William Sommerwerck wrote:

One of the best phase-analysis tools is an oscilloscope -- left channel to
vertical, right to horizontal.

OK, but how does one quantify said output?

Later...

Ron capik
--

Ron Capik wrote:
William Sommerwerck wrote:

One of the best phase-analysis tools is an oscilloscope -- left channel to
vertical, right to horizontal.

OK, but how does one quantify said output?

This is what the Tek phase display does.... it's really a scope in disguise.

If a signal is mono and common to both channels, you get a diagonal line /.
If it's got reversed polarity, you get a reversed line \.
As there are phase differences between channels, the line opens up into a
circle.

With some practice you get a sense of what things should look like with
a minimalist miking set, which can help you place mikes when you cannot
trust the imaging of your speakers in the field.

You also get a sense of problems like out-of-phase bass. If the display
skews vividly northwest when there is a kick drum hit, you have a problem
trying to cut that to LP.
--scott


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

In article writes:

One of the best phase-analysis tools is an oscilloscope -- left channel to
vertical, right to horizontal.

OK, but how does one quantify said output?

With your ears.

With a fixed delay between two sources containing the same signal, the
phase difference will vary with frequency, so you can really only
quantify the phase shift at a single frequency. An oscilloscope
display of a Lissajous pattern as William describes will show you a
trend. If the display tilts toward the right, the trend is toward "in
phase." If it tilts toward the left, the trend is toward "out of
phase."

If you want to know how many degrees at a fixed frequency, you can
display the reference on one channel of a dual channel scope and the
"shifted" signal on the other channel, measure the time delay between
them, then calculate the phase by dividing the delay by the period
(1/frequency) then multiplying by 360 degrees.



--
I'm really Mike Rivers )
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 he double-m-eleven-double-zero at yahoo

On Thu, 13 Jan 2005 01:10:31 GMT, Ron Capik
wrote:

One of the best phase-analysis tools is an oscilloscope -- left channel to
vertical, right to horizontal.

OK, but how does one quantify said output?

Why do you want numbers? The display will tell you what's wrong and
when you get it right. So will your ears.

But 'scopes do have calibration marks.

CubaseFAQ www.laurencepayne.co.uk/CubaseFAQ.htm
"Possibly the world's least impressive web site": George Perfect

Laurence Payne wrote:

On Thu, 13 Jan 2005 01:10:31 GMT, Ron Capik
wrote:

One of the best phase-analysis tools is an oscilloscope -- left channel to
vertical, right to horizontal.

OK, but how does one quantify said output?

Why do you want numbers? The display will tell you what's wrong and
when you get it right. So will your ears.

But 'scopes do have calibration marks.


Well, first off I didn't say I needed "numbers", but the simple answer to
that question is: I don't know. The more complete answer is that I spent
way too many years in a physics lab documenting experimental details
before I entered the more artistic world of audio engineering. When doing
new experiments one never quite knows exactly what data will be needed
or how it may fit an evolving model.
Seems I've carried that over to my corner of the audio world. The x-y scope
shows a real time image of a complex stereo stream. Trends can be seen
in the x-y plot; the plot has an average slope and some shape. So what
information can we take away from this slope and shape? can we correlate
the slope and shape to what we are hearing?
A 45 degree line correlates to mono program material, vertical or horizontal
show just one active channel. A properly phased stereo stream would bounce
about in the first and third quadrants. The extent of excursions from 45 degrees
tells something about the stereo width (or sound stage) ....ummm, etc.
What if there's stuff in the 2nd and 4th quadrant? How much of that can be
allowed before radio stations balk? If the tracks have a temporal offset,
what would that look like?

There may even be statistical programs out there that quantify this
stuff. Don't know...
Maybe I'll find that I have an x-y shape preference for a given musical
style. Don't know...
If I don't take the data I may never know...
I'd hope the x-y scope is more useful (and subtle) than just a way to check for
polarity errors. [But maybe I'm wrong.]

OK, that's just a fraction of the odd questions floating about in my head.

Later...

Ron Capik
--

On Sat, 15 Jan 2005 03:00:04 GMT, Ron Capik
wrote:

lots of very cool thoughts

If I could add just one: the X/Y scope plot is a visual
representation of a M/S microphone's raw output.

Or, if I could add another, the stereo phono cartridge's
raw output. (Although it's geometrically matrixed, usually
((but there is at least one exception)), and was likely
invented by a guy named Bloomline, or something, so,
go figure.)

Chris Hornbeck
"Happiness isn't something you experience; it's something you remember."
-Oscar Levant

Seems I've carried that over to my corner of the audio world. The x-y scope
shows a real time image of a complex stereo stream. Trends can be seen
in the x-y plot; the plot has an average slope and some shape. So what
information can we take away from this slope and shape? Can we correlate
the slope and shape to what we are hearing?
A 45 degree line correlates to mono program material, vertical or horizontal
show just one active channel. A properly phased stereo stream would bounce
about in the first and third quadrants. The extent of excursions from 45
degrees
tells something about the stereo width (or sound stage) ....ummm, etc.
What if there's stuff in the 2nd and 4th quadrant? How much of that can be
allowed before radio stations balk? If the tracks have a temporal offset,
what would that look like?

You've figured out most of this for yourself...

Assume two mics on one instrument. If the two channels have exactly the same
amplitude and phase (ie, they're the same signal), they form a straight, thin
line at 45 degrees. Phase errors turn the line into an ellipse, which becomes a
circle at 90 degrees. I'm not sure what the visible effect of timing errors
would be, but I'm certain it would be much more complex.

The "stuff in the 2nd and 4th quadrant[s]" is the anti-phase or random-phase
components, which are principally due to ambience, reverberation, etc. Relative
errors between the channels also contribute to the 2nd and 4th quadrants.

Note that recordings deliberately encoded for surround sound will show
significantly more signal in the 2nd and 4th quadrants.

Think of the 1st and 3rd quadrants as direct sound, the 2nd and 4th as ambience.
That's an over-simplification, but it's a good starting point.

If I could add just one: the X/Y scope plot is a visual
representation of a M/S microphone's raw output.

If you think of the 45-degree motion as the M, and motion at right angles to
that as S.

In article writes:

Seems I've carried that over to my corner of the audio world. The x-y scope
shows a real time image of a complex stereo stream. Trends can be seen
in the x-y plot; the plot has an average slope and some shape. So what
information can we take away from this slope and shape? can we correlate
the slope and shape to what we are hearing?

Not much. We can tell if the trend is "wrong" but your ears can tell
you that, too, if you have decent monitoring. Sometimes, though, it's
easier to get a clue from the visual display that there's something
that you should listen for and make a subjective judgement. You can
listen to a stereo recording with the polarity of a channel reversed
and to someone who's astute, it will sound wrong. But some people who
are unaware that polarity is an issue hook up their speakers wrong
when they take the stereo out of the box and leave it that way for 20
years until they move. Someone who knows can listen to that and
recognize the problem instantly.

Where it starts becoming truly wrong is when you can no longer hear
things. If, for example, a bass is recorded with a mic and a DI
combined in mono (or even panned around the center in stereo), and
either due to position of the mic or careless wiring or routing, at
the lowest frequencies, the phase relationship of the two signals is
near 180 degrees and they're near the same amplitude, you'll get
noticable cancellation of those frequencies. But you probably already
know that. If you see the scope display tilting to the left when you
bring up the second of the two bass channels in the mix, it's time to
take a closer listen to the bass to see if you're getting the effect
you want (you may want to null out some lows). A good engineer will
hear what's happening immediately, but some may not, in the heat of
the moment.


--
I'm really Mike Rivers )
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 he double-m-eleven-double-zero at yahoo

AEA makes an XY scope for observing phase called the Winkie Blinkie.
Has anyone here used one of those? If so, how well did it work?
Thanks

In article .com,
Haolemon wrote:

AEA makes an XY scope for observing phase called the Winkie Blinkie.
Has anyone here used one of those? If so, how well did it work?

I tried it and didn't like it. The resolution isn't as high as a real
CRT. It'll fit in a bag, though.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

how about just inverting the track.

Bob

wrote in message
ups.com...
hello everyone,

let's say you 2-miked an instrument and recorded it as a stereo file
into the DAW. then when you listened back, you thought it sounded out
of phase.

would it make sense to simply de-couple the stereo file in a wave
editor and then nudge one of the left or right tracks a few samples to
get it in phase?

I was not aware the one could hear phase inless it is changing.

Bob

"Mike Rivers" wrote in message
news:znr1105461896k@trad...

In article
writes:

Yes, that will work for a PHASE problem.
It will not work on a POLARITY problem.

Hey, he said "sounds out of phase." Most people don't know what "out
of polarity" sounds like.

When you put two mics on a guitar, unless thery're coincident, there
will be a time offset between them, which can indeed sound "out of
phase." If there's a time offset as well as a polarity difference,
they'll still sound "out of phase" only different.

--
I'm really Mike Rivers )
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 he double-m-eleven-double-zero at yahoo

sycochkn wrote:
I was not aware the one could hear phase inless it is changing.

Depends on what "phase" is.

Absolute or relative polarity? Group delay in a channel? Comb filtering
effects? All get called "phasing" by people.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

In article . net writes:

I was not aware the one could hear phase inless it is changing.

You can hear polarity sometimes, and you can hear when a pair of
speakers that are supposed to work together are wired with opposite
polarity. This is not the same as phase, though it's usually called
that.

You can hear the effect of cancellation due to phase difference
between two sources, and this is what you hear when you have two mics
on the same source that are sufficiently out of phase. Of course it
will be more obvious that something's going on if it's changing, but
what you hear is the loss of level at certain frequencies - that is, a
non-flat frequency response that doesn't sound like the live source.


--
I'm really Mike Rivers )
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 he double-m-eleven-double-zero at yahoo

Mark Plancke wrote:
would it make sense to simply de-couple the stereo file in a wave
editor and then nudge one of the left or right tracks a few samples
to
get it in phase?

Yes, you can do that. Or alternatively you could try one of the
allpass filters available in a plugin form.

http://www.voxengo.com/pha979/

Mark

Does this plug essentially do (for already recorded files) what the
Little Labs Phase tool does for tracks being recorded?

Thanks

Is there software that would do this? I presume that there are scope
applications, but I've never used a real one and have always been told
that the learning curve is steep. The simplicity of a dedicated device
like the Winkie Blinkie is appealing for that reason.

how about just inverting the track.

Bob

wrote in message
oups.com...
hello everyone,

let's say you 2-miked an instrument and recorded it as a stereo file
into the DAW. then when you listened back, you thought it sounded out
of phase.

would it make sense to simply de-couple the stereo file in a wave
editor and then nudge one of the left or right tracks a few samples to
get it in phase?



Inverting the track will only work if there is a polarity problem.

The OP was asking about a phase problem. His Idea of nudging by a few samples
may work.

When recording a guitar with two microphones, I can tell which mic is closer
by looking at the waveform. If the sound is what I want, i wouldn't mess with
trying to align them.
Richard H. Kuschel
"I canna change the law of physics."-----Scotty

In article .com,
"Haolemon" wrote:

Is there software that would do this? I presume that there are scope
applications, but I've never used a real one and have always been told
that the learning curve is steep. The simplicity of a dedicated device
like the Winkie Blinkie is appealing for that reason.


I'll be releasing an AudioUnit (OSX only, sorry PC guys) that does just
this in the next couple of weeks. It's already written, but I need to
polish it up a bit and get a web page running before I "release" it.
It's OpenGL accelerated and thus very low on CPU -- easy to stick on
your master bus and glance over at while mixing without worrying about
it effecting track count. If you want to try it out with the
understanding it's a prerelease, send me an email and I'll send you a
link.

Screenshot at:
http://goldbarterholdings.com/todd/pscope.jpg

Haven't decided pricing yet. What would people pay for a simple thing
like this? $15 asking too much?

-Todd

In article writes:

I'll be releasing an AudioUnit (OSX only, sorry PC guys) that does just
this in the next couple of weeks.

Haven't decided pricing yet. What would people pay for a simple thing
like this? $15 asking too much?

Hey, it's for a Mac and Mac users got plenty of money. Make it $25.
Then make a Windows version and drop the price to $10.

(Need I say g? These days, probably so)

Is it any more than an X-Y scope display?





--
I'm really Mike Rivers )
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 he double-m-eleven-double-zero at yahoo

In article znr1106078832k@trad, (Mike Rivers)
wrote:

In article
writes:

I'll be releasing an AudioUnit (OSX only, sorry PC guys) that does just
this in the next couple of weeks.

Haven't decided pricing yet. What would people pay for a simple thing
like this? $15 asking too much?

Hey, it's for a Mac and Mac users got plenty of money. Make it $25.
Then make a Windows version and drop the price to $10.


:-D


(Need I say g? These days, probably so)

Is it any more than an X-Y scope display?


Not yet. Planning on writing a whole analysis bundle eventually -- this
plus a 1/6th octave RTA, VU meters, waterfall plot (2D or 3D), function
generator, etc. But I'd probably charge more than $15 for that.

-Todd

In article writes:

Planning on writing a whole analysis bundle eventually -- this
plus a 1/6th octave RTA, VU meters, waterfall plot (2D or 3D), function
generator, etc. But I'd probably charge more than $15 for that.

Sounds a bit like SpectraFoo (which costs considerably more than $15)
but which still isn't available for Windows. Get with the program!!!

--
I'm really Mike Rivers )
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 he double-m-eleven-double-zero at yahoo

In article znr1106102614k@trad, (Mike Rivers)
wrote:

In article
writes:

Planning on writing a whole analysis bundle eventually -- this
plus a 1/6th octave RTA, VU meters, waterfall plot (2D or 3D), function
generator, etc. But I'd probably charge more than $15 for that.

Sounds a bit like SpectraFoo (which costs considerably more than $15)
but which still isn't available for Windows. Get with the program!!!


Yup. I'm familiar with SpectraFoo, Smaart, SIM, etc. Not quite trying to
compete with them -- I'd probably go for 50-75% of the feature set and
50% of the price. The $15 price was for just the simple phase meter.
Probably more like $200-$300 for the "full suite". We'll see.

-Todd