes clipping, but it certainly CAN!
>
> Trevor.

" As I have been saying ALL along, BUT will still clip past the knee. "

So any limiting is clipping, and if it isn't clipping it isn't limiting !

I give up. You win. Bye.

geoff

[Trevor]

On 14/10/2015 2:29 PM, geoff wrote:
> On 14/10/2015 3:55 p.m., Trevor wrote:
>>
>> Not all limiting causes clipping, but it certainly CAN!
> " As I have been saying ALL along, BUT will still clip past the knee. "
^^^^^^^^^^^^^

> So any limiting is clipping,

Wow you *DO* have a problem with reading comprehension!


> and if it isn't clipping it isn't limiting !

Not totally wrong actually, it's really just compression until it
"limits", ie clips. But then the word "limiter" is used for a reason
after all, it's NOT just another compressor.

Trevor

[Trevor]

On 14/10/2015 2:23 PM, geoff wrote:
> On 14/10/2015 4:05 p.m., Trevor wrote:
>> On 14/10/2015 8:09 AM, geoff wrote:
>>> On 14/10/2015 9:43 a.m., wrote:
>>>> geoff:
>>>>
>>>> STOP telling me what I don't understand! You don't
>>>> know me at all.
>>>>
>>>> That said, here is a good diagram illustrating
>>>> clipping: http://www.gmarts.org/pix/fx/fx_clip1.jpg
>>>>
>>>> When I utilize "hard-limiter" in a DAW, the result
>>>> zoomed in looks exactly like the red lines in that
>>>> diagramSo does anyone know what the F- limiting
>>>> looks like in a DAW?!
>>>
>>> Depends on the DAW and what it is doing when you implement it's 'hard
>>> limiting'.
>>>
>>> If it clips (actually clips) and that's not what you want, tweak the
>>> controls, use a different plugin, or buy a different DAW (or editor).
>>
>>
>> Love to hear YOUR definition of HARD limiting? And the difference
>> between HARD limiting and clipping? :-)
>>
>>
>>
> Clipping = individual cycles of waveforms clipped level, 3 or more
> cycles (yes, even if subsequently reduced in level).
>
> Limiting = reduced variation of waveform envelope, with the threshold
> usually higher which differentiates it from compression.
>
> Hard limiting = very small variation in the amplitude of the waveform
> envelope. Arguably including a degree of individual cycle waveform
> distortion resembling the previously-linked 'soft-clipping' picture (as
> long as no more than 3 samples over 0dBFS, even in a prior processing
> stage and subsequently reduced).
>

OK, if that very subjective definition suits you, but obviously would
only work with post processing, because no real time limiter can know
where the peak level input is ahead of time to make sure it doesn't clip
with a high threshold setting and small knee. So what do you call all
those analog limiters I wonder? Clippers perhaps? Maybe not a bad idea! :-)

Trevor.

[Gray_Wolf]

On 10/13/2015 1:46 PM, Mike Rivers wrote:
> On 10/13/2015 2:34 PM, wrote:
>> Hmmm.. When I limit(flat top) a pure sine wave, even by
>> just .5dB, it sounds different to me
>
> That's good, because when you clip off the tops it has more frequencies in it
> than without. But this isn't always a bad thing. Adding harmonics can make a
> boring timbre sound more musically interesting. It's done all the time,
> sometimes even naturally. But whether it's beneficial or harmful, it's still
> distortion.

I presume this is where the exciter comes in. Adding harmonics while making not
it sound like a fuzz box .

[Luxey]

I think, ...
Problem with invisible clipping is not in the sound, because presumably, all interested parties have heard it prior to commiting to it.
It is in "silent" overloading of output stages and what comes with it, possibly all the way to blown out tweeters, depending on quality of playback gear and the way digital and analog meters were alligned to each other.

[John Williamson]

On 14/10/2015 03:25, Trevor wrote:
> On 14/10/2015 7:23 AM, wrote:
>> If a lone sine wave sounds (and looks) different, then it is clipping.
>
> Not necessarily. *Squash* a sine wave without clipping, ie it has
> slightly flattened but still rounded tops, and it will sound different
> due to the extra harmonics added. Only when it has flat tops can it
> accurately be said to be clipped, by definition. Until then it is just
> compressed or soft limited.
>
Some amplification stages don't clip the top of a waveform dead flat
when overdriven. In fact, I have seen ringing on the top section before
now. Do you call that clipping or limiting?

--
Tciao for Now!

John.

[John Williamson]

On 14/10/2015 03:13, Trevor wrote:
> On 14/10/2015 7:20 AM, wrote:
>> On Tuesday, October 13, 2015 at 2:34:38 PM UTC-4,
>> wrote:
>>> :
>>> Hmmm.. When I limit(flat top) a pure sine wave, even by
>>> just .5dB, it sounds different to me
>>
>> if you flat top it, then it is not limiting, it is clipping
>> that is the essesnce of the difference
>
> So you think a limiter cannot cause clipping? I guess you have no idea
> what one does then!
>
Apart from the simplest case of a pair of back to back diodes across the
signal path, a limiter running within its design spec shuold not
noticeably clip the waveform, as they work by reducing the gain in an
amplifier stage according to the input level.


--
Tciao for Now!

John.

[Mike Rivers[_2_]]

On 10/14/2015 1:03 AM, gray_wolf wrote:
> I presume this is where the exciter comes in. Adding harmonics while
> making not it sound like a fuzz box .

If you're talking about the Aphex Aural Exciter, that added 2nd harmonics.

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

[Trevor]

On 14/10/2015 6:44 PM, John Williamson wrote:
> On 14/10/2015 03:25, Trevor wrote:
>> On 14/10/2015 7:23 AM, wrote:
>>> If a lone sine wave sounds (and looks) different, then it is clipping.
>>
>> Not necessarily. *Squash* a sine wave without clipping, ie it has
>> slightly flattened but still rounded tops, and it will sound different
>> due to the extra harmonics added. Only when it has flat tops can it
>> accurately be said to be clipped, by definition. Until then it is just
>> compressed or soft limited.
>>
> Some amplification stages don't clip the top of a waveform dead flat
> when overdriven.

Not until you overdrive them a bit further anyway. everything in a
finite world will clip eventually.


> In fact, I have seen ringing on the top section before
> now. Do you call that clipping or limiting?

Why in hell would I call ringing, clipping *OR* limiting?
Of course the ringing may drive the output into clipping as well, but
that a separate issue.

Trevor.

[Mike Rivers[_2_]]

On 10/14/2015 3:47 AM, John Williamson wrote:
> Apart from the simplest case of a pair of back to back diodes across the
> signal path, a limiter running within its design spec shuold not
> noticeably clip the waveform, as they work by reducing the gain in an
> amplifier stage according to the input level.

A limiter reduces the gain above the threshold, but with sufficiently
fast rise time, it doesn't change the slew rate of the waveform. This
changes the shape of the waveform, and that's, by definition, distortion.

"Clipping" is a specific case of distortion, but it seems to have
entered the public domain to mean anything that makes all the waveform
peaks the same amplitude.

Compression can, and often is used to modify the rise and fall times of
a waveform, which can significantly alter timbre. You can think of it as
working like a synthesizer that has adjustable attack and release times.
But when used as means to reduce dynamic range, the attack and release
times are slow enough so that they operate on the envelope (many cycles)
rather than individual cycles. The waveform isn't buggered once the gain
change stabilizes for a few cycles.

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

[Trevor]

On 14/10/2015 6:47 PM, John Williamson wrote:
> On 14/10/2015 03:13, Trevor wrote:
>> On 14/10/2015 7:20 AM, wrote:
>>> On Tuesday, October 13, 2015 at 2:34:38 PM UTC-4,
>>> wrote:
>>>> :
>>>> Hmmm.. When I limit(flat top) a pure sine wave, even by
>>>> just .5dB, it sounds different to me
>>>
>>> if you flat top it, then it is not limiting, it is clipping
>>> that is the essesnce of the difference
>>
>> So you think a limiter cannot cause clipping? I guess you have no idea
>> what one does then!
>>
> Apart from the simplest case of a pair of back to back diodes across the
> signal path, a limiter running within its design spec shuold not
> noticeably clip the waveform, as they work by reducing the gain in an
> amplifier stage according to the input level.

Absolute BS. They work however the operator sets the knobs. BUT the
whole point of a limiter as opposed to a compressor is to catch
unexpected transients, and since the output is finite, at some point it
MAY clip when doing the job it is designed to do! Of course if there are
NO unexpected transients that exceed the knee of the limiter it will not
clip, in which case you probably didn't need it and a compressor would
have done the job better. But in a live situation you don't always know
that. And in a studio I would never use a limiter myself. (as opposed to
a compressor)
And to be clear, just because a single box MAY contain both a compressor
and a limiter, they are still 2 separate processes from a technical
aspect. Many here just don't seem to grasp that.

Trevor.

[None]

< thekma @ thekma.thekma.thek.ma > ranted and drooled in message
...
> "You've been making a huge stinking public display of what you don't
> understand, for some years now. What makes you think nobody would
> notice"
>
> YOU HAVEN'T CONTRIBUTED ANYTHING TO THIS CONVERSATION - SO BUTT OUT,

Despite your little caps-lock toddler tantrum, the fact remains that
you have made a huge display of what you don't understand. Anyone
who's been reading this groups for any significant amount of time
knows a lot about what you don't understand, and your denial of that
fact is pathetic. And here you are, back to your tricks of not
understanding something that's been explained to you in detail, over
and over and over and over again. And whining like the little bitch
you are, when someone mentions what you don't understand.

> SO BUTT OUT,

Despite your little caps-lock toddler tantrum, the fact remains that
you are not the moderator. You aren't allowed in moderated groups,
because you're not smart enough to follow the rules. So here you are
on Usenet, where nobody can moderate you. And you can't moderate
anyone. Sucks to be you.

> I DON'T KNOW WHAT I DID TO OFFEND YOU, AND YOU NEVER TOLD ME.

Despite your little caps-lock toddler tantrum, the fact remains that
you have been told this, repeatedly, and you just ignore it. Every day
is like groundhog day to you, you start with a clean slate of
dumb****ery.

(The rest of your "filthy diatribe" has been flushed.)

Please, put on your hockey helmet, and get back on the short bus. If
you want to know the answers to your little questions, just go back
and re-read all the explanations that have been given to you in the
years since you decided to ride your little hobby horse in this group.
All the answers are there, if you only had a brain. (Doo doot, doo
doodoodoodoo doot). YRATSAY? FCKWAGFA!

[Gray_Wolf]

On 10/14/2015 5:26 AM, Trevor wrote:
> On 14/10/2015 6:44 PM, John Williamson wrote:
>> On 14/10/2015 03:25, Trevor wrote:
>>> On 14/10/2015 7:23 AM, wrote:
>>>> If a lone sine wave sounds (and looks) different, then it is clipping.
>>>
>>> Not necessarily. *Squash* a sine wave without clipping, ie it has
>>> slightly flattened but still rounded tops, and it will sound different
>>> due to the extra harmonics added. Only when it has flat tops can it
>>> accurately be said to be clipped, by definition. Until then it is just
>>> compressed or soft limited.
>>>
>> Some amplification stages don't clip the top of a waveform dead flat
>> when overdriven.
>
> Not until you overdrive them a bit further anyway. everything in a finite world
> will clip eventually.


I don't always overdrive but when I do I use a Schmitt trigger. ;-)

[None]

"Randy Yates" > wrote in message
A transformation T is linear iff
>
> T(a*x1 + b*x2) = a*T(x1) + b*T(x2)
>
> So let's consider a very simple limiter that does this:
>
> L(x) = x, |x| < 1
> L(x) = sgn(x) * 1, |x| >= 1.
>
> Is this linear by the definition above? Nope. Here's
> a simple counterexample.
>
> Let x1 = 0.75, x2 = 0.75, a = 1, and b = 1. Then
>
>
> a*T(x1) + b*T(x2) = 1 * 0.75 + 1 * 0.75
> = 1.5
>
> but
>
> T(a*x1 + b*x2) = T(1.5) = 1.
>
> Not linear.

Signals and Systems 101. Nothing better than to the heart of the
issue.

However, the debate will rage on, fueled by opinion, lore, ignorance,
sloppy terminology, and of course, hobbyhorse dumb****ery.

[Scott Dorsey]

In article >, Trevor > wrote:
>On 14/10/2015 2:02 AM, Scott Dorsey wrote:
>> In article >, Trevor > wrote:
>>>> Many samples?
>>>
>>> At least 3 or more. (genuine square waves excluded) And as I have said,
>>> some CD's have HUNDREDS of consecutive samples at maximum level, and
>>> often many similar groups in one song. IF there are only 3 in fact I'm
>>> not worried in the slightest, when there are hundreds I am. That is far
>>> more common with pop CD's these days than many people seem to think,
>>> simply because they never look.
>>
>> Well, that's the degenerate case. Those CDs are clearly clipped. But I'm
>> talking about the borderline cases, because that's where it gets interesting.
>
>Few borderline cases in the pop world any more.

I don't live in the pop world. I'm an engineer. I live in the math world.
--scott

--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Scott Dorsey]

In article >, Trevor > wrote:
>
>> But you do appear to be claiming that all limiting is 'clipping',
>> which is crap.
>
>Dunno where how you managed to extrapolate that out of anything I've
>said! In fact I have been VERY clear to spell it out even if you can't
>read or understand. Let me try once more for the slow learners, Limiting
>CAN become clipping when the knee (if there is one) is past.
>Not all limiting causes clipping, but it certainly CAN!

If you haven't passed the knee, you aren't limiting. Below the knee, the
transfer function is 1:1. Above the knee, it's not.
--scott

--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[[email protected]]

On Wednesday, October 14, 2015 at 8:55:19 AM UTC-4, None wrote:
> "Randy Yates" > wrote in message
> A transformation T is linear iff
> >
> > T(a*x1 + b*x2) = a*T(x1) + b*T(x2)
> >
> > So let's consider a very simple limiter that does this:
> >
> > L(x) = x, |x| < 1
> > L(x) = sgn(x) * 1, |x| >= 1.
> >
> > Is this linear by the definition above? Nope. Here's
> > a simple counterexample.
> >
>

Hi Randy,
agreed, except that what is described above I would call a clipper, not a limiter...

If it alters the envelope AND the waveform, its a clipper.

If it alters only the envelope and NOT the waveform, its a limiter, compressor or AGC.

Mark

[Luxey]

Limiting is not more than special case of compression ie compression with
infinite ratio. All the attack, release, envelope, waveform ... talk is an
unnecessarry waste of energy.

Also, IMO, signal may be clipped, but clipping is the gear, pushed to work out of
speced range of operaation. The result is seen as charachteristic distortion
of the waveform, but that is just a sign the gear was pushed into clipping, so it
produced such a waveform which we've conveniently named "clipped" after the
clipping gear that produced it.

[Randy Yates[_2_]]

writes:

> On Wednesday, October 14, 2015 at 8:55:19 AM UTC-4, None wrote:
>> "Randy Yates" > wrote in message
>> A transformation T is linear iff
>> >
>> > T(a*x1 + b*x2) = a*T(x1) + b*T(x2)
>> >
>> > So let's consider a very simple limiter that does this:
>> >
>> > L(x) = x, |x| < 1
>> > L(x) = sgn(x) * 1, |x| >= 1.
>> >
>> > Is this linear by the definition above? Nope. Here's
>> > a simple counterexample.
>> >
>>
>
> Hi Randy,
> agreed, except that what is described above I would call a clipper, not a limiter...
>
> If it alters the envelope AND the waveform, its a clipper.
>
> If it alters only the envelope and NOT the waveform, its a limiter, compressor or AGC.
>
> Mark

Hi Mark,

Define envelope. Define waveform.
--
Randy Yates
Digital Signal Labs
http://www.digitalsignallabs.com

[Randy Yates[_2_]]

Randy Yates > writes:

> writes:
>
>> On Wednesday, October 14, 2015 at 8:55:19 AM UTC-4, None wrote:
>>> "Randy Yates" > wrote in message
>>> A transformation T is linear iff
>>> >
>>> > T(a*x1 + b*x2) = a*T(x1) + b*T(x2)
>>> >
>>> > So let's consider a very simple limiter that does this:
>>> >
>>> > L(x) = x, |x| < 1
>>> > L(x) = sgn(x) * 1, |x| >= 1.
>>> >
>>> > Is this linear by the definition above? Nope. Here's
>>> > a simple counterexample.
>>> >
>>>
>>
>> Hi Randy,
>> agreed, except that what is described above I would call a clipper, not a limiter...
>>
>> If it alters the envelope AND the waveform, its a clipper.
>>
>> If it alters only the envelope and NOT the waveform, its a limiter, compressor or AGC.
>>
>> Mark
>
> Hi Mark,
>
> Define envelope. Define waveform.

The bottom line, in my opinion, is this: If the system is not linear,
then you will introduce frequencies that weren't in the original input

I also do not know what you mean by the terms "envelope" and
"waveform" in this context.
--
Randy Yates
Digital Signal Labs
http://www.digitalsignallabs.com

[Luxey]

Limiting is not more than special case of compression ie compression with
infinite ratio. All the attack, release, envelope, waveform ... talk is an
unnecessarry waste of energy.

Also, IMO, signal may be clipped, but clipping is the gear, pushed to work out of
speced range of operation. The result is seen as charachteristic distortion
of the waveform, but that is just a sign the gear was pushed into "clipping",
so it produced such a waveform, with clipped tops, so we've conveniently named process after the looks of it. What I want to say, clipped waveform is not a
problem unless produced by clipping gear. As I already said in my previous post.

[Luxey]

�реда, 14. октобар 2015. 17.38.38 UTC+2, Randy Yates је напи�ао/ла:
> I also do not know what you mean by the terms "envelope" and
> "waveform" in this context.

Mike Rivers introduced those into discussion, Mark is merely building on it..

[Mike Rivers[_2_]]

On 10/14/2015 11:38 AM, Randy Yates wrote:
> I also do not know what you mean by the terms "envelope" and
> "waveform" in this context.

Envelope is a line drawn through the peaks of every cycle over whatever
period of time you want to look at. In the context that y'all are
arguing about here, it would likely be a whole song. If it's darn near a
straight line with no squiggles, that's what we call "toothpaste tube"
limiting. The amplitude value may not be full scale but it might as well
be - it won't sound any worse than if the envelope is flat-topped at -6
dBFS, but it'll be louder.

The waveform is what you see when you blow up the display so that you
can see individual cycles. You can see the peak level of each cycle, and
whatever goes up, comes down. It's the peaks of the waveform that you
connect to get the envelope.

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

[Frank Stearns]

Mike Rivers > writes:

>arguing about here, it would likely be a whole song. If it's darn near a
>straight line with no squiggles, that's what we call "toothpaste tube"
>limiting. The amplitude value may not be full scale but it might as well

Must be a regional thing. Out west we call it "cinderblock" limiting. Picture a raw
cinderblock in profile, and then a bunch of them lined up in a row (the waveforms of
the songs on your favorite mashed album displayed in your DAW, for example).

Hard, rough, gray, brittle, unmoving. "Toothpaste" sounds too benign. <w>

Frank
Mobile Audio

--

[[email protected]]

On Wednesday, October 14, 2015 at 11:34:52 AM UTC-4, Randy Yates wrote:
> writes:
>
> > On Wednesday, October 14, 2015 at 8:55:19 AM UTC-4, None wrote:
> >> "Randy Yates" > wrote in message
> >> A transformation T is linear iff
> >> >
> >> > T(a*x1 + b*x2) = a*T(x1) + b*T(x2)
> >> >
> >> > So let's consider a very simple limiter that does this:
> >> >
> >> > L(x) = x, |x| < 1
> >> > L(x) = sgn(x) * 1, |x| >= 1.
> >> >
> >> > Is this linear by the definition above? Nope. Here's
> >> > a simple counterexample.
> >> >
> >>
> >
> > Hi Randy,
> > agreed, except that what is described above I would call a clipper, not a limiter...
> >
> > If it alters the envelope AND the waveform, its a clipper.
> >
> > If it alters only the envelope and NOT the waveform, its a limiter, compressor or AGC.
> >
> > Mark
>
> Hi Mark,
>
> Define envelope. Define waveform.
> --
> Randy Yates
> Digital Signal Labs
> http://www.digitalsignallabs.com


see https://en.wikipedia.org/wiki/Envelope_(waves)

Blue is the waveform

red and green are the envelopes which for audio work are almost always the inverse of each other

or see Richard Lyons's book page 366 in the 2nd edition

he defines it mathematically as the abs value of the complex analytical signal

which = sqrt of real part squared + imaginary part squared

[Scott Dorsey]

Luxey > wrote:
>Limiting is not more than special case of compression ie compression with
>infinite ratio. All the attack, release, envelope, waveform ... talk is an
>unnecessarry waste of energy.

This is true. But clipping is just a special case of limiting if you look
at it that way... with a much more discontinuous knee.

>Also, IMO, signal may be clipped, but clipping is the gear, pushed to work out of
>speced range of operaation. The result is seen as charachteristic distortion
>of the waveform, but that is just a sign the gear was pushed into clipping, so it
>produced such a waveform which we've conveniently named "clipped" after the
>clipping gear that produced it.

The thing is, even soft knee limiting can produce some amount of that
distortion. The question is how much makes it clipping? That puts us back
to the definition of "if you can hear it working it's clipping and not limiting"
which is good but not very rigorous.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Scott Dorsey]

Luxey > wrote:
>=D1=81=D1=80=D0=B5=D0=B4=D0=B0, 14. =D0=BE=D0=BA=D1=82=D0=BE=D0=B1=D0=B0=D1=
>=80 2015. 17.38.38 UTC+2, Randy Yates =D1=98=D0=B5 =D0=BD=D0=B0=D0=BF=D0=B8=
>=D1=81=D0=B0=D0=BE/=D0=BB=D0=B0:
>> I also do not know what you mean by the terms "envelope" and=20
>> "waveform" in this context.
>
>Mike Rivers introduced those into discussion, Mark is merely building on it=
>.

It's true though that you can't alter one without altering the other in some
way. They are both different ways of _thinking about_ an arbitrary function.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[geoff]

On 15/10/2015 5:19 a.m., Mike Rivers wrote:
> On 10/14/2015 11:38 AM, Randy Yates wrote:
>> I also do not know what you mean by the terms "envelope" and
>> "waveform" in this context.
>
> Envelope is a line drawn through the peaks of every cycle over
> whatever period of time you want to look at. In the context that y'all
> are arguing about here, it would likely be a whole song. If it's darn
> near a straight line with no squiggles, that's what we call
> "toothpaste tube" limiting. The amplitude value may not be full scale
> but it might as well be - it won't sound any worse than if the
> envelope is flat-topped at -6 dBFS, but it'll be louder.
>
> The waveform is what you see when you blow up the display so that you
> can see individual cycles. You can see the peak level of each cycle,
> and whatever goes up, comes down. It's the peaks of the waveform that
> you connect to get the envelope.
>

Or 'waveform' is the shape of individual cycles, and 'envelope' is the
outline of a bunch of cycles over time (as when zoomed out).

geoff

[geoff]

On 15/10/2015 7:08 a.m., Frank Stearns wrote:
> Mike Rivers > writes:
>
>> arguing about here, it would likely be a whole song. If it's darn near a
>> straight line with no squiggles, that's what we call "toothpaste tube"
>> limiting. The amplitude value may not be full scale but it might as well
> Must be a regional thing. Out west we call it "cinderblock" limiting. Picture a raw
> cinderblock in profile, and then a bunch of them lined up in a row (the waveforms of
> the songs on your favorite mashed album displayed in your DAW, for example).
>
> Hard, rough, gray, brittle, unmoving. "Toothpaste" sounds too benign. <w>
>

Much of the world has no idea what a cinderblock is or looks like. I
guess 'toothpaste tube or sausage may fail by that logic too though ...

;-)

geoff

[geoff]

On 15/10/2015 9:22 a.m., geoff wrote:
>
>>
>
> Or 'waveform' is the shape of individual cycles, and 'envelope' is
> the outline of a bunch of cycles over time (as when zoomed out).
>
> geoff


.... but I'm sure Trev will find a way to lambast me for my apparent
ignorance on this subject too ;-)

geoff

[Scott Dorsey]

geoff > wrote:
>
>Or 'waveform' is the shape of individual cycles, and 'envelope' is the
>outline of a bunch of cycles over time (as when zoomed out).

If you take the waveform and take the absolute value and low-pass it, you
get the envelope. But how far do you have to low-pass it for it to be the
envelope? Depends.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[[email protected]]

Randy Yates wrote: "- show quoted text -
Hi Mark,

Define envelope. Define waveform.
--
Randy Yates
Digital Signal Labs
http://www.digitalsignallabs.com "

THANK YOU VERY MUCH for asking this!

Because frankly, NOBODY has satisfactorily
distinguished the two!

[geoff]

On 15/10/2015 10:09 a.m., Scott Dorsey wrote:
> geoff > wrote:
>> Or 'waveform' is the shape of individual cycles, and 'envelope' is the
>> outline of a bunch of cycles over time (as when zoomed out).
> If you take the waveform and take the absolute value and low-pass it, you
> get the envelope. But how far do you have to low-pass it for it to be the
> envelope? Depends.
> --scott

Depends on the context of what is being described. Envelope of a
half-dozen cycles, or a whole song, or album.

The joys of 'visual low pass filtering' - a simple flick of the
mouse-wheel to achieve each scenario !

geoff

[geoff]

On 15/10/2015 10:29 a.m., wrote:
> Randy Yates wrote: "- show quoted text -
> Hi Mark,
>
> Define envelope. Define waveform.

Clearly explained to you many many times, including with pictures.
Apparently to no avail ....

geoff

[None]

< thekma @ gmail.com > wrote in message news:7335a01f-be1a-4e3e-9659
> Because frankly, NOBODY has satisfactorily distinguished the two!

Dorsey just posted a concise and useful definition (below). Perhaps
you were too busy ranting to notice.

"If you take the waveform and take the absolute value and low-pass it,
you get the envelope. But how far do you have to low-pass it for it
to be the envelope? Depends."

[Mike Rivers[_2_]]

On 10/14/2015 4:24 PM, geoff wrote:
> Much of the world has no idea what a cinderblock is or looks like. I
> guess 'toothpaste tube or sausage may fail by that logic too though ...

I don't know about toothpaste, but practically every nationality has
sausage of some sort, though the word might need translation.

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

[[email protected]]

Scott Dorsey:

An analogy with a city skyline: The individual buildings are
the waveform, while an imaginary line from the top of each
building to the next would be the envelope. But remember:
the 'envelope' is just an outline.

[Mike Rivers[_2_]]

On 10/14/2015 6:09 PM, None wrote:
> Dorsey just posted a concise and useful definition (below). Perhaps you
> were too busy ranting to notice.
>
> "If you take the waveform and take the absolute value and low-pass it,
> you get the envelope. But how far do you have to low-pass it for it to
> be the envelope? Depends."

That's not a definition, it's a mathematical process. The terms here
have been explained several times. Anyone who cares and doesn't
understand what they mean must not understand something more fundamental.

The pictures are good. They explain all you need to know. What you do
with that information, however, depends.

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

[[email protected]]

No^#%.>\!{:

I said ZIP IT unless something other than ****
is going to flow from your keyboard!

[geoff]

On 15/10/2015 11:40 a.m., wrote:
> Scott Dorsey:
>
> An analogy with a city skyline: The individual buildings are
> the waveform, while an imaginary line from the top of each
> building to the next would be the envelope. But remember:
> the 'envelope' is just an outline.


Yeah, the visual analogy is far easier to understand than "low-pass
filtering", especially for the non-technical. But still, it has been
explained already so many times ....

geoff

[None]

< krissie krybaby @ dumb****sRthekma.shortbus.edu > wrote in message
...
> No^#%.>\!{:
>
> I said ZIP IT unless something other than ****
> is going to flow from your keyboard!

I see that you're still ignoring the hell out of me. Have you noticed
that when you tell me to shut up, it doesn't work? No, just because
it's happened hundreds of times, you haven't noticed, because you're
an idiot. Not a moderator; an idiot. IHTSIYC. WAFTH.

[None]

< krissie krybaby @ dumb****sRthekma.shortbus.edu > wrote in message
...
> No^#%.>\!{:
>
> I said ZIP IT unless something other than ****
> is going to flow from your keyboard!

I see that you're still ignoring the hell out of me. Have you noticed
that when you tell me to shut up, it doesn't work? No, just because
it's happened hundreds of times, you haven't noticed, because you're
an idiot. Not a moderator; an idiot. IHTSIYC. WAFTH.

[[email protected]]

Mike Rivers wrote: "understand what they mean must not understand something more fundamental.

The pictures are good. They explain all you need to know. What you do
with that information, however, depends. "

Word games played all too often on
here....

Hey Mike: you good friends with Donald
Rumsfeld? 'Cause you sure sound like
him above! lol

[geoff]

On 15/10/2015 12:23 p.m., wrote:
> geoff wrote: "- show quoted text -
> Or 'waveform' is the shape of individual cycles, and 'envelope' is the
> outline of a bunch of cycles over time (as when zoomed out). "
>
>
> This is all just nitpicking, semantics,
> and the word games typical of
> being played within this group.
>
> I call a duck a duck, and you guys
> can just DEAL. sheesh!



Um no, not semantics ! In audio iIf you can discern individual cycles it
is a waveform. If all the cycles are bunched together , then the overall
outline (which is all that you can discern) is the envelope.

You asked. Many times. And have been clearly and concisely answered.
Many times. It really isn't rocket surgery.

And now it seems you actually prefer to not know. So why ask ? Which is
why many assume that you are a troll rather than simply stupid.

geoff

[Trevor]

On 15/10/2015 12:38 AM, Scott Dorsey wrote:
> In article >, Trevor > wrote:
>>
>>> But you do appear to be claiming that all limiting is 'clipping',
>>> which is crap.
>>
>> Dunno where how you managed to extrapolate that out of anything I've
>> said! In fact I have been VERY clear to spell it out even if you can't
>> read or understand. Let me try once more for the slow learners, Limiting
>> CAN become clipping when the knee (if there is one) is past.
>> Not all limiting causes clipping, but it certainly CAN!
>
> If you haven't passed the knee, you aren't limiting.

You mean if you haven't *reached* the knee, then of course.


> Below the knee, the transfer function is 1:1. Above the knee, it's not.

Well duh.

Trevor.

[Trevor]

On 15/10/2015 2:31 AM, Luxey wrote:
> Limiting is not more than special case of compression ie compression with
> infinite ratio. All the attack, release, envelope, waveform ... talk is an
> unnecessarry waste of energy.
>
> Also, IMO, signal may be clipped, but clipping is the gear, pushed to work out of
> speced range of operaation. The result is seen as charachteristic distortion
> of the waveform, but that is just a sign the gear was pushed into clipping, so it
> produced such a waveform which we've conveniently named "clipped" after the
> clipping gear that produced it.

Agreed, if the waveform is clipped, I can't see the point in arguing HOW
it was done. Other than the to the person who did it perhaps.

Trevor.

[[email protected]]

geoff wrote: "- show quoted text - Which is why many assume that you are a
troll rather than simply stupid.

geoff "

Actually geoff: neither! And you know about the first part
of assuming. ;)

I disdain nit-picking, and prefer to keep the conversation concise
and on a roll. If I make a statement:

"You know, the waveform of that latest George Benson tune
is surprisingly spiky for something modern"

I don't expect a twenty-reply theological debate over
"envelope vs waveform" to hijack a good thread.

I know a lot more about this audio thing we love than
most of you care to give me credit for. But I use simple
straight-forward terms to convey or absorb concepts,
rather than delve into semantics and word-play. This is
rec.audio.pro - not the Pentagon. ;)

[[email protected]]

geoff wrote: "why ask?"

Because you guys got all nit-picky over
what's a waveform and what's not, instead
of just sticking to the topic of whatever
thread the splitting hairs took place in.

Kind of like "used car-NO, it's a "pre-owned
car". Totally gay(not the homosexual kind
either!).

[geoff]

On 15/10/2015 4:17 p.m., wrote:
> geoff wrote: "why ask?"
>
> Because you guys got all nit-picky over
> what's a waveform and what's not, instead
> of just sticking to the topic of whatever
> thread the splitting hairs took place in.
>
> Kind of like "used car-NO, it's a "pre-owned
> car". Totally gay(not the homosexual kind
> either!).
>


No. It's not nit-picking at all. Say what you mean instead of what you
steadfastly refuse to understand, and people will understand you.

It's more like you calling a single skyscraper a skyline (to use Scott's
analogy).

Really, it's not that difficult.

geoff

[Trevor]

On 15/10/2015 4:31 PM, geoff wrote:
> On 15/10/2015 4:17 p.m., wrote:
>> geoff wrote: "why ask?"
>>
>> Because you guys got all nit-picky over
>> what's a waveform and what's not, instead
>> of just sticking to the topic of whatever
>> thread the splitting hairs took place in.
>>
>> Kind of like "used car-NO, it's a "pre-owned
>> car". Totally gay(not the homosexual kind
>> either!).
>
> No. It's not nit-picking at all. Say what you mean instead of what you
> steadfastly refuse to understand, and people will understand you.
>
> It's more like you calling a single skyscraper a skyline (to use Scott's
> analogy).

And yet most people don't draw an imaginary line between building tops
and call that the skyline.


> Really, it's not that difficult.

That only one is real and one is imaginary still seems to be beyond many
though. A waveform is still a waveform whether you zoom in or not,
whether you can see detail or not. But the "envelope" is still a useful
*concept* IF you understand it.

Trevor.

[geoff]

On 15/10/2015 8:24 p.m., Trevor wrote:

>
> That only one is real and one is imaginary still seems to be beyond many
> though. A waveform is still a waveform whether you zoom in or not,
> whether you can see detail or not. But the "envelope" is still a useful
> *concept* IF you understand it.


Thought you'd find a nit to pick !

If you zoom out to the point you cannot see individual cycles, then all
you can see is the envelope. Not imaginary - it's the outline shape that
you see.

How about the term "waveform envelope" - cover enough bases for you ?


geoff

[[email protected]]

geoff wrote: "How about the term "waveform envelope" - cover enough bases for you ?"

Uhh geoff, the ****ing Titanic is completely
bow-down; are you still trying to discern
the BRAND OF PAINT used on that lifeboat,
or are you jumping in?!

sheez!..

[Mike Rivers[_2_]]

On 10/14/2015 9:34 PM, geoff wrote:
> Um no, not semantics ! In audio iIf you can discern individual cycles it
> is a waveform. If all the cycles are bunched together , then the overall
> outline (which is all that you can discern) is the envelope.

You don't have to be able to see a waveform in order to see it. And when
you "zoom out" the waveform doesn't disappear, to be eclipsed by the
envelope. But understand that these are just names for graphic
representations of a changing voltage.

If a tree falls in the forest and there's nobody around to hear it, it
still makes a sound. If you zoom out of a graphic representation of a
signal so that you can see an envelope, the waveform is still there.

Is this semantics? Or philosophy?


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

[[email protected]]

Mike Rivers wrote: "You don't have to be able to see a waveform in order to see it."

Pure https://www.google.com/search?q=donald+rumsfeld+dumb&client=safari&hl=en&prmd=ivn&source=lnms&tbm=isch&sa=X&ved=0CAgQ_AUoAWoVChMI3Om1iqXEyAIVSRs-Ch3OVAZp&biw=480&bih=208&dpr=2#imgrc=FiCJ1_HVsondqM%3A !

Now how about those 'known-unknowns', Mikey? ;)

[[email protected]]

"Is this semantics? Or philosophy? "

If I zoom in on a duck's feathers or
stand twenty feet away it's still a DUCK
to me!

[None]

> wrote in message
news:da6b6322-78d7-4275-a419-
> I know a lot more about this audio thing we love than most of you
> care to give me credit for.

Then why do yo go to such trouble to prove that you're a dumb ****?

[None]

> wrote in message
...
> "Is this semantics? Or philosophy? "
>
> If I zoom in on a duck's feathers or
> stand twenty feet away it's still a DUCK
> to me!

Li'l Krissie is having a meltdown again.

[[email protected]]

geoff wrote: "Say what you mean"

Admittedly I have trouble putting words
together, even if I DO get the concept.

"instead of what you
steadfastly refuse to understand, "

ARE YOU INSIDE MY HEAD or something,
wiseguy?! Don't try to ASSume you know
my motives, GEOFF. As if I would purposely
insist that grass is pink when it is plainly green.


Nobody, except maybe a politician, might
DELIBERATELY try not to understand
anything!

[Mike Rivers[_2_]]

On 10/15/2015 6:16 AM, wrote:
> geoff wrote: "How about the term "waveform envelope" - cover enough bases for you ?"

I would consider that the proper term for what's been shortened to
"envelope" by the lazy duffers.

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

[Mike Rivers[_2_]]

On 10/15/2015 6:42 AM, wrote:
> Mike Rivers wrote: "You don't have to be able to see a waveform in order to see it."

Geez, did I write that? I must have needed another cup of coffee and
didn't realize it. What I really meant was that you don't have to be
able to see a waveform in order for it to be there.

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

[Mike Rivers[_2_]]

On 10/15/2015 6:48 AM, wrote:
> "Is this semantics? Or philosophy?"
>
> If I zoom in on a duck's feathers or
> stand twenty feet away it's still a DUCK
> to me!

But without seeing the duck from 20 or 5 feet away, but only saw a
couple of feathers, would you know it was a duck?

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

[None]

< thick-headed-toddler-whines @gmail.com> wrote in message
...
> geoff wrote: "Say what you mean"
>
> Admittedly I have trouble putting words
> together, even if I DO get the concept.
>
> "instead of what you
> steadfastly refuse to understand, "
>
> ARE YOU INSIDE MY HEAD or something,
> wiseguy?!

Your refusal to understand is on display in your posts. Nobody has to
go inside your head. You keep missing that essential point. It's
another thing you refuse to understand. Anyone who reads this
newsgroup knows that you prefer endless childish arguments to
understanding. This recent string of toddler-tantrums from you is yet
another example. Of course, you don't understand that, because you
refuse to understand.

> Don't try to ASSume you know
> my motives, GEOFF. As if I would purposely
> insist that grass is pink when it is plainly green.

But when anybody points out that the grass is sometimes brown, your
head explodes in rage, and you start ranting and drooling and issuing
personal attacks and filthy diatribes against whoever pointed out the
simple fact that the grass is sometimes brown. The world (including
audio) isn't simple enough for a simpleton like you. What a tragedy.

> Nobody, except maybe a politician, might
> DELIBERATELY try not to understand
> anything!

So you're a politician now? A dumb**** politician in a dumbuck
politician's hockey helmet, on the politiclans' short campaign bus?
Boy, it really sucks to be you.

FCKISSFDSWAHHTHRAOTP. AHETHHS. STBY. Right, Krissie?

[Scott Dorsey]

> wrote:
>geoff wrote: "- show quoted text -
>Or 'waveform' is the shape of individual cycles, and 'envelope' is the
>outline of a bunch of cycles over time (as when zoomed out). "
>
>
>This is all just nitpicking, semantics,
>and the word games typical of
>being played within this group.

No, it's important. Because it's important to know if you're looking at
the actual waveform sample by sample, or just an averaged envelope.

A lot of things like crossover distortion isn't visible at all on the
envelope but are very visible on an actual waveform display.

>I call a duck a duck, and you guys
>can just DEAL. sheesh!

Distinctions like the difference between peak and average levels, between
reference levels and full scale levels, and between waveforms and envelopes
are absolutely critical to understand dynamics processing.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Scott Dorsey]

geoff > wrote:
>On 15/10/2015 4:17 p.m., wrote:
>> geoff wrote: "why ask?"
>>
>> Because you guys got all nit-picky over
>> what's a waveform and what's not, instead
>> of just sticking to the topic of whatever
>> thread the splitting hairs took place in.
>>
>> Kind of like "used car-NO, it's a "pre-owned
>> car". Totally gay(not the homosexual kind
>> either!).
>>
>
>
>No. It's not nit-picking at all. Say what you mean instead of what you
>steadfastly refuse to understand, and people will understand you.
>
>It's more like you calling a single skyscraper a skyline (to use Scott's
>analogy).

I actually didn't use that analogy... it was attributed to me but I can't
take credit. It's a good analogy, though, and I like it.
--scott

--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Scott Dorsey]

In article >,
> wrote:
>"Is this semantics? Or philosophy? "
>
>If I zoom in on a duck's feathers or
>stand twenty feet away it's still a DUCK
>to me!

If the waveform is a duck, the envelope is a drawing of a duck.
--scott

--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Peter Larsen[_3_]]

On 12-10-2015 04:03, JackA wrote:

> On Friday, October 9, 2015 at 9:47:34 PM UTC-4, wrote:
>> Whoever decreed that ALL audio tracks MUST peak within
>> 1/1,000,00dB of full scale must be on something strong!!

> I was told, no clipping! Otherwise, Loudness Wars would have never existed.

Hogwash, the best way to get nice bricks is to clip all bands of a
multiband compressor.

Take some time to actually learn the tools you rant about.

> Jack

- Peter Larsen

[Frank Stearns]

(Scott Dorsey) writes:

>In article >,
> > wrote:
>>"Is this semantics? Or philosophy? "
>>
>>If I zoom in on a duck's feathers or
>>stand twenty feet away it's still a DUCK
>>to me!

>If the waveform is a duck, the envelope is a drawing of a duck.

And if heavily processed as per some of today's fashion, what was once music now
sounds just like a continuously quacking duck, only louder.

Sorry, couldn't resist.

Frank
Mobile Audio

--

[geoff]

On 15/10/2015 11:32 p.m., Mike Rivers wrote:
> On 10/14/2015 9:34 PM, geoff wrote:
>> Um no, not semantics ! In audio iIf you can discern individual cycles it
>> is a waveform. If all the cycles are bunched together , then the overall
>> outline (which is all that you can discern) is the envelope.
>
> You don't have to be able to see a waveform in order to see it. And
> when you "zoom out" the waveform doesn't disappear, to be eclipsed by
> the envelope. But understand that these are just names for graphic
> representations of a changing voltage.
>
> If a tree falls in the forest and there's nobody around to hear it, it
> still makes a sound. If you zoom out of a graphic representation of a
> signal so that you can see an envelope, the waveform is still there.
>
> Is this semantics? Or philosophy?
>
>

Semantics. Zoom out of the forest. In that forest, you can no longer
discern the tree, but you can see the forest. The forest is not 'a tree'.

geoff

[geoff]

On 15/10/2015 11:48 p.m., wrote:
> "Is this semantics? Or philosophy?"
>
> If I zoom in on a duck's feathers or
> stand twenty feet away it's still a DUCK
> to me!
If you zoom in so all you can see is a feather, you are looking at a
feather, not 'a duck'.

Yes, it is a duck's feather, but if you had not seen the zoomed out
version, you wouldn't know it was a duck's feather.

geoff

[geoff]

On 16/10/2015 4:44 a.m., Frank Stearns wrote:
> (Scott Dorsey) writes:
>
> And if heavily processed as per some of today's fashion, what was once music now
> sounds just like a continuously quacking duck, only louder.
>
> Sorry, couldn't resist.
>
> Frank
> Mobile Audio
>


Lump of paté ?

geoff

[[email protected]]

geoff wrote: "- show quoted text -
Semantics. Zoom out of the forest. In that forestZZZZHHHHHHIPPPPP!!!
(needle being dragged across record)

geoff "


I told you - no "semantics", only ENGLISH. So by your forest
analogy, one tree = one up & down cycle = one waveform.

What you say might be true, correct, but when I Google up
"audio waveform", this is what I get THOUSANDS of as
results: https://s-media-cache-ak0.pinimg.com/736x/e0/ab/53/e0ab5380d984bdeb9101a2234454be16.jpg

So you go on ahead Geoff and start correcting all these
thousands of users, for the sake of the survival of the
audio universe!!!

And to be fair:
"audio envelope" returns a sizable majority of this: http://www-sipl.technion.ac.il/new/Teaching/Projects/Projects-Pages/MPEG-7-Audio-Descriptors/pic10.png
as hits. Very different from the majority of hits from "waveform.

Again, a colossal waste of all our time here, seriously!

[geoff]

On 16/10/2015 11:37 a.m., wrote:
>
> Again, a colossal waste of all our time here, seriously!
>

You are.

Enough of my time. Bye.

geoff

[Ralph Barone[_2_]]

Scott Dorsey > wrote:
> In article >,
> > wrote:
>> "Is this semantics? Or philosophy? "
>>
>> If I zoom in on a duck's feathers or
>> stand twenty feet away it's still a DUCK
>> to me!
>
> If the waveform is a duck, the envelope is a drawing of a duck.
> --scott
>

If the waveform is a duck, the envelope is the shadow of a duck.

[[email protected]]

Ralph Barone wrote: "If the waveform is a duck, the envelope is the
shadow of a duck. "


https://abiadams.files.wordpress.com/2011/12/head-in-hands-e1298825206674.jpg

[david gourley[_2_]]

said...news:c7ab072a-e2ae-4ad1-beab-6e76a5f9bbe6
@googlegroups.com:

> Ralph Barone wrote: "If the waveform is a duck, the envelope is the
> shadow of a duck. "
>
>
> https://abiadams.files.wordpress.com/2011/12/head-in-hands-
e1298825206674.jpg

Presumably the look of 'still not getting it?'

david

[Les Cargill[_4_]]

Ralph Barone wrote:
> Scott Dorsey > wrote:
>> In article >,
>> > wrote:
>>> "Is this semantics? Or philosophy?"
>>>
>>> If I zoom in on a duck's feathers or
>>> stand twenty feet away it's still a DUCK
>>> to me!
>>
>> If the waveform is a duck, the envelope is a drawing of a duck.
>> --scott
>>
>
> If the waveform is a duck, the envelope is the shadow of a duck.
>


I shall use my laaargest scaaales....

--
Les Cargill

[None]

< thickskullocks @ gmail.com> wrote in message
...
> geoff wrote: "- show quoted text -
> Semantics.
> I told you - no "semantics", only ENGLISH.

1. You're not the moderator. You don't get to tell others what to
post. It makes no difference what you told anybody. Have you forgotten
that already, since the last time you were reminded? It was only this
morning.

2. Without semantics, English is useless. The only conclusion is that
you have no idea what at least one of those words means. You get
yourself in trouble when you use words without having any idea what
they mean. Like basically all words about audio.

> Again, a colossal waste of all our time here, seriously!

No ****, Dumb ****. And it's entirely your fault. You've polluted this
newsgroup so much, even you can't stand the stench. Maybe you should
go crap all over some other newsgroup, far away, YSFSJBSC.

[None]

"Ralph Barone" > wrote in message
news:1319924126.466653813.861076.address_is-invalid.invalid@shawnews...
> If the waveform is a duck, the envelope is the shadow of a duck.

If the waveform is a flat line, the envelope is a flat line. It's
thikskull's brain function.

[John Williamson]

On 15/10/2015 23:37, wrote:

> What you say might be true, correct, but when I Google up
> "audio waveform", this is what I get THOUSANDS of as
> results: https://s-media-cache-ak0.pinimg.com/736x/e0/ab/53/e0ab5380d984bdeb9101a2234454be16.jpg
>
> So you go on ahead Geoff and start correcting all these
> thousands of users, for the sake of the survival of the
> audio universe!!!
>
That is an excellent picture of an envelope, showing a hint of the
waveform of the sounds. Given sufficient detail on the original
screenshot, if you looked more closely, you could see the individual
waveforms within the envelope. At just the right zoom level, where the
waveforms are just becoming visible, defocus your eyes and you will see
the envelope, and you can switch between the two views at will.

> And to be fair:
> "audio envelope" returns a sizable majority of this: http://www-sipl.technion.ac.il/new/Teaching/Projects/Projects-Pages/MPEG-7-Audio-Descriptors/pic10.png
> as hits. Very different from the majority of hits from "waveform.
>
That is the graphic depiction of the output voltage from an Attack,
Decay, Sustain, Release (ADSR) generator as used on many synthesisers
(Such as the original Moog, or my little Casio pocket jobby) to control
the amplitude of the audio output from the waveform generator's Voltage
Controlled Amplifier when a key is struck, and bears no relationship to
the audio waveform.

> Again, a colossal waste of all our time here, seriously!
>
Only because you are not paying attention to what many very
knowledgeable people are trying to tell you.

--
Tciao for Now!

John.

[Trevor]

On 15/10/2015 7:28 PM, geoff wrote:
> On 15/10/2015 8:24 p.m., Trevor wrote:
>> That only one is real and one is imaginary still seems to be beyond many
>> though. A waveform is still a waveform whether you zoom in or not,
>> whether you can see detail or not. But the "envelope" is still a useful
>> *concept* IF you understand it.
>
>
> Thought you'd find a nit to pick !
>
> If you zoom out to the point you cannot see individual cycles, then all
> you can see is the envelope. Not imaginary - it's the outline shape that
> you see.

You still can't read what I wrote apparently :-(
Just because you choose not to see the actual detail doesn't make the
"envelope" any less of an imaginary construct!


> How about the term "waveform envelope" - cover enough bases for you ?

No, but imaginary waveform envelope would. However envelope alone is
fine for those who actually know what it means already.

Trevor.

[Trevor]

On 15/10/2015 11:35 PM, Mike Rivers wrote:
> On 10/15/2015 6:48 AM, wrote:
>> "Is this semantics? Or philosophy?"
>>
>> If I zoom in on a duck's feathers or
>> stand twenty feet away it's still a DUCK
>> to me!
>
> But without seeing the duck from 20 or 5 feet away, but only saw a
> couple of feathers, would you know it was a duck?

Would that matter to the duck? Oh wait, I guess it might if you were a
duck hunter! :-(

Trevor.

[Gray_Wolf]

On 10/15/2015 5:37 PM, wrote:
> geoff wrote: "- show quoted text -
> Semantics. Zoom out of the forest. In that forestZZZZHHHHHHIPPPPP!!!
> (needle being dragged across record)
>
> geoff "
>
>
> I told you - no "semantics", only ENGLISH. So by your forest
> analogy, one tree = one up & down cycle = one waveform.
>
> What you say might be true, correct, but when I Google up
> "audio waveform", this is what I get THOUSANDS of as
> results: https://s-media-cache-ak0.pinimg.com/736x/e0/ab/53/e0ab5380d984bdeb9101a2234454be16.jpg
>
> So you go on ahead Geoff and start correcting all these
> thousands of users, for the sake of the survival of the
> audio universe!!!
>
> And to be fair:
> "audio envelope" returns a sizable majority of this: http://www-sipl.technion.ac.il/new/Teaching/Projects/Projects-Pages/MPEG-7-Audio-Descriptors/pic10.png
> as hits. Very different from the majority of hits from "waveform.

Both of these are not true waveforms. They are both envelopes. One has been
smoothed and the other has not. It would be impossible to determine the
occurrence of cross-over distortion, low level parasitic oscillation and many
other things. For this you would need to see the detailed waveform. It's the
difference between observing the general form of a razor and close examination
of the quality of the cutting edge.

In any case why would your knowing the difference between wave form and envelope
be of any value to you?

[Mike Rivers[_2_]]

On 10/15/2015 10:04 PM, Ralph Barone wrote:
> If the waveform is a duck, the envelope is the shadow of a duck.

If the envelope is a duck, what the duck had for lunch is somewhere in
the waveform.

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

[None]

"Trevor" > wrote in message
...
>>> "Is this semantics? Or philosophy?"
>
> Would that matter to the duck? Oh wait, I guess it might if you were
> a duck hunter! :-(

It's wabbit season.

[[email protected]]

> >
> > Would that matter to the duck? Oh wait, I guess it might if you were
> > a duck hunter! :-(
>
> It's wabbit season.

Is that why my compressor has a ducking input?

And the duck bill for my ducking compressor came in an envelope?



Mark

[Scott Dorsey]

None > wrote:
>< thickskullocks @ gmail.com> wrote in message
...
>> geoff wrote: "- show quoted text -
>> Semantics.
>> I told you - no "semantics", only ENGLISH.
>
>1. You're not the moderator. You don't get to tell others what to
>post. It makes no difference what you told anybody. Have you forgotten
>that already, since the last time you were reminded? It was only this
>morning.

Part of the problem is that this is basically a mathematical definition
that people are arguing over. Verbal english descriptions are only going
to be approximate analogies; the actual definition is a numerical function.
--scott

--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Peter Larsen[_3_]]

On 16-10-2015 14:47, Scott Dorsey wrote:

> None > wrote:

>> < thickskullocks @ gmail.com> wrote in message
>> ...
>>> geoff wrote: "- show quoted text -
>>> Semantics.
>>> I told you - no "semantics", only ENGLISH.
>
>> 1. You're not the moderator. You don't get to tell others what to
>> post. It makes no difference what you told anybody. Have you forgotten
>> that already, since the last time you were reminded? It was only this
>> morning.

> Part of the problem is that this is basically a mathematical definition

We may be doomed. If they can't see a curve in space like ordinary
people when reading the formula, then they may perhaps never get it.

But at least we now know what it is that is preventing their
understanding. The only thing more dangerous that illiteracy, illmatheracy.

Kind regards

Peter Larsen

> that people are arguing over. Verbal english descriptions are only going
> to be approximate analogies; the actual definition is a numerical function.

> --scott
>

[[email protected]]

Peter Larsen: "ilmatheracy"

Or as I have read, "innumeracy". I cannot
perform even basic math, not because I didn't
want to, but simply because no teacher could
teach it to me.

Others have serious problems grasping basic
grammar. Still others, understanding cause
& effect.

We are all blessed with different talents,
Peter, and sometimes we have challenges
learning others. But no one deserves to
be bullied, ridiculed, or shut out because of
that.

How would a certain bully feel if it were turned
on THEM?

[Peter Larsen[_3_]]

On 16-10-2015 17:33, wrote:

> Peter Larsen: "ilmatheracy"

> Or as I have read, "innumeracy". I cannot
> perform even basic math, not because I didn't
> want to, but simply because no teacher could
> teach it to me.

> Others have serious problems grasping basic
> grammar. Still others, understanding cause
> & effect.

> We are all blessed with different talents,
> Peter, and sometimes we have challenges
> learning others.

Yes. I agree 100 percent. Also please note that I don't give a dam'
about who posts what, I comment on the content and I occasionally devote
quite some effort in getting things understood.

> But no one deserves to
> be bullied, ridiculed, or shut out because of
> that.

Yes, I agree 100 percent.

> How would a certain bully feel if it were turned
> on THEM?

Which is what is happening here since you want to wear the shoe. I
didn't put it on you. It was not you that was at the forefront of my
mind in typing.

Your dragging a dead horse around is just that towards the audio
professionals in this forum. You are very strong on average loudness
level opinions and not so strong on actual audio recording.

Kind regards

Peter Larsen

[[email protected]]

Peter Larsen wrote: "not so strong on actual audio recording."

Average loudness is not an opinion, it is fact that
finished product has gotten louder in the last 20
years. At least I have come to grips with what/who
is driving it.


But please elaborate on that second part - about
audio recording.

[John Williamson]

On 16/10/2015 19:14, wrote:
> Peter Larsen wrote: "not so strong on actual audio recording."
>
> Average loudness is not an opinion, it is fact that
> finished product has gotten louder in the last 20
> years. At least I have come to grips with what/who
> is driving it.
>
So you have finally realised that the loudness wars are driven by the
listeners, not by the engineers. Good.
>
> But please elaborate on that second part - about
> audio recording.
>
Just about every post you make demonstrates your total lack of
comprehension of the basics of audio recording as well as your ignorance
about the aims of and the techniques used in the modern audio production
process.

Your posts in this thread show your inability to tell the difference
between a waveform, the amplitude envelope of that waveform and a
graphical description of how the gain of an amplifier inside a
synthesiser varies, even after many repeated attempts to explain it to
you in simple terms.
--
Tciao for Now!

John.

[[email protected]]

John Williamson wrote: "So you have finally realised that the loudness wars are driven by the listeners, not by the engineers. Good. "

WRONG.

They are driven by the artists, producers,
and labels. If I ever met a listener who said "I wish
this song/record was louder", I'd grab them by the
arm and place their hand on the volume control.
(then I might offer to drive them to an eye doctor to
have their vision checked!)

As far as recording goes, I record my church worship
team every Sunday. I have an analog mixer and a
digital audio recorder, and I know exactly what the
meters on each are telling me. At the same time
I make sure each participant has enough monitor
to hear themself, and that I'm not blasting out the
house.

They trust me enough to know what I'm doing.

[John Williamson]

On 16/10/2015 20:05, wrote:
> John Williamson wrote: "So you have finally realised that the loudness wars are driven by the listeners, not by the engineers. Good. "
>
> WRONG.
>
> They are driven by the artists, producers,
> and labels.

AKA the people who listen to and pay for the recordings.

If I ever met a listener who said "I wish
> this song/record was louder", I'd grab them by the
> arm and place their hand on the volume control.
> (then I might offer to drive them to an eye doctor to
> have their vision checked!)
>
If you speak to someone who listens to recordings in a quite location
such as their living room or a dedicated living room, I'd agree with
you. However, most popular music has, for a few decades now, been
"listened to" in places with a high background noise level such as a car
or a location full of noisy people dancing. In these places, it is
desirable to decrease the ratio between the quietest and loudest parts,
so that the quiet parts are audible above the background noise, and the
loud parts do not cause damage to either the listeners' hearing or the
reproduction equipment. It is not practicable to force the end user to
install a compressor on their player, so the compression has to be done
elsewhere in the chain, and the easiest place to do it is during the
mastering process, so that the sound is not drastically altered by
compression added at the transmitter. This will inevitably have the
effect of increasing the average loudness of the music.

> As far as recording goes, I record my church worship
> team every Sunday. I have an analog mixer and a
> digital audio recorder, and I know exactly what the
> meters on each are telling me. At the same time
> I make sure each participant has enough monitor
> to hear themself, and that I'm not blasting out the
> house.
>
And with this great experience, you still don't know the difference
between the waveform of a sound and the envelope of that waveform.

> They trust me enough to know what I'm doing.
>
I've given the conductor of a full symphony orchestra a copy of a
recording showing a full dynamic range as performed under his
leadership, peaking at -0.5 dB, and his response was "It's too quiet."
He expected to hear some compression on the recording.

--
Tciao for Now!

John.

[Scott Dorsey]

John Williamson > wrote:
>>
>I've given the conductor of a full symphony orchestra a copy of a
>recording showing a full dynamic range as performed under his
>leadership, peaking at -0.5 dB, and his response was "It's too quiet."
>He expected to hear some compression on the recording.

I get that too... and I suggest turning it up. Sometimes there is no
way around it, but you'd be surprised how many people like it after
they try it.

It's also true, though, that a little peak limiting can help you bring
those levels up a good bit without restricting the dynamics or sacrificing
any sound quality. You can bring the horn peaks on that crescendo down closer
to the average orchestral level without actually changing perceived loudness
much at all, but it allows you to then bring all the levels up a bit across
the board.... no perceived dynamic range lost. Mind you if there is a lot
of tympani going on, this won't help you so much.
--scott

--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[[email protected]]

On Friday, October 16, 2015 at 5:12:39 PM UTC-4, Scott Dorsey wrote:
> John Williamson > wrote:
> >>
> >I've given the conductor of a full symphony orchestra a copy of a
> >recording showing a full dynamic range as performed under his
> >leadership, peaking at -0.5 dB, and his response was "It's too quiet."
> >He expected to hear some compression on the recording.
>
> I get that too... and I suggest turning it up. Sometimes there is no
> way around it, but you'd be surprised how many people like it after
> they try it.
>
> It's also true, though, that a little peak limiting can help you bring
> those levels up a good bit without restricting the dynamics or sacrificing
> any sound quality. You can bring the horn peaks on that crescendo down closer
> to the average orchestral level without actually changing perceived loudness
> much at all, but it allows you to then bring all the levels up a bit across
> the board.... no perceived dynamic range lost. Mind you if there is a lot
> of tympani going on, this won't help you so much.
> --scott
>
> --
> "C'est un Nagra. C'est suisse, et tres, tres precis."

I would go ahead and clip (yes clip) non musical sounds such as hand claps or an occasional lound beat, and I stick one of these logos and the link on the recording.

http://turnmeup.org/

Mark

[None]

< thikskull @ dumb****shortbus.edu > wrote in message
news:34487d6f-9549-4cc6-8568
> I cannot perform even basic math, not because I didn't want to, but
> simply because no teacher could teach it to me.

Blame the teachers, as you always do. Did you throw tantrums and argue
with them? Did you insist that they were wrong, and two plus two
didn't equal four; that you knew better? Did you accuse them of having
a vested economic interest in two plus two? No wonder they gave you a
"mercy diploma" to get rid of you after seven years at a two-year
junior college!

By the way, the surest way to get me to insult you is for you to
mention me in a post. I know this won't help you, because you will
have forgotten it moments after you read it.

> I know exactly what the meters on each are telling me.

Based on everything you've posted, that statement is a huge stinking
pile of bull****. You have no idea what the meters are telling you.
You've gone to great effort to prove that.

> They trust me enough to know what I'm doing.

Yeah, their standards are about as low as they get. FCABFGAT?

[geoff]

On 17/10/2015 2:47 a.m., Scott Dorsey wrote:
> None > wrote:
>> < thickskullocks @ gmail.com> wrote in message
>> ...
>>> geoff wrote: "- show quoted text -
>>> Semantics.
>>> I told you - no "semantics", only ENGLISH.
>>
>> 1. You're not the moderator. You don't get to tell others what to
>> post. It makes no difference what you told anybody. Have you forgotten
>> that already, since the last time you were reminded? It was only this
>> morning.
>
> Part of the problem is that this is basically a mathematical definition
> that people are arguing over. Verbal english descriptions are only going
> to be approximate analogies; the actual definition is a numerical function.
> --scott
>


Disagree to a point. Mathematical in nature, but aural in importance,
and most easily and conveniently observed visually (for those of us
operating in this century with more than a teletype screen !)...

geoff

[geoff]

On 17/10/2015 7:39 a.m., John Williamson wrote:

>>
> Just about every post you make demonstrates your total lack of
> comprehension of the basics of audio recording as well as your ignorance
> about the aims of and the techniques used in the modern audio production
> process.
>
> Your posts in this thread show your inability to tell the difference
> between a waveform, the amplitude envelope of that waveform and a
> graphical description of how the gain of an amplifier inside a
> synthesiser varies, even after many repeated attempts to explain it to
> you in simple terms.


The maths teachers apparently had a similar problem.

geoff

[[email protected]]

geoff wrote: "The maths teachers apparently had a similar problem"

You know what geoff - and John W, et al?

If you were in the same boat I was, cognitively,
I'd show you something not one of you has
shown me: A little COMPASSION.

Thank you!

[[email protected]]

John Williamson wrote: "AKA the people who listen to and pay for the recordings. "

WRONG again!

The artists, labels and such are the ones
MAKING and MARKETING the music,
NOT buying it.

And the amount of compression used in
the 1970s and '80s was sufficient and served
the purposes you described, not like today
where they remove any hint of transients
or swell of a chorus.

[None]

< thickmama @gmail.com> wrote in message
...
> geoff wrote: "The maths teachers apparently had a similar problem"
>
> You know what geoff - and John W, et al?

You know what, ThickSkull?

> I'd show you something not one of you has
> shown me: A little COMPASSION.

You have not shown the group compassion. You have treated the readers
of the group with contempt, even those few who remain polite to you.
Nobody on this newsgroup is at fault for your inability to comprehend
anything. Stop blaming everyone else. It's pathetic.

> Thank you!

**** you, asshole. YFSBDF.

[Peter Larsen[_3_]]

On 17-10-2015 00:49, wrote:

> John Williamson wrote: "AKA the people who listen to and pay for the recordings. "

> WRONG again!

> The artists, labels and such are the ones
> MAKING and MARKETING the music,
> NOT buying it.

> And the amount of compression used in
> the 1970s and '80s was sufficient and served
> the purposes you described,

No. It is only the digitally deployed multiband compressor that has
allowed imperceptible compression of natural recordings. It is with
audio dynamics as with image dynamics, to make it sound or look right,
with all details available it is required to fit the dynamics to what
the equipment and playback situation allows.

> not like today
> where they remove any hint of transients
> or swell of a chorus.

Blanket statement that does not now allow for technical and musical
genre, also what you are describing is limiting. Some of the time
limiting is a good thing, because it allows deploying processing only
for the actual duration of the problem.

If the use of a compressor is audible, then it is also incorrect unless
used as an artistic effect. So called parallel compression is often
preferred by live sound engineers because it does a good job of raising
low levels without doing anything about/to the peaks.

Also please note that transients are not just the loud stuff. It is very
often the noise components of instrument sound, such rattle from guitar
strings and the squeak of left hand fingers on the strings and all those
fascinating noises that paper makes. One of the problems in deploying
compression is to avoid that such become unnaturally audible.

Example of natural recording multiband compressed for the medium:

https://soundcloud.com/jexper-holmen/ravnholm

Kind regards

Peter Larsen

[Les Cargill[_4_]]

<snip>

I am so glad I read that. Thank you for sharing your invaluable views
with the assembled group.

Please subscribe me to your newsletter.

--
Les Cargill

[John Williamson]

On 16/10/2015 22:12, Scott Dorsey wrote:
> John Williamson > wrote:
>>>
>> I've given the conductor of a full symphony orchestra a copy of a
>> recording showing a full dynamic range as performed under his
>> leadership, peaking at -0.5 dB, and his response was "It's too quiet."
>> He expected to hear some compression on the recording.
>
> I get that too... and I suggest turning it up. Sometimes there is no
> way around it, but you'd be surprised how many people like it after
> they try it.
>
It's not a problem for me, and some gain automation in post can work
well enough, just losing a touch of the purity I was aiming for.

> It's also true, though, that a little peak limiting can help you bring
> those levels up a good bit without restricting the dynamics or sacrificing
> any sound quality. You can bring the horn peaks on that crescendo down closer
> to the average orchestral level without actually changing perceived loudness
> much at all, but it allows you to then bring all the levels up a bit across
> the board.... no perceived dynamic range lost. Mind you if there is a lot
> of tympani going on, this won't help you so much.
> --scott
>
In this case, the problem on one track was the 32 foot pipes on the
organ with all the stops out, swapping the theme with the rest of the
orchestra. On another track, I could have done with spotting the solo
sax, but wasn't allowed to by the cathedral authorities. <Shrug> They
got a copy of the masters to mix as they wish at the college, and it
would be good practice for a student.

--
Tciao for Now!

John.

[John Williamson]

On 17/10/2015 00:41, wrote:
> geoff wrote: "The maths teachers apparently had a similar problem"
>
> You know what geoff - and John W, et al?
>
> If you were in the same boat I was, cognitively,
> I'd show you something not one of you has
> shown me: A little COMPASSION.
>
> Thank you!
>
Sometimes, true compassion includes trying to help rectify the
situation. We are trying to help you understand where you are going
wrong, while you continue insisting we are all wrong and you are right.

Try considering what we could do to help you understand the problem.
However, posting the type of stuff you post, then attacking the person
who politely points out your errors is not the way to endear yourself to
others, and can lead to sarcasm.

--
Tciao for Now!

John.

[John Williamson]

On 16/10/2015 23:16, geoff wrote:
> On 17/10/2015 2:47 a.m., Scott Dorsey wrote:

>> Part of the problem is that this is basically a mathematical definition
>> that people are arguing over. Verbal english descriptions are only going
>> to be approximate analogies; the actual definition is a numerical
>> function.
>> --scott
>>
>
>
> Disagree to a point. Mathematical in nature, but aural in importance,
> and most easily and conveniently observed visually (for those of us
> operating in this century with more than a teletype screen !)...
>
However, to get the best out of the visual representation, it helps a
lot if the basic principles underlying the image are well understood,
and it is used only as an aid to understanding the aural component.


--
Tciao for Now!

John.

[Peter Larsen[_3_]]

On 16-10-2015 19:14, wrote:

> Peter Larsen wrote: "not so strong on actual audio recording."

> Average loudness is not an opinion, it is fact that
> finished product has gotten louder in the last 20
> years. At least I have come to grips with what/who
> is driving it.

The race for loudness is not new. It was probable Stereophile that
documented it quite carefully by comparing calorimetrically determined
average to peak ratio on records and on FM stations in the second half
of the 1970-ties. Once upon a time, probably 1999, I posted additional
documentation of it on something called usenet. If you find the thread
DO take care to read it to the end, most of the data in it are 3 dB off
due to a software oddity as explained in a late post in it.

> But please elaborate on that second part - about
> audio recording.

You say that you do church audio. Acoustic instrumenets or amplified
average us church rockband? - derogation intentional, but not aimed at
you :)

Kind regards

Peter Larsen

[[email protected]]

Peter Larsen wrote: "Blanket statement that does not now allow for technical and musical genre"


I guess I should specify Pop/Top 40. And it is very likely
both compressed and peak limited before regaining back
to full scale. Very fatiguing to listen to, compared to the
top 40 of perhaps 35 years ago.


Even though some of you may not officially recognize the
TT DR(Dynamic Range) meters(static and real-time) as
official production aids, but a modern pop song's DR value
of "7", compared to a DR12 to 14 is telling. It might be
measuring crest-factor, or even something else - but the
lower resulting value on more recent material is telling,
and confirms what I have been told to listen for.

In any case, I can hear the 1977 Linda Ronstadt in my
car as plainly - and more pleasantly - than the 2014
Ariana Grande, just by turning it up a notch. And no,
I'm not afraid of the older more dynamic piece "shredding"
my car speakers(!). And my Best of Chic CD REALLY
gets other drivers' attention at a stop light in warmer
weather - when all our windows are open. Avg. DR is
15!!

[[email protected]]

Peter Larsen: "You say that you do church audio. Acoustic instrumenets or amplified "


Amplified. Most evangelical or born-again use "rock". And yes, I
use the Yamaha mixer's built in compress knobs to allow
the worship vocalists to cut through the music a little, as well
as to even out our Pastor. LOL, In a 12x24m sanctuary, 6m
cathedral ceiling, with proper system gain structure throughout,
most of my faders ride between -5 to -10dBVU(analog board),
feeding Mackie powered 15's set at default 12 o'clock volume
position.


I just crank up the monitor outs to the Tascam digital
recorder, and make sure I don't peak above -6 full scale.
I can bring things up(reasonably!) later in post.

[[email protected]]

Peter Laraen wrote: "Example of natural recording multiband compressed for the medium: "

Excellent! The chords remind me a bit of my hero - Zappa!

Listening carefully, on fullsized system and through good
cans, I can just hear the effects of the multi-band. The
sustains on the piano are just a tad more present than
they would be if listening in person - but some of that
could also be room. Changes in overall loudness are
preserved - and startling when heard for the first time!

Big-label producers: lend Peter your ears!!

[[email protected]]

Probably started with AM radio in the 60s. WABC
Anybody remember WWRL I think it was at 1600 on the dial in NYC.
They must have been running 40dB of comp. When the announcer took a breath, you could hear every little sound and hum in the studio.
Mark

[[email protected]]

wrote:
Probably started with AM radio in the 60s. WABC
Anybody remember WWRL I think it was at 1600 on the dial in NYC.
They must have been running 40dB of comp. When the announcer took a breath, you could hear every little sound and hum in the studio.
Mark "


Wow! Even a toilet flushing down the hallway, or someone farting
in the sales pool?

[Angus Kerr]

On Saturday, October 17, 2015 at 4:31:08 AM UTC+2, Peter Larsen wrote:
> Also please note that transients are not just the loud stuff. It is very
> often the noise components of instrument sound, such rattle from guitar
> strings and the squeak of left hand fingers on the strings and all those
> fascinating noises that paper makes. One of the problems in deploying
> compression is to avoid that such become unnaturally audible.
>
> Example of natural recording multiband compressed for the medium:
>
> https://soundcloud.com/jexper-holmen/ravnholm
>

Afraid I find the content so fatiguing, I can't concentrate the effects of the multiband compressor. Although I can perceive a reduction in dynamic range that is quite subtle and smooth.


Typical modern classical music. Sends shivers down my spine in not a good way.

But kudos Peter

Nice recording

-Angus.

[Peter Larsen[_3_]]

On 18-10-2015 09:45, Angus Kerr wrote:

> On Saturday, October 17, 2015 at 4:31:08 AM UTC+2, Peter Larsen wrote:

>> Also please note that transients are not just the loud stuff. It is very
>> often the noise components of instrument sound, such rattle from guitar
>> strings and the squeak of left hand fingers on the strings and all those
>> fascinating noises that paper makes. One of the problems in deploying
>> compression is to avoid that such become unnaturally audible.

>> Example of natural recording multiband compressed for the medium:

>> https://soundcloud.com/jexper-holmen/ravnholm

> Afraid I find the content so fatiguing, I can't concentrate the effects of the multiband compressor. Although I can perceive a reduction in dynamic range that is quite subtle and smooth.

I had not expected to use earplugs when recording a chamber quartet, but
did.

> Typical modern classical music. Sends shivers down my spine in not a good way.

> But kudos Peter

> Nice recording

[bowing] thank you!

> -Angus.

Kind regards

Peter Larsen

[Angus Kerr]

On Sunday, October 18, 2015 at 7:11:36 PM UTC+2, Peter Larsen wrote:
> On 18-10-2015 09:45, Angus Kerr wrote:
>
-snip-
> > Afraid I find the content so fatiguing, I can't concentrate the effects of the multiband compressor. Although I can perceive a reduction in dynamic range that is quite subtle and smooth.
>
> I had not expected to use earplugs when recording a chamber quartet, but
> did.

Lol. I often wonder what is the point of this dissonance-on-purpose type of music. It seems almost set up to create maximum offense. I mean, surely we should be now post-modern where a clever melody of some sort can exist with dissonance and shocking backing or whatever. But having some across some composition professors they are about as whacko as you can get. I suppose you have to push the boundaries. I remember doing theory exams where you are given a melody to do 4 part harmonies to or whatever, and thinking to myself, "god this melody sucks, why bother to do harmonies to it?"

Also makes me laugh, because in the late seventies and early eighties when John Williams was talking about his Star Wars soundtrack, he said there was a lot of resistance and rebellion in the LSO regarding his 'dissonant and unmusical' soundtrack. They should get a load of your group's recording!

>
> > Typical modern classical music. Sends shivers down my spine in not a good way.
>
> > But kudos Peter
>
> > Nice recording
>
> [bowing] thank you!
>
> > -Angus.
>
> Kind regards
>
> Peter Larsen

-Angus.

[Les Cargill[_4_]]

Angus Kerr wrote:
> On Sunday, October 18, 2015 at 7:11:36 PM UTC+2, Peter Larsen wrote:
>> On 18-10-2015 09:45, Angus Kerr wrote:
>>
> -snip-
>>> Afraid I find the content so fatiguing, I can't concentrate the
>>> effects of the multiband compressor. Although I can perceive a
>>> reduction in dynamic range that is quite subtle and smooth.
>>
>> I had not expected to use earplugs when recording a chamber
>> quartet, but did.
>
> Lol. I often wonder what is the point of this dissonance-on-purpose
> type of music. It seems almost set up to create maximum offense. I
> mean, surely we should be now post-modern where a clever melody of
> some sort can exist with dissonance and shocking backing or whatever.

"Rite of Spring" was first performed in 1913.

> But having some across some composition professors they are about as
> whacko as you can get. I suppose you have to push the boundaries. I
> remember doing theory exams where you are given a melody to do 4 part
> harmonies to or whatever, and thinking to myself, "god this melody
> sucks, why bother to do harmonies to it?"
>
> Also makes me laugh, because in the late seventies and early eighties
> when John Williams was talking about his Star Wars soundtrack, he
> said there was a lot of resistance and rebellion in the LSO regarding
> his 'dissonant and unmusical' soundtrack. They should get a load of
> your group's recording!
>

MOAR FRENCH HORNS! MOAR! MOAR! It's like Bill Moffit "Soundpower"
arrangements.

Eaughhhh... give me Aaron Copeland or Elmer Bernstein any day.

<snip>
--
Les Cargill