Sign in with
Sign up | Sign in
Your question

90 degree phase offset in CoolEdit

Last response: in Home Audio
Share
Anonymous
March 11, 2005 5:41:46 PM

Archived from groups: rec.audio.misc,rec.audio.pro (More info?)

Hi

I am not very familiar with CoolEdit and I want to playback my voice (or any
mp3 file) with a 90 degree phase offset between the 2 channels/speakers. It
must be easy but how do you do this?

Thanks
Bob
Anonymous
March 11, 2005 5:41:47 PM

Archived from groups: rec.audio.misc,rec.audio.pro (More info?)

"Bob Pit" <bobptz@yahoo.gr> wrote in message
news:1110544844.814317@athnrd02
> Hi
>
> I am not very familiar with CoolEdit and I want to playback my voice
> (or any mp3 file) with a 90 degree phase offset between the 2
> channels/speakers. It must be easy but how do you do this?

Effects, Filters, Graphic Phase Shifter
Anonymous
March 11, 2005 5:41:47 PM

Archived from groups: rec.audio.misc,rec.audio.pro (More info?)

In article <1110544844.814317@athnrd02>, Bob Pit <bobptz@yahoo.gr> wrote:
>Hi
>
>I am not very familiar with CoolEdit and I want to playback my voice (or any
>mp3 file) with a 90 degree phase offset between the 2 channels/speakers. It
>must be easy but how do you do this?

It's not easy at all. You basically need an all-pass filter that shifts
90 degrees at all frequencies, which is not a trivial thing and it's not
something any of the audio packages normally have because it's not normally
very useful.

Why do you want to do this?
--scott

--
"C'est un Nagra. C'est suisse, et tres, tres precis."
Related resources
Anonymous
March 11, 2005 8:13:32 PM

Archived from groups: rec.audio.misc,rec.audio.pro (More info?)

"Bob Pit" wrote ...
> I am not very familiar with CoolEdit and I want to playback my voice
> (or any mp3 file) with a 90 degree phase offset between the 2
> channels/speakers. It must be easy but how do you do this?

90 degrees of phase shift depends on what frequency you are talking
about, doesn't it? It is 1000x longer for 20Hz than it is for 20KHz
Anonymous
March 11, 2005 9:10:10 PM

Archived from groups: rec.audio.misc,rec.audio.pro (More info?)

Bob Pit wrote:
> Hi
>
> I am not very familiar with CoolEdit and I want to playback my voice (or any
> mp3 file) with a 90 degree phase offset between the 2 channels/speakers. It
> must be easy but how do you do this?

You need a filter called a Hilbert transformer. Not a thing
readily available with DAW software but an FIR can be
generated in Matlab which does a pretty good job and can be
applied via Cool Edit's convolution function.


Bob
--

"Things should be described as simply as possible, but no
simpler."

A. Einstein
Anonymous
March 11, 2005 11:31:09 PM

Archived from groups: rec.audio.misc,rec.audio.pro (More info?)

Scott Dorsey wrote:
> In article <1110544844.814317@athnrd02>, Bob Pit <bobptz@yahoo.gr>
wrote:
> >Hi
> >
> >I am not very familiar with CoolEdit and I want to playback my voice
(or any
> >mp3 file) with a 90 degree phase offset between the 2
channels/speakers. It
> >must be easy but how do you do this?
>
> It's not easy at all. You basically need an all-pass filter that
shifts
> 90 degrees at all frequencies, which is not a trivial thing and it's
not
> something any of the audio packages normally have because it's not
normally
> very useful.

Funny - thing is I have on occasion seen in the course of my TV
Audio duties program sources that show themselves on my Tektronics
scope to be 90% out of phase, but I was unaware there was anything
potentially subliminal involved. <g>

I forget exactly the details though, something to do with one side
of a balanced stereo feed being half patched in the patchbay , with
maybe the two sides having reversed polarity.

Will Miho
NY Music and TV Audio Guy
Staff Audio / Fox News / M-AES
"The large print giveth and the small print taketh away..." Tom Waits
Anonymous
March 12, 2005 12:46:38 AM

Archived from groups: rec.audio.misc,rec.audio.pro (More info?)

>> I am not very familiar with CoolEdit and I want to playback my voice
(or any
>> mp3 file) with a 90 degree phase offset between the 2
channels/speakers. It
>> must be easy but how do you do this?



>You need a filter called a Hilbert transformer. Not a thing
>readily available with DAW software but an FIR can be
>generated in Matlab which does a pretty good job and can be
>applied via Cool Edit's convolution function.

Yes, and wouldn't he also have to compensate for the delay in the
Hilbert FIR by delaying the other unprocessed channel by 1/2 the length
of the Hilbert FIR?

Joe
Anonymous
March 12, 2005 1:29:37 AM

Archived from groups: rec.audio.misc,rec.audio.pro (More info?)

Joseph Meditz wrote:
>>>I am not very familiar with CoolEdit and I want to playback my voice
>
> (or any
>
>>>mp3 file) with a 90 degree phase offset between the 2
>
> channels/speakers. It
>
>>>must be easy but how do you do this?
>
>
>
>
>>You need a filter called a Hilbert transformer. Not a thing
>>readily available with DAW software but an FIR can be
>>generated in Matlab which does a pretty good job and can be
>>applied via Cool Edit's convolution function.
>
>
> Yes, and wouldn't he also have to compensate for the delay in the
> Hilbert FIR by delaying the other unprocessed channel by 1/2 the length
> of the Hilbert FIR?

Yes. I'd do that by creating a stereo FIR with one channel
just having a 0 dB impulse in the middle and the Hilbert
transform on the other channel.

If you were to mix those two channels you'd have what's
called a quadrature mixer and the result would be a constant
phase shift of between 0 and 90 at every frequency, the
angle determined by the mix ratio.


Bob
--

"Things should be described as simply as possible, but no
simpler."

A. Einstein
Anonymous
March 12, 2005 1:42:10 AM

Archived from groups: rec.audio.misc,rec.audio.pro (More info?)

> Effects, Filters, Graphic Phase Shifter
I did a +90 degree phase shift on the left channel. It seems something like
what I am looking for, but still far from it.

Bob


"Arny Krueger" <arnyk@hotpop.com> wrote in message
news:TPidnVQi2KutAqzfRVn-2A@comcast.com...
>
> "Bob Pit" <bobptz@yahoo.gr> wrote in message
> news:1110544844.814317@athnrd02
>> Hi
>>
>> I am not very familiar with CoolEdit and I want to playback my voice
>> (or any mp3 file) with a 90 degree phase offset between the 2
>> channels/speakers. It must be easy but how do you do this?
>
> Effects, Filters, Graphic Phase Shifter
>
>
Anonymous
March 12, 2005 1:42:11 AM

Archived from groups: rec.audio.misc,rec.audio.pro (More info?)

Voxengo makes a phase shift plugin.

Al

On Fri, 11 Mar 2005 22:42:10 +0200, "Bob Pit" <bobptz@yahoo.gr> wrote:

>> Effects, Filters, Graphic Phase Shifter
>I did a +90 degree phase shift on the left channel. It seems something like
>what I am looking for, but still far from it.
>
>Bob
>
>
>"Arny Krueger" <arnyk@hotpop.com> wrote in message
>news:TPidnVQi2KutAqzfRVn-2A@comcast.com...
>>
>> "Bob Pit" <bobptz@yahoo.gr> wrote in message
>> news:1110544844.814317@athnrd02
>>> Hi
>>>
>>> I am not very familiar with CoolEdit and I want to playback my voice
>>> (or any mp3 file) with a 90 degree phase offset between the 2
>>> channels/speakers. It must be easy but how do you do this?
>>
>> Effects, Filters, Graphic Phase Shifter
>>
>>
>
Anonymous
March 12, 2005 1:42:11 AM

Archived from groups: rec.audio.misc,rec.audio.pro (More info?)

In article <1110573659.111832@athnrd02>, Bob Pit <bobptz@yahoo.gr> wrote:
>> Effects, Filters, Graphic Phase Shifter
>I did a +90 degree phase shift on the left channel. It seems something like
>what I am looking for, but still far from it.

What are you looking for?
--scott


--
"C'est un Nagra. C'est suisse, et tres, tres precis."
Anonymous
March 12, 2005 1:43:55 AM

Archived from groups: rec.audio.misc,rec.audio.pro (More info?)

How about the "Effects, Filters, Graphic Phase Shifter" that Arny suggested?

> Why do you want to do this?
The reason I want it is that the effect of a voiced processed like this, is
very hypnotic. I want to make hypnotic/subliminal mp3 files for myself.

Bob


"Scott Dorsey" <kludge@panix.com> wrote in message
news:D 0sa7t$gfk$1@panix2.panix.com...
>
> In article <1110544844.814317@athnrd02>, Bob Pit <bobptz@yahoo.gr> wrote:
>>Hi
>>
>>I am not very familiar with CoolEdit and I want to playback my voice (or
>>any
>>mp3 file) with a 90 degree phase offset between the 2 channels/speakers.
>>It
>>must be easy but how do you do this?
>
> It's not easy at all. You basically need an all-pass filter that shifts
> 90 degrees at all frequencies, which is not a trivial thing and it's not
> something any of the audio packages normally have because it's not
> normally
> very useful.
>
> Why do you want to do this?
> --scott
>
> --
> "C'est un Nagra. C'est suisse, et tres, tres precis."
Anonymous
March 12, 2005 2:26:39 AM

Archived from groups: rec.audio.misc,rec.audio.pro (More info?)

> What are you looking for?
As I said, I want to make hypnotic/subliminal mp3 files for myself. The
only specs I have is "90 degree phase offset". I cannot describe it in
other specific ways. I can send you a very small segment of a professional
created mp3 file that use this technique if you want. If you are
experienced, then you will know how they did it.

Bob
Anonymous
March 12, 2005 2:26:40 AM

Archived from groups: rec.audio.misc,rec.audio.pro (More info?)

In article <1110576331.158713@athnrd02>, Bob Pit <bobptz@yahoo.gr> wrote:
>> What are you looking for?
>As I said, I want to make hypnotic/subliminal mp3 files for myself. The
>only specs I have is "90 degree phase offset". I cannot describe it in
>other specific ways. I can send you a very small segment of a professional
>created mp3 file that use this technique if you want. If you are
>experienced, then you will know how they did it.

No, a 90-degree phase shift is really not very audible at all. Again, it's
an all-pass thing.

Are you thinking about some kind of comb filtering, and not really phase
shift at all?

Since MP3s don't exactly preserve phase very well anyway, any of this stuff
is going to be mostly lost in MP3 encoding in any event.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."
Anonymous
March 12, 2005 2:34:10 AM

Archived from groups: rec.audio.misc,rec.audio.pro (More info?)

In theory, a phase shift of 90 degrees is done by a capacitor.To calculate
the capacitance, for the desired frequency (20 Hz-20 kHz) you have to use
the following formula:Xc= 1/(omega*c)where omega=2*pi*f, f=frequency in
Hz.To give you an idea, for fm and tv frequency (VHF and UHF) we use 1nF and
for AM 100 nF capacitors.So, for audio frequencies, you need a 10
micro-farad capacitor (100 times that of the AM).
Tzortzakakis Dimitriïs
major in electrical engineering, freelance electrician
FH von Iraklion-Kreta, freiberuflicher Elektriker
dimtzort AT otenet DOT gr
Ï "Bob Pit" <bobptz@yahoo.gr> Ýãñáøå óôï ìÞíõìá
news:1110544844.814317@athnrd02...
> Hi
>
> I am not very familiar with CoolEdit and I want to playback my voice (or
any
> mp3 file) with a 90 degree phase offset between the 2 channels/speakers.
It
> must be easy but how do you do this?
>
> Thanks
> Bob
>
>
Anonymous
March 12, 2005 2:34:11 AM

Archived from groups: rec.audio.misc,rec.audio.pro (More info?)

Can you post a link to what you're talking about? It would be easier to
offer help.
Thanks


"Dimitrios Tzortzakakis" <use@address.below> wrote in message
news:D 0t33a$jv8$1@usenet.otenet.gr...
> In theory, a phase shift of 90 degrees is done by a capacitor.To calculate
> the capacitance, for the desired frequency (20 Hz-20 kHz) you have to use
> the following formula:Xc= 1/(omega*c)where omega=2*pi*f, f=frequency in
> Hz.To give you an idea, for fm and tv frequency (VHF and UHF) we use 1nF
> and
> for AM 100 nF capacitors.So, for audio frequencies, you need a 10
> micro-farad capacitor (100 times that of the AM).
> Tzortzakakis Dimitriïs
> major in electrical engineering, freelance electrician
> FH von Iraklion-Kreta, freiberuflicher Elektriker
> dimtzort AT otenet DOT gr
> Ï "Bob Pit" <bobptz@yahoo.gr> Ýãñáøå óôï ìÞíõìá
> news:1110544844.814317@athnrd02...
>> Hi
>>
>> I am not very familiar with CoolEdit and I want to playback my voice (or
> any
>> mp3 file) with a 90 degree phase offset between the 2 channels/speakers.
> It
>> must be easy but how do you do this?
>>
>> Thanks
>> Bob
>>
>>
>
>
Anonymous
March 12, 2005 3:24:54 AM

Archived from groups: rec.audio.misc,rec.audio.pro (More info?)

> No, a 90-degree phase shift is really not very audible at all. Again,
> it's
> an all-pass thing.
>
> Are you thinking about some kind of comb filtering, and not really phase
> shift at all?

I wish I knew what these mean. If I knew all these, probably I would not
ask for help in this forum.

Bob
Anonymous
March 12, 2005 3:24:55 AM

Archived from groups: rec.audio.misc,rec.audio.pro (More info?)

In article <1110579825.435618@athnrd02> bobptz@yahoo.gr writes:

> I wish I knew what these mean. If I knew all these, probably I would not
> ask for help in this forum.

I think you should ask whoever produced the files you're trying to
emulate. A 90 degree phase shift to one channel doesn't make any
sense. But maybe it's not supposed to.

If you'd like to experiment with hardware, the Little Labs IBP is a
box that has an adjustable phase shift. It's most often used to line
up a guitar direct signal with a microphone on an amplifier. It might
be educational to try it if you can find one to borrow.

--
I'm really Mike Rivers (mrivers@d-and-d.com)
However, until the spam goes away or Hell freezes over,
lots of IP addresses are blocked from this system. If
you e-mail me and it bounces, use your secret decoder ring
and reach me here: double-m-eleven-double-zero at yahoo
March 12, 2005 11:47:55 AM

Archived from groups: rec.audio.misc,rec.audio.pro (More info?)

Phil Allison wrote:
> "Phil Allison"
> "Joseph Meditz"
>
> >
> >> Yes, and wouldn't he also have to compensate for the delay in the
> >> Hilbert FIR by delaying the other unprocessed channel by 1/2 the
length
> >> of the Hilbert FIR?
> >>
> >
>
> ** Forget this:
>
> " ** See my post on this - a Hilbert transformer delays and
phase
> shifts
> both channels, one 90 degrees more than the other. "
>
>
> ** The OP will need to use the two all pass chains of the Hilbert
> transformer for Left and Right channels.
>
> AFAIK there is no way to produce a quadrature version of a given
audio
> signal.
>
>
>
>
> ............. Phil


A +45 and -45 degree shifted version of a given audio signal can be
approximated in real time over a limited range of frequencies using a
large number of RC networks.

Mark
Anonymous
March 12, 2005 12:02:49 PM

Archived from groups: rec.audio.misc,rec.audio.pro (More info?)

In article <1110579825.435618@athnrd02>, Bob Pit <bobptz@yahoo.gr> wrote:
>> No, a 90-degree phase shift is really not very audible at all. Again,
>> it's
>> an all-pass thing.
>>
>> Are you thinking about some kind of comb filtering, and not really phase
>> shift at all?
>
>I wish I knew what these mean. If I knew all these, probably I would not
>ask for help in this forum.

Okay, comb filtering is what happens when you sum a signal that has been
shifted slightly in _time_ (not phase) with the original signal. This
results in a frequency response that looks like a toothed comb, and has
a somewhat airy and distant sound to it.

You can take a signal and shift it in _time_, in which every frequency is
shifted the same time factor, or you can shift it in _phase_, in which the
lower notes are shifted longer in time than the higher notes. A tape
machine is designed to shift signals in time... you put signals in and
what you get out is delayed by hours if not years when you play it back.
A delay in _phase_ is like that, but in varying amounts for different
frequencies.
--scott

--
"C'est un Nagra. C'est suisse, et tres, tres precis."
Anonymous
March 12, 2005 12:18:46 PM

Archived from groups: rec.audio.misc,rec.audio.pro (More info?)

Richard Crowley <rcrowley7@xprt.net> wrote:
>"Bob Pit" wrote ...
>> I am not very familiar with CoolEdit and I want to playback my voice
>> (or any mp3 file) with a 90 degree phase offset between the 2
>> channels/speakers. It must be easy but how do you do this?
>
>90 degrees of phase shift depends on what frequency you are talking
>about, doesn't it? It is 1000x longer for 20Hz than it is for 20KHz

Right. That is why a filter with constant phase shift at all frequencies
is very difficult to build. It's possible to do in the digital domain,
but it's not very useful.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."
Anonymous
March 12, 2005 12:18:47 PM

Archived from groups: rec.audio.misc,rec.audio.pro (More info?)

> Right. That is why a filter with constant phase shift at all frequencies
> is very difficult to build. It's possible to do in the digital domain, but
> it's not very useful.

It is, for matrixed surround systems, or situations where you want to reduce the
build-up of mono components when stereo recordings are mixed to mono.

Traditionally, 90-degree phase-shift filters are implemented as an all-pass
filter for the signal that is _not_ to be shifted, plus an all-pass filter that
has the shift of the first filter, plus 90 degrees, for the signal that _is_ to
be shifted. qv, SQ, QS, et al.
Anonymous
March 12, 2005 3:16:01 PM

Archived from groups: rec.audio.misc,rec.audio.pro (More info?)

On Fri, 11 Mar 2005 14:41:46 +0200, "Bob Pit" <bobptz@yahoo.gr> wrote:

>
>I am not very familiar with CoolEdit and I want to playback my voice (or any
>mp3 file) with a 90 degree phase offset between the 2 channels/speakers. It
>must be easy but how do you do this?

We need to find out just what you mean by "90 deg phase offset".

You're going to get a lot of technical willy-waving now from people
who'll delight in taking your request literally. But I bet that
isn't what you want.

Can you point us to the source that suggested "90 deg phase offset"?

CubaseFAQ www.laurencepayne.co.uk/CubaseFAQ.htm
"Possibly the world's least impressive web site": George Perfect
Anonymous
March 12, 2005 3:38:16 PM

Archived from groups: rec.audio.misc,rec.audio.pro (More info?)

"Scott Dorsey"
Bob Pit :

>>
>>I am not very familiar with CoolEdit and I want to playback my voice (or
>>any
>>mp3 file) with a 90 degree phase offset between the 2 channels/speakers.
>>It
>>must be easy but how do you do this?
>
> It's not easy at all. You basically need an all-pass filter that shifts
> 90 degrees at all frequencies, which is not a trivial thing and it's not
> something any of the audio packages normally have because it's not
> normally
> very useful.
>


** Here is how you do it:

http://home.att.net/~wa1sov/technical/allpass/allpass.h...


Enjoy.........



> Why do you want to do this?


** He is loopy.



............. Phil
Anonymous
March 12, 2005 5:22:55 PM

Archived from groups: rec.audio.misc,rec.audio.pro (More info?)

"Bob Pit" <bobptz@yahoo.gr> wrote in news:1110573659.111832@athnrd02:

>> Effects, Filters, Graphic Phase Shifter
> I did a +90 degree phase shift on the left channel. It seems
> something like what I am looking for, but still far from it.

Did you try mixing the two channels to mono? Is that the effect you want?
March 12, 2005 6:23:50 PM

Archived from groups: rec.audio.misc,rec.audio.pro (More info?)

On 3/11/05 5:24 PM, in article 1110579825.435618@athnrd02, "Bob Pit"
<bobptz@yahoo.gr> wrote:

>> No, a 90-degree phase shift is really not very audible at all. Again,
>> it's
>> an all-pass thing.
>>
>> Are you thinking about some kind of comb filtering, and not really phase
>> shift at all?
>
> I wish I knew what these mean. If I knew all these, probably I would not
> ask for help in this forum.
>
> Bob

Hey.. Bob... Ya don;t have to know, just listen to the less-kneejerk-manic
types here (like Scott!) who are still with you rather than being
hyperkinetic foulmouth reactionary intellectia. (?!)

Due to my self-limited time for reading this forum (I wanna help so many but
the incredible amount of sheer beligerance boggles) I start skipping even
the interesting threads like this when they get pointlessly whacky.
Can you (if you haven;t... I might've missed it) actually DESCRIBE the
effect you;re looking to achieve, maybe even an example of a recording that
demonstrates it?

JV
March 12, 2005 6:25:33 PM

Archived from groups: rec.audio.misc,rec.audio.pro (More info?)

On 3/12/05 9:18 AM, in article d0uto6$c8b$1@panix2.panix.com, "Scott Dorsey"
<kludge@panix.com> wrote:

>> 90 degrees of phase shift depends on what frequency you are talking
>> about, doesn't it? It is 1000x longer for 20Hz than it is for 20KHz
>
> Right. That is why a filter with constant phase shift at all frequencies
> is very difficult to build. It's possible to do in the digital domain,
> but it's not very useful.
> --scott

Never thought of this... What the hell does that SOOUND LIKE? Doesn;t the
old COMREX system do this... No wait, that's LINEAR frequency shifting.. Not
the same thing at all.
Anonymous
March 12, 2005 6:48:25 PM

Archived from groups: rec.audio.misc,rec.audio.pro (More info?)

"WillStG"

>
> Funny - thing is I have on occasion seen in the course of my TV
> Audio duties program sources that show themselves on my Tektronics
> scope to be 90% out of phase,


** That could only be so with a test tone in use.





.............. Phil
Anonymous
March 12, 2005 6:59:30 PM

Archived from groups: rec.audio.misc,rec.audio.pro (More info?)

There's no link, I just read through the book electrical circuits, which is
necessary to pass to get the degree in EE in college.

--
Tzortzakakis Dimitriïs
major in electrical engineering, freelance electrician
FH von Iraklion-Kreta, freiberuflicher Elektriker
dimtzort AT otenet DOT gr
Ï "Dave Kowalski" <daveski@optonline.net> Ýãñáøå óôï ìÞíõìá
news:_TpYd.1555$qt2.444@fe10.lga...
> Can you post a link to what you're talking about? It would be easier to
> offer help.
> Thanks
>
>
> "Dimitrios Tzortzakakis" <use@address.below> wrote in message
> news:D 0t33a$jv8$1@usenet.otenet.gr...
> > In theory, a phase shift of 90 degrees is done by a capacitor.To
calculate
> > the capacitance, for the desired frequency (20 Hz-20 kHz) you have to
use
> > the following formula:Xc= 1/(omega*c)where omega=2*pi*f, f=frequency in
> > Hz.To give you an idea, for fm and tv frequency (VHF and UHF) we use 1nF
> > and
> > for AM 100 nF capacitors.So, for audio frequencies, you need a 10
> > micro-farad capacitor (100 times that of the AM).
> > Tzortzakakis Dimitriïs
> > major in electrical engineering, freelance electrician
> > FH von Iraklion-Kreta, freiberuflicher Elektriker
> > dimtzort AT otenet DOT gr
> > Ï "Bob Pit" <bobptz@yahoo.gr> Ýãñáøå óôï ìÞíõìá
> > news:1110544844.814317@athnrd02...
> >> Hi
> >>
> >> I am not very familiar with CoolEdit and I want to playback my voice
(or
> > any
> >> mp3 file) with a 90 degree phase offset between the 2
channels/speakers.
> > It
> >> must be easy but how do you do this?
> >>
> >> Thanks
> >> Bob
> >>
> >>
> >
> >
>
>
Anonymous
March 12, 2005 6:59:31 PM

Archived from groups: rec.audio.pro (More info?)

In article <d0usqq$2tm$1@usenet.otenet.gr> use@address.below writes:

> There's no link, I just read through the book electrical circuits, which is
> necessary to pass to get the degree in EE in college.

If all I had to do in order to get my EE degree was read a book, I'd
get a few more.

I think that what he's asking (and we all want to know) is the source
of the information that describes these subliminal recordings with a
90 degree phase shift. We're working on too simplistic an explanation
of what's really required. If it's really a 90 degree phase shift of
one channel relative to the other one, you have some methods of doing
that. But somehow the effect of this 90 degree thing sounds either
like total horse manure or it's not a correct or complete explanation
of the whole process.


--
I'm really Mike Rivers (mrivers@d-and-d.com)
However, until the spam goes away or Hell freezes over,
lots of IP addresses are blocked from this system. If
you e-mail me and it bounces, use your secret decoder ring
and reach me here: double-m-eleven-double-zero at yahoo
Anonymous
March 12, 2005 7:05:21 PM

Archived from groups: rec.audio.misc,rec.audio.pro (More info?)

John <ssconmag1@verizon.net> wrote:
>On 3/12/05 9:18 AM, in article d0uto6$c8b$1@panix2.panix.com, "Scott Dorsey"
><kludge@panix.com> wrote:
>
>>> 90 degrees of phase shift depends on what frequency you are talking
>>> about, doesn't it? It is 1000x longer for 20Hz than it is for 20KHz
>>
>> Right. That is why a filter with constant phase shift at all frequencies
>> is very difficult to build. It's possible to do in the digital domain,
>> but it's not very useful.
>
>Never thought of this... What the hell does that SOOUND LIKE? Doesn;t the
>old COMREX system do this... No wait, that's LINEAR frequency shifting.. Not
>the same thing at all.

As Sommerwerck points out, this is how the rear channels are encoded with
some of the matrix quadraphonic systems. And what it sounds like, for the
most part, isn't much at all. If you listen to the things undecoded, they
may sound a little "funny" but not offensively so.
--scott

--
"C'est un Nagra. C'est suisse, et tres, tres precis."
Anonymous
March 12, 2005 7:05:22 PM

Archived from groups: rec.audio.misc,rec.audio.pro (More info?)

> As Sommerwerck points out, this is how the rear channels are encoded
> with some of the matrix quadraphonic systems. And what it sounds like,
> for the most part, isn't much at all. If you listen to the things undecoded,
> they may sound a little "funny," but not offensively so.

When played in stereo, on speakers that image well, the rear channels of an SQ
recording appear slightly "outside" the speakers.
March 12, 2005 7:41:49 PM

Archived from groups: rec.audio.misc,rec.audio.pro (More info?)

On 3/11/05 4:26 PM, in article 1110576331.158713@athnrd02, "Bob Pit"
<bobptz@yahoo.gr> wrote:

>> What are you looking for?
> As I said, I want to make hypnotic/subliminal mp3 files for myself. The
> only specs I have is "90 degree phase offset". I cannot describe it in
> other specific ways. I can send you a very small segment of a professional
> created mp3 file that use this technique if you want. If you are
> experienced, then you will know how they did it.

I'll make some wildassed speculation here.
Possible #1:
Simple reverse-polarity between left and right channels.
(while this is almost always thought of as "180deg phase shift". It's only
180deg at ONE frequency... Which is why everybody MUST understand that
REVERSE PHASE really means NOTHING like REVERSE POLARITY).
This results (especially in headphones) in a bizarre sort of 'behind-me'
effect.
It also means that the sound COMPLETELY VANISHES when the audio is
listened to in mono. What it might do to how various processors handle it,
is up for grabs... For instance let's say you want to compress the stereo
pair (which as I said are reverse=polarity) and the compressor senses the
overall level at any instant by COMBINING left-and-right channels, just to
derive the control signal to discern how much gain reduction to effect, then
with the polarity flop in the signal, that control-signal instead of being a
very real SUM of left and right overall levels, becomes ZERO and the
compressor might do NOTHING or merely get 'interesting' in it's actions.

POSSIBLE #2
Simple SHORT time-delay between channels. (on the order of milliseconds)
This is actually phase-shift, but while the TIME might be a constant (say a
few milliseconds) the amount of PHASE shifted varies with frequency, way
more at hi's, very little with lo's. This again results (since it's playing
with arrival-times at left-vs-right ear) in a 'where-IS-that-coming-from'
effect in your head, again especially in headphones.
When THIS sort of thing is summed to mono, the effect goes from a minimal
vacuum-cleaner-hose hollow effect to, with time-delays on the order of 50ms
on up, VERY whacked peaks and cancellations of frequencies all up and down
the spectrum, mathematically related, that, when looked at on an analyser,
show as regularly-spaced peaks and dips wider-and-fewer (for short delay
times), through MASSIVE numbers of lines of peaks and dips closer and closer
together at regular intervals, making the chart look like teeth in a comb.

SUMMATION:
These two are EASY to achieve with most any audio gear you have available.
You might try them to see if they can do what you want to hear. You have
nothing to lose (not even money!)

Beyond these two approaches, you walk into the mathematical REAL world of
serious weird stuff, and what the folks here have been going on (at
wonderful length) about... And THAt has been a real breath of fresh air
(excluding the kneejerk socially-inept inexcusable foulmouth overeactions to
what SHOULD be simple pleasant exploration of a concept) amongst the
mundanity.

Hope this helps
Jv
Anonymous
March 12, 2005 8:02:51 PM

Archived from groups: rec.audio.misc,rec.audio.pro (More info?)

"Joseph Meditz"
>
>
>>You need a filter called a Hilbert transformer. Not a thing
>>readily available with DAW software but an FIR can be
>>generated in Matlab which does a pretty good job and can be
>>applied via Cool Edit's convolution function.
>
> Yes, and wouldn't he also have to compensate for the delay in the
> Hilbert FIR by delaying the other unprocessed channel by 1/2 the length
> of the Hilbert FIR?
>


** See my post on this - a Hilbert transformer delays and phase shifts
both channels, one 90 degrees more than the other.

Remember - the OP wanted " ... a 90 degree phase offset between the 2
channels/speakers. "




................ Phil
Anonymous
March 12, 2005 8:11:40 PM

Archived from groups: rec.audio.misc,rec.audio.pro (More info?)

"Phil Allison"
"Joseph Meditz"

>
>> Yes, and wouldn't he also have to compensate for the delay in the
>> Hilbert FIR by delaying the other unprocessed channel by 1/2 the length
>> of the Hilbert FIR?
>>
>

** Forget this:

" ** See my post on this - a Hilbert transformer delays and phase
shifts
both channels, one 90 degrees more than the other. "


** The OP will need to use the two all pass chains of the Hilbert
transformer for Left and Right channels.

AFAIK there is no way to produce a quadrature version of a given audio
signal.




.............. Phil
Anonymous
March 12, 2005 8:11:41 PM

Archived from groups: rec.audio.misc,rec.audio.pro (More info?)

Phil Allison wrote:

> AFAIK there is no way to produce a quadrature version of a given audio
> signal.

Then modify what you know.


Bob
--

"Things should be described as simply as possible, but no
simpler."

A. Einstein
Anonymous
March 12, 2005 8:11:41 PM

Archived from groups: rec.audio.misc,rec.audio.pro (More info?)

"Phil Allison" <philallison@tpg.com.au> wrote in message
news:39ffgqF61g34eU1@individual.net

> AFAIK there is no way to produce a quadrature version of a given
> audio signal.

That why SSB transmitters, a popular form of which is dependent on the
production of two audio signals that are in quadrature, never work.

;-)


BTW, I generated a stereo test tone composed of 1 second segments of 100 Hz,
500 Hz, 1KHz, 5 KHz and 10 KHz in Audition. I then applied the 90 degree
preset for the graphic phase shifter to the right channel. Each segment of
the right channel of the test tone was in fact phase shifted by 90 degrees
from the corresponding segment of the left channel.

I conclude that since Phil is always right, my test tone composed of 1
second segments of 100 Hz, 500 Hz, 1KHz, 5 KHz and 10 KHz is in fact not a
"given audio signal".

What is it?

;-)
March 13, 2005 12:09:02 AM

Archived from groups: rec.audio.misc,rec.audio.pro (More info?)

On 3/12/05 4:05 PM, in article d0vlih$p1p$1@panix2.panix.com, "Scott Dorsey"
<kludge@panix.com> wrote:

> John <ssconmag1@verizon.net> wrote:
>> On 3/12/05 9:18 AM, in article d0uto6$c8b$1@panix2.panix.com, "Scott Dorsey"
>> <kludge@panix.com> wrote:
>>
>>>> 90 degrees of phase shift depends on what frequency you are talking
>>>> about, doesn't it? It is 1000x longer for 20Hz than it is for 20KHz
>>>
>>> Right. That is why a filter with constant phase shift at all frequencies
>>> is very difficult to build. It's possible to do in the digital domain,
>>> but it's not very useful.
>>
>> Never thought of this... What the hell does that SOOUND LIKE? Doesn;t the
>> old COMREX system do this... No wait, that's LINEAR frequency shifting.. Not
>> the same thing at all.
>
> As Sommerwerck points out, this is how the rear channels are encoded with
> some of the matrix quadraphonic systems. And what it sounds like, for the
> most part, isn't much at all. If you listen to the things undecoded, they
> may sound a little "funny" but not offensively so.

And there's where I get left behind...
Wouldn't constant phase shift mean VARYING time shift across the spectrum
and everything goes to hell in a handbucketbrigadedevice?
Anonymous
March 13, 2005 12:09:03 AM

Archived from groups: rec.audio.misc,rec.audio.pro (More info?)

John <ssconmag1@verizon.net> wrote:
>
>And there's where I get left behind...
>Wouldn't constant phase shift mean VARYING time shift across the spectrum
>and everything goes to hell in a handbucketbrigadedevice?

Right, so an impulse no longer looks like an impulse, but is sort of
smushed to one side or the other. It tends to make detail sound a little
bit smeary... string attacks don't sound sharp any more. Kind of an AM radio
sound thing, since I guess that's where people are most familiar with
massive group delay.
--scott


--
"C'est un Nagra. C'est suisse, et tres, tres precis."
Anonymous
March 13, 2005 12:09:03 AM

Archived from groups: rec.audio.misc,rec.audio.pro (More info?)

John wrote:

> And there's where I get left behind...
> Wouldn't constant phase shift mean VARYING time shift across the spectrum
> and everything goes to hell in a handbucketbrigadedevice?

Yes. Think of it as a magnitude compensated first
derivative of the signal WRT time.


Bob
--

"Things should be described as simply as possible, but no
simpler."

A. Einstein
Anonymous
March 13, 2005 7:33:13 AM

Archived from groups: rec.audio.misc,rec.audio.pro (More info?)

>> And there's where I get left behind...
>> Wouldn't constant phase shift mean VARYING time shift across the
>> spectrum and everything goes to hell in a handbucketbrigadedevice?

> Yes. Think of it as a magnitude compensated first
> derivative of the signal WRT time.

I'm going to make one comment, and then let it drop, because I don't want to get
involved in a technical tsuris.

There is a difference between true delay (such as recording a signal and playing
it back later) and what I call "envelope shift" (what most people call group
delay). I find the whole issue thoroughly confusing, and wish someone would
study it carefully and thoroughly (if no one has already) and publish an article
on it.
March 13, 2005 9:54:35 AM

Archived from groups: rec.audio.pro (More info?)

Chris Hornbeck wrote:
> On Sun, 13 Mar 2005 04:33:13 -0800, "William Sommerwerck"
> <williams@nwlink.com> wrote:
>
> >There is a difference between true delay (such as recording a signal
and playing
> >it back later) and what I call "envelope shift" (what most people
call group
> >delay). I find the whole issue thoroughly confusing, and wish
someone would
> >study it carefully and thoroughly (if no one has already) and
publish an article
> >on it.
>
> I don't know what envelope shift is, but group delay is just an
> expression for the variations in time delay at various frequencies.
> What I find confusing is where the name comes from, so I hope
> that gets included in the article.
>
> Chris Hornbeck

No Actualy group delay, envelope delay and delay by recording and
playback are all the same thing.

But when most people say group delay talking about a type of
distrotion, they really mean group delay VARIATION. That is a change
in delay vs frequency.

It is similar top gian and frequency response. Gain is not bad.
Change in gain vs frequency is bad. Group delay is not bad (in this
context it is simply a time delay) Group delay variation with
frequency can change the waveform. Most people agree that small
amounts of group delay variation vs frequency are not audible. Large
amounts of group delay variation obviously are, they can change a click
into a chirp.

In short, if you start with a click, flat group delay will simply delay
the click in time. ie. 1 second of flat group delay dealys the click 1
second but does not otherwise change it's sound. Group dealy VARIATION
vs frequency can change the click into a chirp. A CHANGE in group delay
vs frquency of 1 second from 20 Hz to 20 kHz will change the click into
a chirp.

Mark
March 13, 2005 12:38:35 PM

Archived from groups: rec.audio.pro (More info?)

Chris Hornbeck wrote:
> On 13 Mar 2005 06:54:35 -0800, "Mark" <makolber@yahoo.com> wrote:
>
> >No Actualy group delay, envelope delay and delay by recording and
> >playback are all the same thing.
>
> We'll have to disagree about this.
>
> But I still want to know where the name comes from, if it's
> not too far over my head.
>
> Chris Hornbeck

Group delay comes from the name for the delay of a group of closely
spaced frequencies. These closely grouped frequencies create an
envelope like the carrier and sidebands of an AM signal create an AM
modulated carrier hence the name envelope delay. Both group delay and
envelope delay are the same thing.

http://ccrma.stanford.edu/~jos/filters/Group_Delay.html

http://www.dspguru.com/info/terms/filtterm/index2.htm

http://doc-telecom.enst-bretagne.fr/doc-telecom/terme.j...

Mark
March 13, 2005 1:57:08 PM

Archived from groups: rec.audio.pro (More info?)

Chris Hornbeck wrote:
> On 13 Mar 2005 09:38:35 -0800, "Mark" <makolber@yahoo.com> wrote:
>
> >Group delay comes from the name for the delay of a group of closely
> >spaced frequencies. These closely grouped frequencies create an
> >envelope like the carrier and sidebands of an AM signal create an AM
> >modulated carrier hence the name envelope delay. Both group delay
and
> >envelope delay are the same thing.
>
> Thanks. That makes sense. I guess I'd assumed the name
> came from some definite mathematical derivation. Shows
> to go ya.
>
>
> >http://ccrma.stanford.edu/~jos/filters/Group_Delay.html
> This one gives a similar derivation to yours.
>
>
> >http://www.dspguru.com/info/terms/filtterm/index2.htm
> This is one of those wacky computer programmer ideas of
> filtering that assumes things like "we'd like that filter's passband
> phase to be as linear as possible with respect to frequency. In other
> words, we'd prefer the filter's group delay to vary as little as
> possible in the passband."
>
> Not from an audio background probably.
>
>
> >http://doc-telecom.enst-bretagne.fr/doc-telecom/terme.j...
> This is empty on my browser (Netscape 7.1)
>
> Thanks for your thoughts,
>
> Chris Hornbeck

Thank you
Mark
Anonymous
March 13, 2005 5:43:16 PM

Archived from groups: rec.audio.misc,rec.audio.pro (More info?)

Bob Pit wrote:
> Hi
>
> I am not very familiar with CoolEdit and I want to playback my voice (or any
> mp3 file) with a 90 degree phase offset between the 2 channels/speakers. It
> must be easy but how do you do this?
>
> Thanks
> Bob
>
>

Ultrafunk makes a phase plugin. Don't remember if it's ActiveX or
VST, nor can I vouch for it beyond the demo phase.

--
Les Cargill
Anonymous
March 13, 2005 5:45:50 PM

Archived from groups: rec.audio.pro (More info?)

On Sun, 13 Mar 2005 04:33:13 -0800, "William Sommerwerck"
<williams@nwlink.com> wrote:

>There is a difference between true delay (such as recording a signal and playing
>it back later) and what I call "envelope shift" (what most people call group
>delay). I find the whole issue thoroughly confusing, and wish someone would
>study it carefully and thoroughly (if no one has already) and publish an article
>on it.

I don't know what envelope shift is, but group delay is just an
expression for the variations in time delay at various frequencies.
What I find confusing is where the name comes from, so I hope
that gets included in the article.

Chris Hornbeck
Anonymous
March 13, 2005 6:19:36 PM

Archived from groups: rec.audio.misc,rec.audio.pro (More info?)

Bob Cain wrote:

>
>
> Phil Allison wrote:
>
>> AFAIK there is no way to produce a quadrature version of a given
>> audio signal.
>
>
> Then modify what you know.
>
>
> Bob

Aren't the two elements of an X/Y pair roughly
+/-90 degrees with respect to each other?

That's not exactly "quadrature" per se... i
think of quadrature as more 'a discrete
transducer kinda thing....

--
Les Cargill
Anonymous
March 13, 2005 6:19:37 PM

Archived from groups: rec.audio.misc,rec.audio.pro (More info?)

Les Cargill wrote:
> Bob Cain wrote:
>
>>
>>
>> Phil Allison wrote:
>>
>>> AFAIK there is no way to produce a quadrature version of a given
>>> audio signal.
>>
>>
>>
>> Then modify what you know.
>>
>>
>> Bob
>
>
> Aren't the two elements of an X/Y pair roughly
> +/-90 degrees with respect to each other?

That's a different kind of angle, a spatial one. It's a
measure of the direction in which they point. The kind of
angle we are talking about here is a time delay in a signal
measured as a fraction of a cycle at any given frequency.

Quadrature is just short for 90 degrees or pi/4 in radians
and this is measured, again, for any given frequency as a
portion of it's cycle (360 degrees or 2*pi radian per
cycle.) This means that the time an angle represents varies
with the frequency.


Bob
--

"Things should be described as simply as possible, but no
simpler."

A. Einstein
Anonymous
March 13, 2005 6:38:15 PM

Archived from groups: rec.audio.pro (More info?)

On 13 Mar 2005 06:54:35 -0800, "Mark" <makolber@yahoo.com> wrote:

>No Actualy group delay, envelope delay and delay by recording and
>playback are all the same thing.

We'll have to disagree about this.

But I still want to know where the name comes from, if it's
not too far over my head.

Chris Hornbeck
Anonymous
March 13, 2005 6:46:37 PM

Archived from groups: rec.audio.misc,rec.audio.pro (More info?)

"Les Cargill" <lNOcargill@cfl.Arr.com> wrote in message
news:cSYYd.139174$qB6.11555@tornado.tampabay.rr.com


> Aren't the two elements of an X/Y pair roughly
> +/-90 degrees with respect to each other?

Nope. Intensity stereo has outputs that are perfectly in phase, and differ
only in amplitude.
!