Dual mono vs. mono mono interrogative...
Keith G
they present me with an interesting dilmma - whether to record them as 'dual
mono' or 'mono mono'..??
So, is there any convention or meaningful reason why they should not be
recorded as 'dual mono' if I choose, or indeed is there any merit in
recording them thus?? One thing I notice is the spitchy bits (no cleaning on
these samples) are nicely moved into the middle and in some way buried in
the mono transcriptions, but I do hear other interesting differences!
Here are a couple of near-identical samples to compare:
http://www.moirac.adsl24.co.uk/showntell/DBdualmono.mp3
http://www.moirac.adsl24.co.uk/showntell/DBmonomono.mp3
Anyone with any thoughts?
TIA, as usual....
Dave Plowman (News)
Keith G <k...@moirac.adsl24.co.uk> wrote:
> I have a number of (predominantly jazz) mono albums I want to 'digitise'
> and they present me with an interesting dilmma - whether to record them
> as 'dual mono' or 'mono mono'..??
Do you mean by 'dual mono' using a stereo cart and keeping things stereo
throughout? Or feeding a 'proper' mono signal to left and right legs of a
stereo signal?
--
*Certain frogs can be frozen solid, then thawed, and survive *
Dave Plowman da...@davenoise.co.uk London SW
To e-mail, change noise into sound.
Phil Allison
"Keith G"
>I have a number of (predominantly jazz) mono albums I want to 'digitise'
>and they present me with an interesting dilmma - whether to record them as
>'dual mono' or 'mono mono'..??
>
> So, is there any convention or meaningful reason why they should not be
> recorded as 'dual mono' if I choose, or indeed is there any merit in
> recording them thus?? One thing I notice is the spitchy bits (no cleaning
> on these samples) are nicely moved into the middle and in some way buried
> in the mono transcriptions, but I do hear other interesting differences!
** The interesting thing about surface noise on a vinyl LP is that it is
always in stereo !!
In the case of a mono LP, the music is gonna appear smack in the centre of a
pair of stereo speaker - if everything is well matched up.
This makes it possible to mentally " tune out " such surface noise as it is
not coming from the same direction as the music but rather from the far left
and far right of it.
...... Phil
Don Pearce
wrote:
You can use the dual mono signal quite handily. Put it into your audio
software and use the facility (which most have) of centre channel or
vocal extraction. That way you will lose the spitches (which are 99%
left or right, but never centre), and keep the good stuff.
Also if you need to do any real repairs to remove a pop, you can
select the option to copy and paste from one channel to the other.
Then save it as pure mono, and it will halve the file size.
d
David Looser
news:4af0c60a....@news.eternal-september.org...
> On Mon, 2 Nov 2009 23:24:36 -0000, "Keith G" <k...@moirac.adsl24.co.uk>
> wrote:
>
>>
>
> You can use the dual mono signal quite handily. Put it into your audio
> software and use the facility (which most have) of centre channel or
> vocal extraction. That way you will lose the spitches (which are 99%
> left or right, but never centre),
> Also if you need to do any real repairs to remove a pop, you can
> select the option to copy and paste from one channel to the other.
>
The click energy can be mostly in the L, R, difference or sum channels
depending on how the damage was done. In the case of a click that is mainly
L or R then a copy & paste from the "good" channel to the "bad" can be very
effective, when the click energy is mainly difference averaging the signal
can largely eliminate it. But there are a hard core (far more than 1%) of
cases when none of those work, and we are back to removing the click the
hard way.
> Then save it as pure mono, and it will halve the file size.
>
And will remove a lot (though nothing like 99%) of the surface crackle and
much of the distortion.
David.
Brian Gaff
removed, and nasty wear artefacts do not spread over the sound stage.
Brian
--
Brian Gaff....Note, this account does not accept Bcc: email.
graphics are great, but the blind can't hear them
Email: bri...@blueyonder.co.uk
______________________________________________________________________________________________________________
"Keith G" <k...@moirac.adsl24.co.uk> wrote in message
news:hcnpnh$hu5$1...@energise.enta.net...
Keith G
"Brian Gaff" <Bri...@blueyonder.co.uk> wrote in message
news:dvSHm.1221$Ym4...@text.news.virginmedia.com...
> Well, I prefer mono after recording so at least out of phase crackles are
> removed, and nasty wear artefacts do not spread over the sound stage.
There are times when 'stereo' helps (large scale, orchestral) but most of
the time it's a gimmick I don't *have* to have, I find!! Done badly (20 foot
wide pianos and ping pong solo instruments) it is atrocious and has me
reaching for the 'mono button' I haven't got on any of my amps!!
Said it before - I do a lot (if not most) of my listening 'off axis'
anyway!!
But then, that's just me!! ;-)
Keith G
"Dave Plowman (News)" <da...@davenoise.co.uk> wrote in message
news:50b425d...@davenoise.co.uk...
> In article <hcnpnh$hu5$1...@energise.enta.net>,
> Keith G <k...@moirac.adsl24.co.uk> wrote:
>> I have a number of (predominantly jazz) mono albums I want to 'digitise'
>> and they present me with an interesting dilmma - whether to record them
>> as 'dual mono' or 'mono mono'..??
>
>
> Do you mean by 'dual mono' using a stereo cart and keeping things stereo
> throughout? Or feeding a 'proper' mono signal to left and right legs of a
> stereo signal?
The former: a 100% two channel 'stereo' setup all the way from the pickup to
the soundcard; then either capturing/saving as two, L+R channels - what I
call 'dual mono' when from a 'mono' source, or one channel - what I call
'mono mono' in this instance, to differentiate from a more usual mono
capture from a stereo recording!
Keith G
"Don Pearce" <sp...@spam.com> wrote
> You can use the dual mono signal quite handily. Put it into your audio
> software and use the facility (which most have) of centre channel or
> vocal extraction. That way you will lose the spitches (which are 99%
> left or right, but never centre), and keep the good stuff.
That is what I suspect happens automatically when the capture is set to
'mono' and the file is saved as such - I can't see any facilities in the
software to enable you to choose it as a process..??
>
> Also if you need to do any real repairs to remove a pop, you can
> select the option to copy and paste from one channel to the other.
Serious pops are easily removed in SoundForge; it's the continuous crackle
that can be a nuisance (if it bothers you), but I am hoping for a
breakthrough *offlist* on that a little later on.
Might be worth a mention that the 'mono' clips I posted are off the Brubeck
'Jazz Goes To College' album which dates from 1954 - my copy is an original
Columbia CL566 in stunning condition. Neither of those clips have had any
cleaning whatsoever!
>
> Then save it as pure mono, and it will halve the file size.
Yes - another bonus of mono recordings.
Keith G
"Phil Allison" <phi...@tpg.com.au> wrote in message
news:7l9dkcF...@mid.individual.net...
>
> "Keith G"
>>I have a number of (predominantly jazz) mono albums I want to 'digitise'
>>and they present me with an interesting dilmma - whether to record them as
>>'dual mono' or 'mono mono'..??
>>
>> So, is there any convention or meaningful reason why they should not be
>> recorded as 'dual mono' if I choose, or indeed is there any merit in
>> recording them thus?? One thing I notice is the spitchy bits (no cleaning
>> on these samples) are nicely moved into the middle and in some way buried
>> in the mono transcriptions, but I do hear other interesting differences!
>
>
> ** The interesting thing about surface noise on a vinyl LP is that it is
> always in stereo !!
Certainly is when you are playing it, but it gets nicely buried (to a point)
when transcribed to a mono recording!
>
> In the case of a mono LP, the music is gonna appear smack in the centre of
> a pair of stereo speaker - if everything is well matched up.
A 'dual mono' waveform is a good visual check for the overall channel
balance of the recording/replay sysytem!
>
> This makes it possible to mentally " tune out " such surface noise as it
> is not coming from the same direction as the music but rather from the far
> left and far right of it.
And is always 'up front and in your face' when the music 'soundstage' goes
*way back over there*! The trick I suspect successful vinylistas manage to
achieve most of the time is simply to follow the music and ignore the
fireworks off to the side!
Don Pearce
wrote:
>
>"Don Pearce" <sp...@spam.com> wrote
>
>
>> You can use the dual mono signal quite handily. Put it into your audio
>> software and use the facility (which most have) of centre channel or
>> vocal extraction. That way you will lose the spitches (which are 99%
>> left or right, but never centre), and keep the good stuff.
>
>
>That is what I suspect happens automatically when the capture is set to
>'mono' and the file is saved as such - I can't see any facilities in the
>software to enable you to choose it as a process..??
>
Nope, it will just add the two channels together. The crackling will
become a little less evident because it is now coming from the same
spot as the music, and a bit better hidden than when it was separated
spatially.
I don't know what Sound Forge (is that what you use?) does. It is
there in Audition.
>
>>
>> Also if you need to do any real repairs to remove a pop, you can
>> select the option to copy and paste from one channel to the other.
>
>
>Serious pops are easily removed in SoundForge; it's the continuous crackle
>that can be a nuisance (if it bothers you), but I am hoping for a
>breakthrough *offlist* on that a little later on.
>
Is this what you are after?
http://81.174.169.10/odds/DBnoclicks.mp3
I used Izotope RX to do that. If I had some groove noise without music
to sample, I could have got rid of loads more.
>Might be worth a mention that the 'mono' clips I posted are off the Brubeck
>'Jazz Goes To College' album which dates from 1954 - my copy is an original
>Columbia CL566 in stunning condition. Neither of those clips have had any
>cleaning whatsoever!
>
Remarkably good given the age.
>
>>
>> Then save it as pure mono, and it will halve the file size.
>
>
>Yes - another bonus of mono recordings.
>
>
But not what I would describe as a selling point.
d
Phil Allison
"Keith Git "
>
> "Phil Allison"
>>
>
>> ** The interesting thing about surface noise on a vinyl LP is that it is
>> always in stereo !!
>
>
> Certainly is when you are playing it, but it gets nicely buried (to a
> point) when transcribed to a mono recording!
** Totally false assertion.
Surface noise is independent of the signal level impressed on the disk.
Fuckhead.
>> In the case of a mono LP, the music is gonna appear smack in the centre
>> of a pair of stereo speaker - if everything is well matched up.
>
>
> A 'dual mono' waveform is a good visual check for the overall channel
> balance of the recording/replay sysytem!
** Meaningless drivel.
Yawwwwnnnnnnnnnnnnnnnnnnn......
>> This makes it possible to mentally " tune out " such surface noise as it
>> is not coming from the same direction as the music but rather from the
>> far left and far right of it.
>
>
> And is always 'up front and in your face' when the music 'soundstage' goes
> *way back over there*!
** This trolling cunthead asked folk for information.
But all the fucking stupid ass REALLY wanted was a fight.
Game on.
..... Phil
Keith G
>
> "Keith Git "
??
Uh oh, he's *gone* again....
<silly nutter crap snipped - saves wasting time....>
> ** This trolling cunthead asked folk for information.
'Trolling' is a bit strong - more of a 'roll call' if anything (it has been
very quiet in here)....
>
> But all the fucking stupid ass REALLY wanted was a fight.
No, that's you....
>
> Game on.
I think not: a) you're a nutter and I was once advised never to waste time
reasoning with nutters and b) you really aren't that interesting!
(Sorry...)
So - guess again, sunshine!
;-)
Keith G
> On Tue, 3 Nov 2009 13:13:50 -0000, "Keith G" <k...@moirac.adsl24.co.uk>
> wrote:
>
>>
>>"Don Pearce" <sp...@spam.com> wrote
>>
>>
>>> You can use the dual mono signal quite handily. Put it into your audio
>>> software and use the facility (which most have) of centre channel or
>>> vocal extraction. That way you will lose the spitches (which are 99%
>>> left or right, but never centre), and keep the good stuff.
>>
>>
>>That is what I suspect happens automatically when the capture is set to
>>'mono' and the file is saved as such - I can't see any facilities in the
>>software to enable you to choose it as a process..??
>>
> Nope, it will just add the two channels together. The crackling will
> become a little less evident because it is now coming from the same
> spot as the music, and a bit better hidden than when it was separated
> spatially.
OK, that's really what I said originally.
>
> I don't know what Sound Forge (is that what you use?) does. It is
> there in Audition.
I've not looked at it since I stopped combining Swim's se[parate clart and
piano recordings - I find SoundForge just too handy for chopping up and
trimming recordings and the easy removing of serious pops and farts.
> Is this what you are after?
>
> http://81.174.169.10/odds/DBnoclicks.mp3
Yes, that is nice - seems this sort of software has come on a bit since I
first heard it!
>
> I used Izotope RX to do that. If I had some groove noise without music
> to sample, I could have got rid of loads more.
OK. I've noted the name (and found the download sites) - thanks for the
tip....
Brian Gaff
of radiograms to show off with, never mind realism. I think you have
forgotten that there are times when exciting stereo can be fun. Its not
realism, of course its not, but if you really want to hear what happens
when you let a demented sound mixer loose on a multitrack master, look out
for a track called Mandrill, by Mandrill and beware, do not listen on
headphones.
Brian
--
Brian Gaff....Note, this account does not accept Bcc: email.
graphics are great, but the blind can't hear them
Email: bri...@blueyonder.co.uk
______________________________________________________________________________________________________________
"Keith G" <k...@moirac.adsl24.co.uk> wrote in message
news:hcpahq$2tje$4...@energise.enta.net...
Iain Churches
>
> "Phil Allison" <phi...@tpg.com.au> wrote in message
> news:7laru3F...@mid.individual.net...
>>
>> "Keith Git "
>
>
> ??
>
> Uh oh, he's *gone* again....
>
Haven't you noticed, Keith?
There's a full moon.
Iain
Phil Allison
"Keith the Trolling Git "
>
> "Phil Allison"
>>
>
>> ** The interesting thing about surface noise on a vinyl LP is that it is
>> always in stereo !!
>
>
> Certainly is when you are playing it, but it gets nicely buried (to a
> point) when transcribed to a mono recording!
** Totally false assertion.
Surface noise is independent of the signal level impressed on the disk.
Fuckhead.
>> In the case of a mono LP, the music is gonna appear smack in the centre
>> of a pair of stereo speaker - if everything is well matched up.
>
>
> A 'dual mono' waveform is a good visual check for the overall channel
> balance of the recording/replay sysytem!
** Meaningless drivel.
Yawwwwnnnnnnnnnnnnnnnnnnn......
>> This makes it possible to mentally " tune out " such surface noise as it
>> is not coming from the same direction as the music but rather from the
>> far left and far right of it.
>
>
> And is always 'up front and in your face' when the music 'soundstage' goes
> *way back over there*!
** This trolling cunthead asked folk for information.
But all the fucking stupid ass REALLY wanted was a fight.
Game on.
..... Phil
UnsteadyKen
> There's a full moon.
Aye, young 'un, we here in UKRA know to fear the full moon, tis said
that at this time of the month a fearful apparition called the
"keithG" comes roaring through the newsgroup on a terrible dark steed
called "Moaty Bike" and as he comes he cries foul blasphemies the likes
of which us goodly Strictly watchers should never have to hear. Tis
said that old Gran-pa LeSurf was out in the fields tending his crop of
speaker cables and hear the KeithG cry "Hell Pees sound alright" and
his beard turned white overnight. Tis also said that Mother Allison was
in her allotment planting seedies one night when the foul thing came
a'roaring past and she heard the awful cry of "Devils Audio
Broadcasting is OK for casual listening purposes" Well, the poor old
things brain was addled and she hasn't spoken a word of sense since.
Some old 'uns tell that the KeithG's head was turned when he bought a
copy of Des O'Connors Greatest Hit with his paper round money and he's
been a'haunting the neighbourhood ever since. Some also say that on
some days he can be seen in underpasses fiddling with his equipment and
flashing.
Mark my words, no good'll come of it. Beware.
--
Ken O'Meara
http://www.btinternet.com/~unsteadyken/
Keith G
"Iain Churches" <Iai...@kolumbus.fi> wrote in message
news:Bv_Hm.35120$La7....@uutiset.elisa.fi...
:-)
Keith G
"Nutter Allison" <phi...@tpg.com.au> wrote in message
news:7lbpkuF...@mid.individual.net...
>
> "Keith the Trolling Git "
>>
>> "Phil Allison"
>>>
>>
>>> ** The interesting thing about surface noise on a vinyl LP is that it is
>>> always in stereo !!
>>
>>
>> Certainly is when you are playing it, but it gets nicely buried (to a
>> point) when transcribed to a mono recording!
>
>
> ** Totally false assertion.
Oh dear, I know he's a nutter but wot a twat also....
>
> Surface noise is independent of the signal level impressed on the disk.
So what? When recording to mono *everything* the pickup gathers is centred,
be it surface noise, groove damage or the music groove itself and
consequently a lot of the unwanted noise is buried in the music. I posted
two examples of the same clip, one mono and one stereo (dual mono) which
demonstrates this - fucking play them, nutter...
>
> Fuckhead.
This is a newsgroup - there's no need to *sign* your posts, nutter....
Keith G
"UnsteadyKen" <unste...@btinternet.com> wrote in message
news:MPG.255b4b94f...@news.btinternet.com...
'Strictly' watchers...??
:-)
And don't joke about Des O Connor's Greatest Hits - I'm sure that's kicking
about somewhere around here, or has done in the past!!
Keith G
> Well, The two pianos of.. etc, were a thing of their time. Made in the
> days of radiograms to show off with, never mind realism. I think you have
> forgotten that there are times when exciting stereo can be fun.
Hah! I've heard no end of 'Stereo Spectacular' type 'demo records' in the
past and almost certainly have one or two kicking about here somewhere!!
Some of them weren't too bad, as I recall (??) - I'll have to fish them out
and give 'em a spin!
Its not
> realism, of course its not, but if you really want to hear what happens
> when you let a demented sound mixer loose on a multitrack master, look out
> for a track called Mandrill, by Mandrill and beware, do not listen on
> headphones.
Gawd - sounds like it's summat 'orrible like a baboon's purple arse...??
Phil Allison
"Keith the Moronic Git"
"Keith the Trolling Git "
>>>
>>> "Phil Allison"
>>>>
>>>
>>>> ** The interesting thing about surface noise on a vinyl LP is that it
>>>> is always in stereo !!
>>>
>>>
>>> Certainly is when you are playing it, but it gets nicely buried (to a
>>> point) when transcribed to a mono recording!
>>
>>
>> ** Totally false assertion.
>
>
>
> So what?
** How fucking stupid is this DUMB POMMY CUNT !!!
The NOISE is NOT gonna be buried during quiet passages !!!
IMBECILE !!!
... Phil
Iain Churches
> And don't joke about Des O Connor's Greatest Hits - I'm sure that's
> kicking about somewhere around here, or has done in the past!!
Yes, Des O'Connor CBE, He probably still lives in that wacking great
house down in Sussex and drives his maroon and grey turbo Bentley. :-)
Poor chap :-((
Iain Churches
> Well, The two pianos of.. etc, were a thing of their time. Made in the
> days of radiograms to show off with, never mind realism. I think you have
> forgotten that there are times when exciting stereo can be fun. Its not
> realism, of course its not,
>
The Two Pianos Of... and recordings of that ilk, were very much a
fashion statement, as is much popular music. I worked on many
of the Phase Four recordings at Decca.
After the initial stereo showcase recordings, many of them were
multi-microphone multitrack productions. And very popular
they were too:-) No one pretended it had anything to do with
realism.
> but if you really want to hear what happens when you let a demented sound
> mixer loose on a multitrack master, look out for a track called Mandrill,
> by Mandrill and beware, do not listen on headphones.
Fun to blame the sound mixer isn't it? One needs to consider
why the particlar project was made, what it is trying to achieve
and for whom it was intended, Brian. Maybe (probably) not
for the likes of you or I. If it is made for "headbangers", then you
should ask those headbangers if they enjoyed it. If they say
"yes" then the project, even with the demented sound mixer has
reached its goal, and pleased the adiencve for which the music
was intended.
It's rather like asking people who enjoy a Peter
Katin rendition of Debussy, to listen to Thelonius
Monk or Charles Bell playing "And Satan Said"
I find the various categtories of music and also the
styles within those categories fascinating. I was at
a lecture recently where we were asked to listen to
some saxophone clips. Some of the very best playing
dated from the early thirties.(there were probably more
professional saxophone players then than there are now,
so the standard was high and competition very stiff.
The recordings were of course primitive by modern
standards, and the playing (due to the fast vibrato
which was the fashion then) was deemed to be "corny"
Most people couldn't see past these two obstacles
and realise that there were in fact some very good players
indeed at work
Iain
James Perrett
> On Tue, 3 Nov 2009 13:13:50 -0000, "Keith G" <k...@moirac.adsl24.co.uk>
> wrote:
>
>>
>> "Don Pearce" <sp...@spam.com> wrote
>>
>>
>>> You can use the dual mono signal quite handily. Put it into your audio
>>> software and use the facility (which most have) of centre channel or
>>> vocal extraction. That way you will lose the spitches (which are 99%
>>> left or right, but never centre), and keep the good stuff.
>>
>>
>> That is what I suspect happens automatically when the capture is set to
>> 'mono' and the file is saved as such - I can't see any facilities in the
>> software to enable you to choose it as a process..??
>>
> Nope, it will just add the two channels together. The crackling will
> become a little less evident because it is now coming from the same
> spot as the music, and a bit better hidden than when it was separated
> spatially.
>
> I don't know what Sound Forge (is that what you use?) does. It is
> there in Audition.
>
Don - this is actually something that is unique to Audition as far as I
know. The Centre Channel Extractor is much cleverer than it may first
appear as it actually looks at the correlation between channels and leaves
(or removes) signals with the correlation that you choose.
Cheers
James.
Don Pearce
Is that so? I didn't know it was unique. I really only use Audition
because I have sort of grown up with it throughout its CoolEdit
incarnations, and I now use it more or less by instinct.
d
Iain Churches
But AFAIK this Centre Channel Extractor does not exist in CEP Pro
(or at least in the beta testers version that I am familiar with)
Iain
>
> d
Don Pearce
<Iai...@kolumbus.fi> wrote:
>> Is that so? I didn't know it was unique. I really only use Audition
>> because I have sort of grown up with it throughout its CoolEdit
>> incarnations, and I now use it more or less by instinct.''
>
>But AFAIK this Centre Channel Extractor does not exist in CEP Pro
>(or at least in the beta testers version that I am familiar with)
Did it only appear once Adobe bought it? I'm glad they did something
more useful than just making the interface "pretty".
d
Jim Lesurf
Pardon me for hijacking this thread, but the mentions of CEP prompt me to
ask a question about it.
I don't use CEP or know anything about how it works. However in a thread on
a couple of tv/broadcasting technical groups I've been discussing the
problem of intersample peaks that can produce 'overshoots' that can go
above 0dBFS of someone scales up the samples to be too close to 0dBFS.
I've been told that CEP shows the shape inbetween samples if you 'zoom in'
and that it uses an approx to the formally correct sinc function to do
this. But in the discussion there is also mention of using 'sinusoid
curves' as if the process were a simple one of generating a 'smooth fit'
using a spline fit (or similar) of sinusoidal curves.
Can someone here who uses and understands CEP comment on how it displays
waveforms from LPCM data samples, and - critically - does it do the
formally correct sinc method to show the correctly defined output waveform
between samples?
Slainte,
Jim
--
Please use the address on the audiomisc page if you wish to email me.
Electronics http://www.st-and.ac.uk/~www_pa/Scots_Guide/intro/electron.htm
Armstrong Audio http://www.audiomisc.co.uk/Armstrong/armstrong.html
Audio Misc http://www.audiomisc.co.uk/index.html
Don Pearce
I think the drawing method is actually pretty crude. You can get all
sorts of artefacts showing up as you zoom in from a simply block of
colour to individual waves.
Here's an example. I have generated a sine sweep from 100Hz to 10kHz.
Checking it, it is a good sweep, with no amplitude variations.
http://81.174.169.10/odds/sweep.gif
Now zooming in a bit, you can see amplitude anomalies.
http://81.174.169.10/odds/sweep2.gif
But zooming in enough to resolve samples and the waveform, it appears
to have sorted itself out enough that the curve doesn't appear to go
above -6dB, which was the chosen value.
http://81.174.169.10/odds/sweep3.gif
Does that answer the question? I can prod a bit deeper if you like.
d
Don Pearce
Better still, here's a short flash movie showing a slow zoom from the
complete sweep to a single wave.
http://81.174.169.10/odds/sweep.html
I trust you can play flash?
d
Jim Lesurf
> >On Sat, 07 Nov 2009 09:46:40 +0000 (GMT), Jim Lesurf
>
> >>
> >>I've been told that CEP shows the shape inbetween samples if you 'zoom
> >>in' and that it uses an approx to the formally correct sinc function
> >>to do this. But in the discussion there is also mention of using
> >>'sinusoid curves' as if the process were a simple one of generating a
> >>'smooth fit' using a spline fit (or similar) of sinusoidal curves.
> >
> >I think the drawing method is actually pretty crude. You can get all
> >sorts of artefacts showing up as you zoom in from a simply block of
> >colour to individual waves.
Yes. if you don't zoom in to be able to resolve individual samples it looks
like it just skips over samples and then joins the remainder, so giving
artefacts.
[snip]
> >But zooming in enough to resolve samples and the waveform, it appears
> >to have sorted itself out enough that the curve doesn't appear to go
> >above -6dB, which was the chosen value.
Thanks for the above. Alas the snag is that using a sinusoid as the test
waveform may not sort out the question I have in mind!
*If* CEP uses a 'sinusoidal fit' approach then the result could look fine
when the test waveform *is* a simple sinusoid as the waveform is of the
shape the fit presumes. But what if the waveform is nothing like a
sinusoid?
Obvious examples being.
1) An impulse - maybe time-offset as shown on
http://www.audiomisc.co.uk/HFN/OverTheTop/OTT.html
2) Squarewave with an even integer number of samples per period. (i.e. can
be defined by a regular alternation between two values.)
The formally correct method is to use a process equivalent to filter with a
time symmetric sinc founction. Or some near approx. Other forms of
interpolation or 'smooth fit' will be OK for specific examples of waveform
shape, but be in error for others.
> Better still, here's a short flash movie showing a slow zoom from the
> complete sweep to a single wave.
> http://81.174.169.10/odds/sweep.html
> I trust you can play flash?
Yes. ...Or rather FireFox on my new Xubuntu laptop can. :-)
Thanks for the above. The 'movie' is quite useful for me as I hadn't any
idea what kind of presentation CEP would give as you adjusted the 'zoom'.
problem is that with a sinusoid the result doesn't resolve what I was
wondering about!
FWIW what prompted this was someone saying it was a good idea to always
normalise so the max came to -0.5dBFS. I was then pointing out this could
be a mistake if you only looked at the sample values - for reasons shown on
the page I reference above. The question then became, what does CEP
actually display? Does it show the user a waveform that would allows them
to see if this problem was causing their output to exceed 0dBFS or not for
arbitrary waveforms?
Don Pearce
Just as a quickie, here is a piece of music normalized to 0dB, just
looking at the one sample that maxed out. There clearly is overshoot
of the waveform above the zero line; it goes up to about + 0.1dB. The
place I can see that being a possible problem is in the oversampler,
which will presumably try to reproduce that curve as a trajectory of
points, all above full scale. So maybe they have a point, although
0.5dB may be a little excessive.
Now of course if the normalizer worked at the oversampled rate, this
would cease to be a problem.
http://81.174.169.10/odds/over.gif
d
Jim Lesurf
<sp...@spam.com> wrote:
> On Sat, 07 Nov 2009 12:09:41 +0000 (GMT), Jim Lesurf
> <no...@audiomisc.co.uk> wrote:
> Just as a quickie, here is a piece of music normalized to 0dB, just
> looking at the one sample that maxed out. There clearly is overshoot of
> the waveform above the zero line; it goes up to about + 0.1dB.
Alas, that still doesn't necessarily answer my question.
*How* is the waveform line being computed? if it is on an incorrect basis
it may still go though the samples and show some overshoot. But not be the
correct shape or amount of overshoot.
The gif you reference below does look to me like just one sample point with
a section of a sinusoid drawn though it. But is the waveform shown part of
a sinusoid - or something else? Afraid I can't tell by looking at the gif.
> The place
> I can see that being a possible problem is in the oversampler, which
> will presumably try to reproduce that curve as a trajectory of points,
> all above full scale. So maybe they have a point, although 0.5dB may be
> a little excessive.
> Now of course if the normalizer worked at the oversampled rate, this
> would cease to be a problem.
> http://81.174.169.10/odds/over.gif
> d
--
Don Pearce
<no...@audiomisc.co.uk> wrote:
>> Just as a quickie, here is a piece of music normalized to 0dB, just
>> looking at the one sample that maxed out. There clearly is overshoot of
>> the waveform above the zero line; it goes up to about + 0.1dB.
>
>Alas, that still doesn't necessarily answer my question.
>
>*How* is the waveform line being computed? if it is on an incorrect basis
>it may still go though the samples and show some overshoot. But not be the
>correct shape or amount of overshoot.
>
>The gif you reference below does look to me like just one sample point with
>a section of a sinusoid drawn though it. But is the waveform shown part of
>a sinusoid - or something else? Afraid I can't tell by looking at the gif.
>
I'm afraid only the programmer could tell you how that line is
computed, but I can at least show you the context of that last gif.
Here it is a bit wider. The data point you saw is the highest one on
the top trace - just to the right of the marker line. It is part of a
snare drum transient, so not a sine wave, although I have to say that
if we are talking about something narrow enough to poke up between
adjacent data points, it is going to have major frequency content
right up just below Nyquist, so it is to all intents and purposes
sinusoidal as far as the digital domain is concerned.
http://81.174.169.10/odds/wider.gif
Audition, obviously, has no say the way the DAC goes about creating
that trajectory, so unless every DAC uses the same FIR filter, which
will define the line, the displayed line may as well do whatever it is
doing here. I guess they used whatever method they found
computationally easiest - spline probably
d
Keith G
"Phil 'I can't help it, it's the Mixie' Allison" <phi...@tpg.com.au> wrote
in message news:7leii6F...@mid.individual.net...
>
> "Keith the Moronic Git"
Maybe...
> "Keith the Trolling Git "
No, not unless the word 'troll' is in the subject line....
>>>>
>>>> "Phil Allison"
>>>>>
>>>>
>>>>> ** The interesting thing about surface noise on a vinyl LP is that it
>>>>> is always in stereo !!
>>>>
>>>>
>>>> Certainly is when you are playing it, but it gets nicely buried (to a
>>>> point) when transcribed to a mono recording!
>>>
>>>
>>> ** Totally false assertion.
>>
>>> Surface noise is independent of the signal level impressed on the disk.
>>
>>
>> So what?
>
>
> ** How fucking stupid is this DUMB POMMY CUNT !!!
>
> The NOISE is NOT gonna be buried during quiet passages !!!
:-)
Ooh, you *are* a wag.....
Keith G
Yes, never underestimate the power of the *ample-bosomed matron* bloc to
make or break anyone's career in the entertainment industry! No strong
feelings either way about the bloke myself - not my sort of thing by a
country mile, but good luck to him anyway!!
Another one of the Old School who has achieved nobility through longevity is
Bruce Forsyth - same difference and good luck to him also!!
Keith G
"Don Pearce" <sp...@spam.com> wrote
> I'm afraid only the programmer could tell you how that line is
> computed, but I can at least show you the context of that last gif.
> Here it is a bit wider. The data point you saw is the highest one on
> the top trace - just to the right of the marker line. It is part of a
> snare drum transient, so not a sine wave, although I have to say that
> if we are talking about something narrow enough to poke up between
> adjacent data points, it is going to have major frequency content
> right up just below Nyquist, so it is to all intents and purposes
> sinusoidal as far as the digital domain is concerned.
>
> http://81.174.169.10/odds/wider.gif
>
> Audition, obviously, has no say the way the DAC goes about creating
> that trajectory, so unless every DAC uses the same FIR filter, which
> will define the line, the displayed line may as well do whatever it is
> doing here. I guess they used whatever method they found
> computationally easiest - spline probably
That looks so *lo-res* - is it the best Audition can do? What 'zoom factor'
applies?
Here's a similar strength waveform in Sound Forge at 8:1
http://www.moirac.adsl24.co.uk/showntell/waveform.jpg
(It can be flattened out considerably at 24: 1 with correspondingly more
'dots' which I suspect are sampling points..??)
Ken
wrote:
> That looks so *lo-res* - is it the best Audition can do?
No.
Don Pearce
wrote:
>
No, that is doing pretty much the same thing. Apart from the fact that
Sound Forge clearly just joins the dots with straight lines, while
Audition makes curves.
d
Keith G
>
> "Brian Gaff" <Bri...@blueyonder.co.uk> wrote in message
> news:ZCYHm.1341$Ym4...@text.news.virginmedia.com...
>> Well, The two pianos of.. etc, were a thing of their time. Made in the
>> days of radiograms to show off with, never mind realism. I think you
>> have
>> forgotten that there are times when exciting stereo can be fun. Its not
>> realism, of course its not,
>
>>
>
> The Two Pianos Of... and recordings of that ilk, were very much a
> fashion statement, as is much popular music. I worked on many
> of the Phase Four recordings at Decca.
>
> After the initial stereo showcase recordings, many of them were
> multi-microphone multitrack productions. And very popular
> they were too:-) No one pretended it had anything to do with
> realism.
The trouble with 'stereo' is that it encourages the geeks who think they can
replicate a real live experience in their own living rooms like '3D
depth/spatiality' is *everything*!! (Even now there are twats in here who
will blather on about 'proper stereo'!!)
Once the 'audio ping pong' trick was played out I suspect Joe Ordinaire just
ignored it and listened to the music. A 'stereo' label on a record doesn't
particularly excite me - like I have said often before, I guess I'm too 'off
axis' to care much of the time!!
That all said, of course, ignoring the 'stereo capabilities' of one's
'stereo hifi' is one thing - the said hifi' not having any decent stereo
capabilities to start with is quite another!
Some of the very best playing
> dated from the early thirties.(there were probably more
> professional saxophone players then than there are now,
> so the standard was high and competition very stiff.
> The recordings were of course primitive by modern
> standards, and the playing (due to the fast vibrato
> which was the fashion then) was deemed to be "corny"
> Most people couldn't see past these two obstacles
> and realise that there were in fact some very good players
> indeed at work
Oh yes and I don't care how 'primitive' the recordings are by today's
standards (?? are you kidding??) - all part and parcel of the experience for
me!
(Caruso had a little tinny voice in real life, didn't he? :-)
Keith G
"Don Pearce" <sp...@spam.com> wrote
> No, that is doing pretty much the same thing. Apart from the fact that
> Sound Forge clearly just joins the dots with straight lines, while
> Audition makes curves.
OK, but at what 'zoom'? - Sound Forge smooths out at 6:1 as per:
http://www.moirac.adsl24.co.uk/showntell/smoothwaveform.jpg
But I like that SF isn't too smoothed out, it makes it easier to do creative
editing - normally straightening out the waveform, but here's a bit of
hi-res 24:1 editing going fairly deliberately the other way:
http://www.moirac.adsl24.co.uk/showntell/editedwaveform.jpg
:-)
Jim Lesurf
<sp...@spam.com> wrote:
> On Sat, 07 Nov 2009 13:17:46 +0000 (GMT), Jim Lesurf
> <no...@audiomisc.co.uk> wrote:
> >
> >*How* is the waveform line being computed? if it is on an incorrect
> >basis it may still go though the samples and show some overshoot. But
> >not be the correct shape or amount of overshoot.
> I'm afraid only the programmer could tell you how that line is computed,
> but I can at least show you the context of that last gif. Here it is a
> bit wider.
> http://81.174.169.10/odds/wider.gif
OK.
> The data point you saw is the highest one on the top trace -
> just to the right of the marker line. It is part of a snare drum
> transient, so not a sine wave, although I have to say that if we are
> talking about something narrow enough to poke up between adjacent data
> points, it is going to have major frequency content right up just below
> Nyquist, so it is to all intents and purposes sinusoidal as far as the
> digital domain is concerned.
I don't really agree with the last point for a number of reasons. For
example, consider how high an f2/s sinusoid could go above the samples. Yet
fs/2 is well below fs. Another is to see some of the real-world examples I
found are reported on the webpage. Overshoots in the +1dBFS to +2dBFs
region certainly occur with some 'max loudness' CDs.
Another reason is to bear in mind that a waveform composed of a series of
harmonics can - with them in phase - generate very sharp peaks with
relatively little HF content. (And if you look at other measurements these
peaked waveforms certainly arise for some instruments like violin or
trumpet.)
> Audition, obviously, has no say the way the DAC goes about creating that
> trajectory, so unless every DAC uses the same FIR filter, which will
> define the line, the displayed line may as well do whatever it is doing
> here. I guess they used whatever method they found computationally
> easiest - spline probably
That shows the problem. The point here is that a correctly working DAC
should use the well-defined sinc TDA or equivalent. So far as I know, that
approach has been pretty much the common standard since the first Philips
chipsets for CDDA replay. It also accords well with what the sampling
theorem specifies for correct reconstruction. So if CEP or some other
program uses another method because the programmer found it "easier" we
have no idea if what CEP shows is what most domestic DACs would produce.
The problem with a programmer treating this as an exercise in 'getting a
smooth fit' is that this is *not* the basis in Information Theory. So the
result of a correct reconstruction may look 'less smooth' but actually be
the waveform the samples define.
I'll ask the person who said CEP does use sinc if this is stated in the
documentation for the program.
Thanks.
Jim
Don Pearce
wrote:
>
So-many-to-1 just isn't relevant as a figure. What you see at what
zoom will depend on how long the original bit of music was. What you
need from your software is what Audition tells you down the bottom.
The exact start and stop times of the visible window and whatever you
have selected inside it. You can see that easily without the box when
it is zoomed in far enough to show the sample points.
Now, as for that second edit waveform, I'm afraid it shows the
limitations of Sound Forge. And in Audition, I don't think anyone
would ever bother to "paint" out a click.
d
Don Pearce
<no...@audiomisc.co.uk> wrote:
OK, let me think about that. You're probably right, because you
obviously have thought about it - which has prompted your posts.
>> Audition, obviously, has no say the way the DAC goes about creating that
>> trajectory, so unless every DAC uses the same FIR filter, which will
>> define the line, the displayed line may as well do whatever it is doing
>> here. I guess they used whatever method they found computationally
>> easiest - spline probably
>
>That shows the problem. The point here is that a correctly working DAC
>should use the well-defined sinc TDA or equivalent. So far as I know, that
>approach has been pretty much the common standard since the first Philips
>chipsets for CDDA replay. It also accords well with what the sampling
>theorem specifies for correct reconstruction. So if CEP or some other
>program uses another method because the programmer found it "easier" we
>have no idea if what CEP shows is what most domestic DACs would produce.
>
>The problem with a programmer treating this as an exercise in 'getting a
>smooth fit' is that this is *not* the basis in Information Theory. So the
>result of a correct reconstruction may look 'less smooth' but actually be
>the waveform the samples define.
>
>I'll ask the person who said CEP does use sinc if this is stated in the
>documentation for the program.
>
Remember that CEP started in days when processor power was at a bit of
a premium, and a programmer couldn't afford to waste cycles on errors
which were probably going to be contained within the width of the
trace for 99.99% of users.
d
Keith G
"Don Pearce" <sp...@spam.com> wrote
> So-many-to-1 just isn't relevant as a figure. What you see at what
> zoom will depend on how long the original bit of music was.
Hmm, that's not my understanding of how the programme works but, whatever,
it's academic to me - zoomimg in and out is simply a question of spinning
the (mouse) wheel back and forth to show me what amd how much I want to see
on the screen. It's very fast and easy to do in SF; the only constraint is
that the zoom factor has to be 1:32, or bigger, to be able to use the pencil
tool.
What you
> need from your software is what Audition tells you down the bottom.
> The exact start and stop times of the visible window and whatever you
> have selected inside it. You can see that easily without the box when
> it is zoomed in far enough to show the sample points.
>
> Now, as for that second edit waveform, I'm afraid it shows the
> limitations of Sound Forge.
Like how? (Not that I GAS - I'm not selling it for a living or
anything...??)
And in Audition, I don't think anyone
> would ever bother to "paint" out a click.
Quick and easy to do in SF but I'm picking up Izotope RX in a little while
and that may well alter the way I do things - my only interest in 'sound
editing' atm is cutting whole side/whole disc LP recordings into tracks,
trimmimg them and removing bothersome pops and clicks. If Izotope can do
that well enough ('de-clicking' software I tried in the misty past was
pretty much NFG) it will alter the way I do things and obviously speed up
the workflow....
(Except that I don't *need* the workflow speeded up....???)
Don Pearce
wrote:
>
>"Don Pearce" <sp...@spam.com> wrote
>
>> So-many-to-1 just isn't relevant as a figure. What you see at what
>> zoom will depend on how long the original bit of music was.
>
>
>Hmm, that's not my understanding of how the programme works but, whatever,
>it's academic to me - zoomimg in and out is simply a question of spinning
>the (mouse) wheel back and forth to show me what amd how much I want to see
>on the screen. It's very fast and easy to do in SF; the only constraint is
>that the zoom factor has to be 1:32, or bigger, to be able to use the pencil
>tool.
>
>
> What you
>> need from your software is what Audition tells you down the bottom.
>> The exact start and stop times of the visible window and whatever you
>> have selected inside it. You can see that easily without the box when
>> it is zoomed in far enough to show the sample points.
>>
>> Now, as for that second edit waveform, I'm afraid it shows the
>> limitations of Sound Forge.
>
>
>Like how? (Not that I GAS - I'm not selling it for a living or
>anything...??)
>
>
Because what you see on the screen bears only the most passing
resemblance to what emerges from the DAC.
>
> And in Audition, I don't think anyone
>> would ever bother to "paint" out a click.
>
>
>Quick and easy to do in SF but I'm picking up Izotope RX in a little while
>and that may well alter the way I do things - my only interest in 'sound
>editing' atm is cutting whole side/whole disc LP recordings into tracks,
>trimmimg them and removing bothersome pops and clicks. If Izotope can do
>that well enough ('de-clicking' software I tried in the misty past was
>pretty much NFG) it will alter the way I do things and obviously speed up
>the workflow....
>
>(Except that I don't *need* the workflow speeded up....???)
>
>
When I digitize vinyl I tend not to trim (except at the finish of a
side, I leave the needle drop in place). I leave the inter-track
spaces exactly as they are, and just drop in zero delay track markers
to separate them. That way I can play what sounds exactly like a whole
side of the original lp.
d
Jim Lesurf
<k...@moirac.adsl24.co.uk> wrote:
> "Don Pearce" <sp...@spam.com> wrote
> > No, that is doing pretty much the same thing. Apart from the fact that
> > Sound Forge clearly just joins the dots with straight lines, while
> > Audition makes curves.
> OK, but at what 'zoom'? - Sound Forge smooths out at 6:1 as per:
> http://www.moirac.adsl24.co.uk/showntell/smoothwaveform.jpg
Can't see any samples represented on that so no idea what it is doing.
> But I like that SF isn't too smoothed out, it makes it easier to do
> creative editing - normally straightening out the waveform, but here's
> a bit of hi-res 24:1 editing going fairly deliberately the other way:
> http://www.moirac.adsl24.co.uk/showntell/editedwaveform.jpg
As with your earlier example that seems to just 'join dots' with straight
lines. Which is almost worthless as a representation of the actual waveform
you'd get from a correctly working player. For the kinds of examples I've
been discussing the results would be quite different.
So if you were using software with such a display you'd have to keep all
the samples below -5dBFS to be certain no out-of-range peaks were being
produced on replay - even though the display didn't show them. Although in
practice you'd *probably* be safe with waveforms that hadn't already been
boogered if you kept well below -2dBFS.
That's fine if you were aware of this problem. But if not, using such a
display to adjust/edit the sound and make it 'louder' would be bad news for
the listeners.
Slainte,
Jim
Jim Lesurf
<k...@moirac.adsl24.co.uk> wrote:
> "Don Pearce" <sp...@spam.com> wrote
> > I'm afraid only the programmer could tell you how that line is
> > computed, but I can at least show you the context of that last gif.
> > Here it is a bit wider.
> >
> > http://81.174.169.10/odds/wider.gif
> That looks so *lo-res* - is it the best Audition can do? What 'zoom
> factor' applies?
> Here's a similar strength waveform in Sound Forge at 8:1
> http://www.moirac.adsl24.co.uk/showntell/waveform.jpg
Not clear what you mean by "8:1". However the above seems to just show
straight lines between adjacent pairs of (sample?) points. That would be a
useless form of display for detecting overshoots as it can't represent them
at all.
Slainte,
Jim
Keith G
Nice touch! Get an 'auto return' deck for that lovely little 'syonara/see
you later' lift-off sound!! :-)
I leave the inter-track
> spaces exactly as they are, and just drop in zero delay track markers
> to separate them. That way I can play what sounds exactly like a whole
> side of the original lp.
OK, but no good for me - I need individual numbered and titled tracks that
can be found in a search. I frequently know a track I want to hear but don't
know/can't remember which album it's from!! (An album is a folder with
individual track files in it; if I want to play the whole disc I just hit
'Play All' - I cba with playlists and the like!)
Messing with the 'vinyl digitisations' is no chore to me - I get to listen
(over and over, if I want or just let a track run) as I cut up the 'sausage
string', trim the individual tracks to the right 'lead in' and lead out'
lengths, add fades where I want and mute intertrack 'dead wax' surface noise
so the **quiet bits aren't noisy**!!
(Nutter Allison, are you reading this? :-)
Don Pearce
wrote:
The tracks are still numbered, named and findable. It is just that
when you play the whole thing you don't hear the joins. It sounds
exactly like the LP. This is only for ripping to CD, you understand.
On the PC they are like yours - files in a folder.
>Messing with the 'vinyl digitisations' is no chore to me - I get to listen
>(over and over, if I want or just let a track run) as I cut up the 'sausage
>string', trim the individual tracks to the right 'lead in' and lead out'
>lengths, add fades where I want and mute intertrack 'dead wax' surface noise
>so the **quiet bits aren't noisy**!!
>
>(Nutter Allison, are you reading this? :-)
>
d
David Looser
>
> When I digitize vinyl I tend not to trim (except at the finish of a
> side, I leave the needle drop in place).
Er... why? Isn't one of the nice things about a CD copy that you can get rid
of the thump as the needle hits the disc and the "snap crackle and pop" of
the lead in?
David.
Don Pearce
No, I digitize vinyl purely for fun. I can always get hold of clean
copies other ways.
d
Iain Churches
>
I agree. Nothing succeeds like success:-)
Remember the famous Liberace quote: Once I used to go laughing all
the way to the bank. Now I own it"
Iain Churches
"Jim Lesurf" <no...@audiomisc.co.uk> wrote in message
news:50b6744...@audiomisc.co.uk...
> FWIW what prompted this was someone saying it was a good idea to always
> normalise so the max came to -0.5dBFS. I was then pointing out this could
> be a mistake if you only looked at the sample values - for reasons shown
> on
> the page I reference above. The question then became, what does CEP
> actually display? Does it show the user a waveform that would allows them
> to see if this problem was causing their output to exceed 0dBFS or not for
> arbitrary waveforms?
>
Jim.
Personally, I would not trust software like CEP at anywhere close to OdBFS.
I have seen .wav files made on CEP (which the person who made them claims
are
*clean*) that are considerable clipped when uploaded to to mastering DAW.
Iain
Don Pearce
Finger trouble, I think, Iain. I've been using first CEP then Audition
for a long time and I've never seen anything like that.
d
Iain Churches
>
> "Iain Churches" <Iai...@kolumbus.fi> wrote in message
> news:i0yIm.35635$La7....@uutiset.elisa.fi...
>>
>> "Brian Gaff" <Bri...@blueyonder.co.uk> wrote in message
>> news:ZCYHm.1341$Ym4...@text.news.virginmedia.com...
>>> Well, The two pianos of.. etc, were a thing of their time. Made in the
>>> days of radiograms to show off with, never mind realism. I think you
>>> have
>>> forgotten that there are times when exciting stereo can be fun. Its not
>>> realism, of course its not,
>>
>>>
>>
>> The Two Pianos Of... and recordings of that ilk, were very much a
>> fashion statement, as is much popular music. I worked on many
>> of the Phase Four recordings at Decca.
>>
>> After the initial stereo showcase recordings, many of them were
>> multi-microphone multitrack productions. And very popular
>> they were too:-) No one pretended it had anything to do with
>> realism.
>
>
> The trouble with 'stereo' is that it encourages the geeks who think they
> can replicate a real live experience in their own living rooms like '3D
> depth/spatiality' is *everything*!! (Even now there are twats in here who
> will blather on about 'proper stereo'!!)
Listening to well-recorded stereo material can be a very exciting
experience.
>
> Once the 'audio ping pong' trick was played out I suspect Joe Ordinaire
> just ignored it and listened to the music. A 'stereo' label on a record
> doesn't particularly excite me - like I have said often before, I guess
> I'm too 'off axis' to care much of the time!!
Many good stereo recordings, especially multi-mic multitrack productions
often offer something new at each listening (well not new, because it has
been there all the time, but something that you had not noticed previously)
One can use "stereo" to greast effect. For example a single tambourine hit
-on the left _ can be sent to a digital delay, repeated with a multi delay
attenuating by 3dB at each hit, and brought back centre-left panning
across to the right. It works a lot better if it is neither too loud or
obvious.
For that reason I greatly enjoyed listening in depth to band like Queen.
The Devil is indeed in the detail.
> Oh yes and I don't care how 'primitive' the recordings are by today's
> standards (?? are you kidding??) - all part and parcel of the experience
> for me!
It's refreshing to hear someone say that.
>
> (Caruso had a little tinny voice in real life, didn't he? :-)
Rumour has it that his re-incarnation frequently stood in for Kate Bush
Iain
Iain Churches
If it had been overshot a couple of dB with the traditional
10 or even 6dB headroom no harm would have been done. As
it was, his material was rejected.
Iain Churches
> On Fri, 6 Nov 2009 21:12:14 +0200, "Iain Churches"
> <Iai...@kolumbus.fi> wrote:
>
>>> Is that so? I didn't know it was unique. I really only use Audition
>>> because I have sort of grown up with it throughout its CoolEdit
>>> incarnations, and I now use it more or less by instinct.''
>>
>>But AFAIK this Centre Channel Extractor does not exist in CEP Pro
>>(or at least in the beta testers version that I am familiar with)
>
> Did it only appear once Adobe bought it? I'm glad they did something
> more useful than just making the interface "pretty".
It seems so. There were several Beta versions of CEP but none AFAIK
have the centre channel extractor. Under what pull down menu is it to be
found on Audition?
This is not commonly found on other audio software either.
I think that some versions of ProTools can do it.
Iain
Don Pearce
<Iai...@kolumbus.fi> wrote:
>
>"Don Pearce" <sp...@spam.com> wrote in message
>news:4af4ae7f....@news.eternal-september.org...
>> On Fri, 6 Nov 2009 21:12:14 +0200, "Iain Churches"
>> <Iai...@kolumbus.fi> wrote:
>>
>>>> Is that so? I didn't know it was unique. I really only use Audition
>>>> because I have sort of grown up with it throughout its CoolEdit
>>>> incarnations, and I now use it more or less by instinct.''
>>>
>>>But AFAIK this Centre Channel Extractor does not exist in CEP Pro
>>>(or at least in the beta testers version that I am familiar with)
>>
>> Did it only appear once Adobe bought it? I'm glad they did something
>> more useful than just making the interface "pretty".
>
>It seems so. There were several Beta versions of CEP but none AFAIK
>have the centre channel extractor. Under what pull down menu is it to be
>found on Audition?
>
>This is not commonly found on other audio software either.
>I think that some versions of ProTools can do it.
>
>Iain
>
>
Effects then Stereo Imagery
d
Laurence Payne
If you leave them in, the distortions of vinyl are less noticeable
afterwards :-)
Chris Isbell
Oh the power of the mind. The psychology of perception is a very
interesting subject and shows just how easily we can be misled.
--
Chris Isbell
Southampton, UK
Jim Lesurf
<sp...@spam.com> wrote:
> On Sat, 07 Nov 2009 14:24:16 +0000 (GMT), Jim Lesurf
> <no...@audiomisc.co.uk> wrote:
[snip lots]
> >
> >Another reason is to bear in mind that a waveform composed of a series
> >of harmonics can - with them in phase - generate very sharp peaks with
> >relatively little HF content. (And if you look at other measurements
> >these peaked waveforms certainly arise for some instruments like violin
> >or trumpet.)
> >
> OK, let me think about that. You're probably right, because you
> obviously have thought about it - which has prompted your posts.
Or at least worried at this problem for some time! :-)
The difference isn't large in many practical cases *if* the user is
cautious and avoid scaling sample values into the region around or about
-2dBFS. But the problem seems to be that people *do* alter values in this
range. So may cause DAC problems they are unaware of.
Hard clipping is essentially an 'all or nothing' problem. So best avoided
entirely unless inevitable. And in this context less than a dB can make
the difference.
What I haven't (yet) done is compare 'sinusoidal spline fit' with sinc
to see how much they differ as I hadn't thought anyone would use that.
Have looked at quadratic interpolation and the distortion that causes,
though.
The problem with sine fitting is that it probably works very well indeed
*when* the input series if for a sinusoid! So may look great when tested
that way. But then not work well for other more general waveforms, like
music!
Actually I should have twigged this ages ago when I noticed that one
of the options with the ALSA (Linux) sound system is for resampling
to use a 'sine' fit. Should have rung warning bells, but I just decided
not to do that an find hardware which didn't need to resample at all.
Thus avoiding any errors due to resampling. Better if you can to avoid
processes that can introduce problems than to worry about minimising the
problems the processes can cause.
> >
> >The problem with a programmer treating this as an exercise in 'getting
> >a smooth fit' is that this is *not* the basis in Information Theory. So
> >the result of a correct reconstruction may look 'less smooth' but
> >actually be the waveform the samples define.
> >
> >I'll ask the person who said CEP does use sinc if this is stated in the
> >documentation for the program.
> >
> Remember that CEP started in days when processor power was at a bit of a
> premium, and a programmer couldn't afford to waste cycles on errors
> which were probably going to be contained within the width of the trace
> for 99.99% of users.
Yes. That's fair enough. Indeed, I also often plot/display with 'join up
the dots' as it is a quick way to visualise the data sample series. It is a
quick way to see what you have.
Indeed, I have a feeling that many people didn't realise that the data can
generate peaks above all the samples until quite recently.
However I've known about the way this is a bit like a map with blank areas
on which should be stamped "Keep clear! Here be dragons!" :-) This was as
much as anything from working with data in contexts that are nowt to do
with audio. So with any recordings I make I avoid every getting any samples
above -4dBFS and then avoid scaling upwards. For me this is just good
practice.
But following the thread elsewhere (uk.tech.digital-tv) it has made me
think some people will be falling into the trap because the software
doesn't display the consequences properly and people don't know about the
problem. And some seem to feel you have to scale up to maximise the of all
the bits available, particularly when then using the result for lossy
reduction systems. Which might make the situation *worse* not better!
Slainte,
Jim Lesurf
FWIW in essence that is also what I do when making recordings. So when a
long recording has been divided into tracks the output series of samples is
the same as if playing the original recording.
Jim Lesurf
<Iai...@kolumbus.fi> wrote:
> Jim.
That seems to support my feeling that a lack of the actual reconstructed
shape being displayed means the user can easily be blissfully unaware of
the problem they create when they scale up the loudness to get as close to
0dBFS as they can.
I am also wondering if the CEP waveform display is even a sinusoidal
spline. Might just be something simple like a cubic as that is trivial and
quick for a programmer. Just reach for your copy of 'Recipies in
<Language>' then copy the required routines into your program.
Jim Lesurf
<sp...@spam.com> wrote:
> On Sat, 7 Nov 2009 20:51:31 +0200, "Iain Churches" <Iai...@kolumbus.fi>
> wrote:
> >
> >Personally, I would not trust software like CEP at anywhere close to
> >OdBFS. I have seen .wav files made on CEP (which the person who made
> >them claims are *clean*) that are considerable clipped when uploaded
> >to to mastering DAW.
> Finger trouble, I think, Iain. I've been using first CEP then Audition
> for a long time and I've never seen anything like that.
Perhaps because you are aware that it is wise to keep the signal levels
down and avoid any samples going above about -4dBFS.
This issue was publicised for professional audio engineers a few years ago
when some papers on it appeared in AES journals and conferences. But I
continue to wonder how many people working with audio are totally unaware
of these problems. Particularly given the established religious belief that
'louder is better'!
Jim Lesurf
> <Iai...@kolumbus.fi> wrote:
> >Personally, I would not trust software like CEP at anywhere close to
> >OdBFS. I have seen .wav files made on CEP (which the person who made
> >them claims are *clean*) that are considerable clipped when uploaded
> >to to mastering DAW.
> Finger trouble, I think, Iain. I've been using first CEP then Audition
> for a long time and I've never seen anything like that.
I've been wondering if you might like to help with an experiment to check
this out. If I were to put up a couple of short LPCM WAV files of, say, a
'waveform from hell' or an offset impulse, could you look at them with CEP
and find the peaks in the zoomed in reconstruction waveforms?
I can predict what actual peak levels to expect. So if what you see agrees
with that it will tend to support that CEP can show a decent waveform. But
if the peak falls short it will show that it isn't reliable as a display
for overshoots.
Don Pearce
<no...@audiomisc.co.uk> wrote:
>In article <4af6c452....@news.eternal-september.org>,
> Don Pearce <sp...@spam.com> wrote:
>> On Sat, 7 Nov 2009 20:51:31 +0200, "Iain Churches"
>> <Iai...@kolumbus.fi> wrote:
>
>> >Personally, I would not trust software like CEP at anywhere close to
>> >OdBFS. I have seen .wav files made on CEP (which the person who made
>> >them claims are *clean*) that are considerable clipped when uploaded
>> >to to mastering DAW.
>
>
>> Finger trouble, I think, Iain. I've been using first CEP then Audition
>> for a long time and I've never seen anything like that.
>
>I've been wondering if you might like to help with an experiment to check
>this out. If I were to put up a couple of short LPCM WAV files of, say, a
>'waveform from hell' or an offset impulse, could you look at them with CEP
>and find the peaks in the zoomed in reconstruction waveforms?
>
>I can predict what actual peak levels to expect. So if what you see agrees
>with that it will tend to support that CEP can show a decent waveform. But
>if the peak falls short it will show that it isn't reliable as a display
>for overshoots.
>
>Slainte,
>
>Jim
Yes of course. Go ahead - I'll be interested.
d
Don Pearce
What Iain is describing here is, I think probably not related to the
visual display. When setting levels it is not usual practice to look
at the screen and judge when the waveform is close enough to the peak.
You use the maths within the programme and choose your peak level that
way. Then you will listen to what you have created before saving it.
If a job reaches the mastering DAW with gross clipping as Iain
describes, then that procedure has not been followed and something
nasty has gone on. At a guess I would say that all the work was done
in floating point, which is a good idea as you don't have to dither
intermediate stages (that noise can build up), then the final
requantisation was done without regard to the fact that the peaks were
well above FS. Just my guess.
d
Jim Lesurf
<sp...@spam.com> wrote:
> On Sun, 08 Nov 2009 10:03:34 +0000 (GMT), Jim Lesurf
> <no...@audiomisc.co.uk> wrote:
> >I've been wondering if you might like to help with an experiment to
> >check this out. If I were to put up a couple of short LPCM WAV files
> >of, say, a 'waveform from hell' or an offset impulse, could you look at
> >them with CEP and find the peaks in the zoomed in reconstruction
> >waveforms?
> Yes of course. Go ahead - I'll be interested.
OK. I've now put a small zip at
http://jcgl.orpheusweb.co.uk/temp/WaveFromHell.zip
This should contain two 'CDDA format' wav files, each of about 6MB and 35
seconds duration. Details as below. However because the data in the files
is highly periodic (no dithering) they compress down into a 53K zip. The
data essentially consists of the same 140 pairs of values over and over as
the waveform cycles.
N.B. I've created these in a quick and rough way so I'm not certain they
are correct. They looked OK when I checked them. BUT I have only recently
started doing new versions of my audio file creation/processing/analysis
software to be able to handle Wave formats. Until a few months ago it was
convenient to use a raw data format for my audio files. But since starting
to use Linux alongside RISC OS I've started producing new versions of my
existing software that can work with Wave files. I've also started doing
Linux versions of the main applications.[1]
That said the two files are both in stereo 44.1k format. They also have a
top and tail of a few seconds of zeros (silence). One file then has a '0dB'
mono (i.e. both channels the same) of the 'Waveform From Hell'. This should
probably generate the largest out-of-range spikes you've ever seen for LPCM
data. :-) Note I've fractionally underscaled to avoid actually getting the
genuine sample values for max to avoid problems with any software that
doesn't understand that sample values aren't normal integers so flips the
sign for one range max value. IIRC I went under max by one bit, but my
memory on that may be wrong as I wrote the code to generate the waveform
some time ago.
The other file has the waveform reduced by 10dB and is in antiphase to make
the two waveforms show up when displayed. (You may fined the mono file only
shows as one line as the channels are identical.)
If you play these though an audio system I have to wash my hands of any
responsibility for damage! :-) Tweeters in particular may not like them.
Nor a poor amp driving a capacitative load.
Hope the above files are OK. If not, I'll have another go at making decent
files.
Cheers,
Jim
[1] If anyone is interested these will all be put on the website for free
use and contain all the 'C' source code so people can examine how they work
to look for bugs, etc.
Don Pearce
<no...@audiomisc.co.uk> wrote:
I've had a look, and again created a couple of flash screen dumps so
you can see how CEP (actually Audition) handles the various phases of
zoom.
http://81.174.169.10/odds/WFH0dBMono.html
http://81.174.169.10/odds/WFH-10dBAntiphase.html
d
Jim Lesurf
<sp...@spam.com> wrote:
> What Iain is describing here is, I think probably not related to the
> visual display. When setting levels it is not usual practice to look at
> the screen and judge when the waveform is close enough to the peak. You
> use the maths within the programme and choose your peak level that way.
> Then you will listen to what you have created before saving it.
OK.
> If a job reaches the mastering DAW with gross clipping as Iain
> describes, then that procedure has not been followed and something nasty
> has gone on. At a guess I would say that all the work was done in
> floating point, which is a good idea as you don't have to dither
> intermediate stages (that noise can build up), then the final
> requantisation was done without regard to the fact that the peaks were
> well above FS. Just my guess.
That is also quite interesting. Although to be honest I'd say that
dithering is still theoretically needed even for floating point since the
values are still quantised into a finite set. Problem then is that dither
and noise shaping become more compilicated as you have a NICAM-like
process to deal with.
But if they are using something like IEEE doubles this isn't likely to be
much of a problem in reality!
FWIW all the internal calculations in my own programs tend to use IEEE
double unless the process is trivial enough not to need this. This is
because I have in the past found that even single float isn't enough to
avoid some artifiact problems for some cases.
In fact, when doing an analysis of SACD a few years ago I was starting to
think that even normal doubles weren't enough to examine the problems with
DSD. However since SACD essentially faded away I decided that not many
people would have cared about the problems it had anyway! :-)
Keith G
??
Hey, hijack the thread and take it where you want; that's normal for Usenet
newsgroups, but I'd suggest it's unsafe to presume the OP (me) is
automatically interested where you go with it - my interest in the
comparative resolution available in the editing softwares mentioned is not,
in this instance, anything to do with sound levels.
Keith G
Now it's probably billions in debt and *we* get to own it, but the best
thing I remember about Liberace was when he demonstrated
'hemidemisemiquavers' on the piano - quite stunning!
Don Pearce
wrote:
>
I think your original bit of the thread is long since answered and put
to bed. It has, as threads do, taken on a new life and new direction
which was prompted by your original one. So don't think of this as
hijacking, which it would have been if it had happened at - say - post
number 2. Rather it is evolution and metamorphosis.
d
Arny Krueger
> In article
> <4af4ae7f....@news.eternal-september.org>, Don
> Pearce <sp...@spam.com> wrote:
>> On Fri, 6 Nov 2009 21:12:14 +0200, "Iain Churches"
>> <Iai...@kolumbus.fi> wrote:
>
>>>> Is that so? I didn't know it was unique. I really only
>>>> use Audition because I have sort of grown up with it
>>>> throughout its CoolEdit incarnations, and I now use it
>>>> more or less by instinct.''
>
>>> But AFAIK this Centre Channel Extractor does not exist
>>> in CEP Pro (or at least in the beta testers version
>>> that I am familiar with)
>> Did it only appear once Adobe bought it? I'm glad they
>> did something more useful than just making the interface
>> "pretty".
Interesting. I own Audition 2.0 but never adapted to the new interface well
enough to shift over to using it as my daily driver.
> Pardon me for hijacking this thread, but the mentions of
> CEP prompt me to ask a question about it.
> I don't use CEP or know anything about how it works.
> However in a thread on a couple of tv/broadcasting
> technical groups I've been discussing the problem of
> intersample peaks that can produce 'overshoots' that can
> go above 0dBFS of someone scales up the samples to be too
> close to 0dBFS.
True.
This can happen in the real world, too.
> I've been told that CEP shows the shape inbetween samples
> if you 'zoom in' and that it uses an approx to the
> formally correct sinc function to do this.
I know of no authoritative discussions of CEP or Audition internals.
Keith G
> On Sat, 7 Nov 2009 16:29:30 -0000, "Keith G" <k...@moirac.adsl24.co.uk>
> wrote:
>
>>
>>"Don Pearce" <sp...@spam.com> wrote in message
>>> On Sat, 7 Nov 2009 15:40:16 -0000, "Keith G" <k...@moirac.adsl24.co.uk>
>>> wrote:
>>>
>>>>
>>>>"Don Pearce" <sp...@spam.com> wrote
>>>>
>>>>> So-many-to-1 just isn't relevant as a figure. What you see at what
>>>>> zoom will depend on how long the original bit of music was.
>>>>
>>>>
>>>>Hmm, that's not my understanding of how the programme works but,
>>>>whatever,
>>>>it's academic to me - zoomimg in and out is simply a question of
>>>>spinning
>>>>the (mouse) wheel back and forth to show me what amd how much I want to
>>>>see
>>>>on the screen. It's very fast and easy to do in SF; the only constraint
>>>>is
>>>>that the zoom factor has to be 1:32, or bigger, to be able to use the
>>>>pencil
>>>>tool.
>>>>
>>>>
>>>> What you
>>>>> need from your software is what Audition tells you down the bottom.
>>>>> The exact start and stop times of the visible window and whatever you
>>>>> have selected inside it. You can see that easily without the box when
>>>>> it is zoomed in far enough to show the sample points.
>>>>>
>>>>> Now, as for that second edit waveform, I'm afraid it shows the
>>>>> limitations of Sound Forge.
>>>>
>>>>
>>>>Like how? (Not that I GAS - I'm not selling it for a living or
>>>>anything...??)
>>>>
>>>>
>>>
>>> Because what you see on the screen bears only the most passing
>>> resemblance to what emerges from the DAC.
>>>
>>>>
>>>> And in Audition, I don't think anyone
>>>>> would ever bother to "paint" out a click.
>>>>
>>>>
>>>>Quick and easy to do in SF but I'm picking up Izotope RX in a little
>>>>while
>>>>and that may well alter the way I do things - my only interest in 'sound
>>>>editing' atm is cutting whole side/whole disc LP recordings into tracks,
>>>>trimmimg them and removing bothersome pops and clicks. If Izotope can do
>>>>that well enough ('de-clicking' software I tried in the misty past was
>>>>pretty much NFG) it will alter the way I do things and obviously speed
>>>>up
>>>>the workflow....
>>>>
>>>>(Except that I don't *need* the workflow speeded up....???)
>>>>
>>>>
>>>
>>> When I digitize vinyl I tend not to trim (except at the finish of a
>>> side, I leave the needle drop in place).
>>
>>
>>
>>you later' lift-off sound!! :-)
>>
>>
>>
>> I leave the inter-track
>>> spaces exactly as they are, and just drop in zero delay track markers
>>> to separate them. That way I can play what sounds exactly like a whole
>>> side of the original lp.
>>
>>
>>can be found in a search. I frequently know a track I want to hear but
>>don't
>>know/can't remember which album it's from!! (An album is a folder with
>>individual track files in it; if I want to play the whole disc I just hit
>>'Play All' - I cba with playlists and the like!)
>>
>
> The tracks are still numbered, named and findable. It is just that
> when you play the whole thing you don't hear the joins. It sounds
> exactly like the LP.
If I want to hear 'exactly like the LP' - I play the LP..!! ;-)
This is only for ripping to CD, you understand.
Why waste time with 'hyphen technology'? - You need one of these:
http://www.brennan.co.uk/home/
> On the PC they are like yours - files in a folder.
>
>>Messing with the 'vinyl digitisations' is no chore to me - I get to listen
>>(over and over, if I want or just let a track run) as I cut up the
>>'sausage
>>string', trim the individual tracks to the right 'lead in' and lead out'
>>lengths, add fades where I want and mute intertrack 'dead wax' surface
>>noise
>>so the **quiet bits aren't noisy**!!
>>
>>(Nutter Allison, are you reading this? :-)
>>
>
> d
Arny Krueger
> "Jim Lesurf" <no...@audiomisc.co.uk> wrote in message
> news:50b6744...@audiomisc.co.uk...
>
>> FWIW what prompted this was someone saying it was a good
>> idea to always normalise so the max came to -0.5dBFS. I
>> was then pointing out this could be a mistake if you
>> only looked at the sample values - for reasons shown on
>> the page I reference above. The question then became,
>> what does CEP actually display? Does it show the user a
>> waveform that would allows them to see if this problem
>> was causing their output to exceed 0dBFS or not for
>> arbitrary waveforms?
> Personally, I would not trust software like CEP at
> anywhere close to OdBFS.
The voice of inexperience speaks!
I've used CEP extensively for the better part of a decade, for recording and
technical testing. It is generally trustworthy - certainly more trustworthy
than many of the people who denigrate it.
> I have seen .wav files made on
> CEP (which the person who made them claims are
> *clean*) that are considerable clipped when uploaded to
> to mastering DAW.
A person with common sense would assign the probable blame to the person
making the assertions.
Arny Krueger
> If it had been overshot a couple of dB with the
> traditional 10 or even 6dB headroom no harm would have
> been done.
There is no headroom in the digital domain.
> As it was, his material was rejected.
Interesting that you would publicly criticize a well-known software product
based on a producer whose recordings you wouldn't trust.
Don Pearce
wrote:
>> The tracks are still numbered, named and findable. It is just that
>> when you play the whole thing you don't hear the joins. It sounds
>> exactly like the LP.
>
>
>
>If I want to hear 'exactly like the LP' - I play the LP..!! ;-)
>
But each time you play it, there is a new collection of pops. Ripping
to CD you preserve it as good as it is.
>
> This is only for ripping to CD, you understand.
>
>
>Why waste time with 'hyphen technology'? - You need one of these:
>
>http://www.brennan.co.uk/home/
>
>
>
No, my way I keep all my options open.
d
Arny Krueger
> In article
> <4af6c452....@news.eternal-september.org>, Don
> Pearce <sp...@spam.com> wrote:
>> On Sat, 7 Nov 2009 20:51:31 +0200, "Iain Churches"
>> <Iai...@kolumbus.fi> wrote:
>
>>>
>>> Personally, I would not trust software like CEP at
>>> anywhere close to OdBFS. I have seen .wav files made on
>>> CEP (which the person who made them claims are *clean*)
>>> that are considerable clipped when uploaded to to
>>> mastering DAW.
>
>
>> Finger trouble, I think, Iain. I've been using first CEP
>> then Audition for a long time and I've never seen
>> anything like that.
>
> Perhaps because you are aware that it is wise to keep the
> signal levels down and avoid any samples going above
> about -4dBFS.
True at tracking time. However, I often allow even more headroom - more like
10 dB. I find that musicians just aren't that reliable or predictable. And
when there is a live audience present, I can pretty much guarantee that
they will perform up to 10 dB louder.
For finished work, I have no problem with peaks going up to -1 dBFS.
I've tested literally 100s of DACs. While many of them are very clean at - 1
dBFS, some won't cleanly handle -0.1 dBFS or even -0.5 dBFS.
Arny Krueger
>
> Effects then Stereo Imagery
"Stereo Imagery" is not in CEP 2.1 Effects. That was the last version of
CEP.
I wonder in what release Audition that was first added?
Don Pearce
wrote:
Unfortunately I have used it so very rarely that I couldn't even guess
when it first appeared.
d
Jim Lesurf
<sp...@spam.com> wrote:
> >
> >OK. I've now put a small zip at
> >
> >http://jcgl.orpheusweb.co.uk/temp/WaveFromHell.zip
> I've had a look, and again created a couple of flash screen dumps so you
> can see how CEP (actually Audition) handles the various phases of zoom.
> http://81.174.169.10/odds/WFH0dBMono.html
> http://81.174.169.10/odds/WFH-10dBAntiphase.html
Thanks, Don. :-)
That does look like CEP is doing a good job of generating the intersample
waveforms when it can. In principle the waveform should give a peak factor
of 1.8 or 5.1dB. i.e. the 0dB waveform should produce +5.1dBFS peaks and
the -10dB produce peak at -4.9dBFS.
Provided you zoom in enough, the display seems to show +5dBFS and -5dBFS
so far as I can see it. Which is close enough given the size of the
displays, and should be fine for avoiding overshoots. So it does look like
a good approximation to the ideal sinc process.
However I'm less happy with lower zooms as the display then often seems to
miss samples and so not all peaks go up to the correct level. Hence if
someone is using a display showing a duration more than a small fraction of
a second then they could miss peaks that weren't visible.
As broad rule then I'd guess that level adjustments would be 'safe' if you
have zoomed in to be able actually distinguish one sample value from its
neighbours. But with a wider view you might miss peaks.
Slainte,
Jim
Keith G
Fully aware of all that Don and I think I made the point clearly enough
before you took the trouble to repeat it back to me. When I start a thread
in this group I do sometimes see it as something of starting a 'ball
rolling' exercise (or 'trolling' as Nutter Allisson would prefer to call it)
and am pleased when it creates a bit of traffic in an otherwise barren
place. (You should try it sometime...)
What I am saying is that is is unsafe to automatically assume anyone from
the origins of the thread has bothered to follow the the new direction -
especially if I'm the OP and the thread has been typically dragged down to
usual mithering, sub-molecular level and has become littered with the word
'problem'....
Don Pearce
I think that pretty well sums it up. The next question is the degree
of similarity between your waveform from hell, and music. If this
stuff really only happens in a transient, it doesn't matter if there
is some clipping, however if it is there in extended instrumental
notes it could be a big problem.
d
Keith G
> On Sun, 8 Nov 2009 12:17:55 -0000, "Keith G" <k...@moirac.adsl24.co.uk>
> wrote:
>
>>> The tracks are still numbered, named and findable. It is just that
>>> when you play the whole thing you don't hear the joins. It sounds
>>> exactly like the LP.
>>
>>
>>
>>If I want to hear 'exactly like the LP' - I play the LP..!! ;-)
>>
>
> But each time you play it, there is a new collection of pops. Ripping
> to CD you preserve it as good as it is.
Hmmm, *mastertape direct to MP3* thinking?
When I play a record, I play a record - fuck the 'pops'...
Don Pearce
wrote:
I can't agree with your assessment of "dragged down", that is a
gratuitous judgment that what isn't interesting to you is somehow
worthless. That isn't either nice nor necessary. If you have got what
you need from the thread, and it has taken a direction that doesn't
hold your interest, just let it go. I don't think that anyone was
assuming that you had bothered to follow it, although it is clear that
you had because you did contribute a couple of useful screen shots
showing how Sound Forge handled the matter (or didn't handle it, as it
turned out). Maybe Jim should have started a new thread, but my
opinion is that his direction did emerge quite naturally from what we
had seen in the previous posts, so I'm fine with it.
d
Keith G
You are entitled to your opinions...
If you have got what
> you need from the thread, and it has taken a direction that doesn't
> hold your interest, just let it go. I don't think that anyone was
> assuming that you had bothered to follow it, although it is clear that
> you had because you did contribute a couple of useful screen shots
> showing how Sound Forge handled the matter (or didn't handle it, as it
> turned out).
No, you were at cross-purposes here trying to answer two different lines of
enquiry - having seen your link, I merely posted my pictures to compare a
similar waveform in Sound Forge from an (apparent) *resolution* POV as I
thought the Audition waveform looked far too 'smoothed out' and with
considerably less 'sample points' (if that is what they were) - nothing to
do with sound levels at all....
Don Pearce
wrote:
>No, you were at cross-purposes here trying to answer two different lines of
>enquiry - having seen your link, I merely posted my pictures to compare a
>similar waveform in Sound Forge from an (apparent) *resolution* POV as I
>thought the Audition waveform looked far too 'smoothed out' and with
>considerably less 'sample points' (if that is what they were) - nothing to
>do with sound levels at all....
>
OK, no. Audition had precisely the same number of sample points (44100
per second in both our cases), and the "smoothing out" of Audition has
turned out to be more or less precisely correct, even with a
pathological signal. What Audition shows is what the DAC provides to
an amplifier.
d
Jim Lesurf
Not quite. It shows what Audition *should* provide *if* the DAC can deal
with overshoots. Measurements on real DACs show examples that are clearly
unable to do that. Even for overshoots much smaller than the one in the
test waveform that deliberately sets out to be an extremely demanding case.
Jim Lesurf
Krueger
<ar...@hotpop.com> wrote:
> "Iain Churches" <Iai...@kolumbus.fi> wrote in message
> news:YkjJm.36688$La7....@uutiset.elisa.fi
> > "Jim Lesurf" <no...@audiomisc.co.uk> wrote in message
> > news:50b6744...@audiomisc.co.uk...
> >
> >> FWIW what prompted this was someone saying it was a good idea to
> >> always normalise so the max came to -0.5dBFS. I was then pointing out
> >> this could be a mistake if you only looked at the sample values - for
> >> reasons shown on the page I reference above. The question then
> >> became, what does CEP actually display? Does it show the user a
> >> waveform that would allows them to see if this problem was causing
> >> their output to exceed 0dBFS or not for arbitrary waveforms?
> > Personally, I would not trust software like CEP at anywhere close to
> > OdBFS.
> The voice of inexperience speaks!
Well, I've never used CEP so have no experience at all with it. However the
results shown by Don's Flash movies apparently show it misses peak values if
you don't zoom in sufficiently. So would be unwise to trust that it was
showing you what you'd need to assess peaking if you don't zoom in and only
look at a short section at a time.
> I've used CEP extensively for the better part of a decade, for recording
> and technical testing. It is generally trustworthy - certainly more
> trustworthy than many of the people who denigrate it.
I can't really comment on what others may have said about it.
From this examination it looks to me to work quite well for the purpose of
assessing peaks *provided* you zoom in sufficiently. And have the
understanding to give yourself enough headroom to deal with any overshoots
you may have not seen. So for someone who know what they are doing, I'd
agree with you. But...
> > I have seen .wav files made on CEP (which the person who made them
> > claims are *clean*) that are considerable clipped when uploaded to to
> > mastering DAW.
> A person with common sense would assign the probable blame to the person
> making the assertions.
Whilst I'd agree I have some reservations. Ideally everyone who uses any
soiftware knows *exactly* how it works, and the details of every process it
performs. And knows enough about all the relevant topics in IT, DSP, etc,
to assess what that all means for the tasks. But in practice I doubt most
people ever do meet this set of requirements.
I'm personally ultra-cautious about processing data because my own
professional interest over the years has mainly been in high performance
measurement, etc. So it is usually necessary to write or check all code and
know what it does. This can be a slow and tedious approach. So I can
understand why people simply use packages that others also seem happy with,
without knowing what they do under the GUI.
And in this case, so far as I could tell from asking on more than one group
where sound editing is relevant, no-one could point me at any documents or
sources that specified what CEP *actually* did to generate 'waveforms' from
the samples, etc. At least one person though it fitted 'sinusoids' in
between the samples. Presumably as a spline fit.
This was also (elsewhere) in the context of using it prior to sending the
LPCM though a low bitrate lossy encoder. Which can then alter the waveforms
and change the peaking. Again with no clear sign that those I'd been taking
to knew the details of what that would do.
Given all that, I decided to find out. Hence my starting to discuss it
here where I thought some people would also have an interest and the
ability to get involved. Which they did. :-)
The outcome seems to me quite a decent one for CEP and its users. It seems
to give a very reliable indication of peaking *provided* it is used
appropriately. The snag then being, how many users know it might *not* do
so if you don't zoom in enough? I can't say as I never use it, so have to
leave that question to others.
I'd agree that bad workmen should not blame their tools. But I do think
that good workmen need to really understand how their tools work, and
to be able to judge how to use them (and when not to).
I have already made an analogy with this in another group by referring
to desktop home publishing. This has given many people the ability
to easily generate superbly well laid out documents that are highly
readable and informative. But it has also give other the ability to
produce eye-wateringly nasty documents that are pain to have to try
and read.
Jim Lesurf
<sp...@spam.com> wrote:
> On Sun, 8 Nov 2009 12:39:36 -0000, "Keith G" <k...@moirac.adsl24.co.uk>
> wrote:
> >
> >What I am saying is that is is unsafe to automatically assume anyone
> >from the origins of the thread has bothered to follow the the new
> >direction - especially if I'm the OP and the thread has been typically
> >dragged down to usual mithering, sub-molecular level and has become
> >littered with the word 'problem'....
> >
> I can't agree with your assessment of "dragged down", that is a
> gratuitous judgment that what isn't interesting to you is somehow
> worthless. That isn't either nice nor necessary.
Fairly predictable from Keith G given his track record in this group,
though. If something doesn't interest him or he doesn't understand it his
reaction is to dismiss it in terms like the above.
> Maybe Jim should have started a new thread, but my opinion is that his
> direction did emerge quite naturally from what we had seen in the
> previous posts, so I'm fine with it.
It did link with what had already been said, and I have since started a new
thread. Where I think we have now arrived at a conclusion that I've found
very helpful. :-)
Jim Lesurf
<sp...@spam.com> wrote:
> On Sun, 08 Nov 2009 12:30:00 +0000 (GMT), Jim Lesurf
> <no...@audiomisc.co.uk> wrote:
> I think that pretty well sums it up. The next question is the degree of
> similarity between your waveform from hell, and music. If this stuff
> really only happens in a transient, it doesn't matter if there is some
> clipping, however if it is there in extended instrumental notes it could
> be a big problem.
Yes. The test waveform is quite deliberately intended to be the ultra
extereme case. I doubt you'd stand much chance of finding it on a normal
music CD. But I am not sure of that given some of the nastier CDs I've
encountered. Something like it might be out there somewhere, albeit a
rare beast.
So here the waveform was just to check and see how well CEP did at getting
the correct result. Which in effect it did quite nicely.
The problem is that the initial context[1] was the way some users choose
to level compress so that the waveforms tend to reach the region where the
max samples are in the range around -0.5dBFS or above. Thus we may be
dealing with 'music' which has been 'improved' to the point of being near
clipping a lot of the time.
In that situation we would only need intersample peaks that go another
0.5dB or so to reach peaks above 0dBFS. So you wouldn't need patterns
as remarkable as the 'Wave from Hell' to do this.
An additional problem is that the result is then fed to lossy compression
systems with low or modest bitrates. That alters the waveforms, and can
invoke overshoots which aren't in the LPCM that was created using CEP or
the equivalent used!
Alas for reasons I mention elsewhere I am less sure that most DACs do
as good a job with out-of-range overshoots than CEP. People generally
don't check this, and the measurements I've seen don't fill me with
confidence.
Slainte,
[1] The discussion elsewhere that prompted me to raise the matter here
in the hope (now justified) that here it could be properly investigated.
Jim Lesurf
Krueger
<ar...@hotpop.com> wrote:
> "Jim Lesurf" <no...@audiomisc.co.uk> wrote in message
> news:50b6ea5...@audiomisc.co.uk
> > In article <4af6c452....@news.eternal-september.org>, Don Pearce
> > <sp...@spam.com> wrote:
> >> Finger trouble, I think, Iain. I've been using first CEP then
> >> Audition for a long time and I've never seen anything like that.
> >
> > Perhaps because you are aware that it is wise to keep the signal
> > levels down and avoid any samples going above about -4dBFS.
> True at tracking time. However, I often allow even more headroom - more
> like 10 dB. I find that musicians just aren't that reliable or
> predictable. And when there is a live audience present, I can pretty
> much guarantee that they will perform up to 10 dB louder.
Yes, I can see that makes good sense! :-) I was referring to contexts
where the source already has a decently defined maximum.
> For finished work, I have no problem with peaks going up to -1 dBFS.
That is fine if talking about the levels including intersample peaks. I'd
be more wary if just talking about sample values in case it produced
overshoots between samples. For the practical reasons I mention below.
> I've tested literally 100s of DACs. While many of them are very clean at
> - 1 dBFS, some won't cleanly handle -0.1 dBFS or even -0.5 dBFS.
I'd be *very* interested in seeing what a series of DACs did with the
"Wave from Hell" if you'd like to try it.
Some while ago Keith Howard did some measurements for me on a variety of
DACs to probe their behaviour with relatively simple waveforms with
overshoots. Many of the results show distortion large enough to be clearly
visible on the output waveforms seen on a 'scope. So far as I know, no-one
systematically measures this.
That was with simple waveforms that would actually mask or avoid some
classes of DAC misbehavior that overshoots could cause. The "Wave from Hell"
would be more demanding. Any DAC that gave the right result could probably
cope with everything on any Audio CD, etc.
However I suspect most consumer DACs would produce results far poorer than
CEP displayed.
I'm afraid that a few decades of observing how the domestic audio biz works
has left me with the feeling that areas where no-one checks are ones where
possible problems tend to become real behaviour. Only the Shadow knows...
:-)
Slainte,
Don Pearce
<no...@audiomisc.co.uk> wrote:
>In article <4afecb84....@news.eternal-september.org>, Don Pearce
><sp...@spam.com> wrote:
>> On Sun, 8 Nov 2009 13:38:38 -0000, "Keith G" <k...@moirac.adsl24.co.uk>
>> wrote:
>
>> >No, you were at cross-purposes here trying to answer two different
>> >lines of enquiry - having seen your link, I merely posted my pictures
>> >to compare a similar waveform in Sound Forge from an (apparent)
>> >*resolution* POV as I thought the Audition waveform looked far too
>> >'smoothed out' and with considerably less 'sample points' (if that is
>> >what they were) - nothing to do with sound levels at all....
>> >
>
>> OK, no. Audition had precisely the same number of sample points (44100
>> per second in both our cases), and the "smoothing out" of Audition has
>> turned out to be more or less precisely correct, even with a
>> pathological signal. What Audition shows is what the DAC provides to an
>> amplifier.
>
>Not quite. It shows what Audition *should* provide *if* the DAC can deal
>with overshoots. Measurements on real DACs show examples that are clearly
>unable to do that. Even for overshoots much smaller than the one in the
>test waveform that deliberately sets out to be an extremely demanding case.
>
>Slainte,
>
>Jim
OK, let me qualify. it is accurate provided there is no overloading.
d
Don Pearce
<no...@audiomisc.co.uk> wrote:
>Alas for reasons I mention elsewhere I am less sure that most DACs do
>as good a job with out-of-range overshoots than CEP. People generally
>don't check this, and the measurements I've seen don't fill me with
>confidence.
No DAC (as far as I know) can deal with these peaks. As I said before,
it is the oversampler that will create the peak, and if it is working
in the same 16 bit domain as the rest of the DAC it simply will be
clipped. Unfortunately that clipping will produce not just harmonics
which we can probably ignore, but also aliasing back into the audio
band.
I wonder if any DAC manufacturer has thought to make the oversampling
part of the DAC 17 bits. That would provide an extra 6dB of headroom
and this problem would cease to be an issue.
d
Iain Churches
Yes. Despiter appearances (!!) he was actually a very good
pianist indeed.
I met him once (fleetingly) when he came to Decca to look
at a Baldwin piano which was surplus to requirements.
I was asked to take him and his entourage down to Studio III.
The piano which was in pristine condition and had been
rarely used. It had a stiff action because it was never played,
and no-one wanted to play it because it had a stiff action!! Well,
Mr Liberace bought the piano and as far as I know it is still part
of the Liberace piano collection now run as a museum by a trust,
along with the Chickering grand owned by George Gershwin on
which he wrote Rhapsody in Blue.
Iain
UnsteadyKen
> Why waste time with 'hyphen technology'? - You need one of these:
>
> http://www.brennan.co.uk/home/
No way Jose.
Vinylise your digits, you know it makes sense.
http://www.vestax.com/v/products/recorders/vrx2000.html
!
--
Ken O'Meara
http://www.btinternet.com/~unsteadyken/