Become a Site Supporter and Never see Ads again!

Author Topic: XY Greater than 90 Degrees?  (Read 6797 times)

0 Members and 1 Guest are viewing this topic.

Offline kuba e

  • Site Supporter
  • Trade Count: (1)
  • Taperssection Member
  • *
  • Posts: 445
  • Gender: Male
Re: XY Greater than 90 Degrees?
« Reply #45 on: March 30, 2022, 04:59:57 AM »
:coolguy:

I have a stereo mic or two (card patterns @ 100'), and have observed a image improvement since running them that way.  It's like what you can accomplish with a smidgen of Ozone Imager width in post.

Yes, it's similar. Ms for xy is a natural (we change the ratio of the pressure and pressure gradient component of the microphone capsule) and is lossless. It is always possible to change the ratio to the original state. Imager makes artificial adjustments and it cannot be returned to its original state. For xy is the best ms, it is the clearest edit. (The Imager never sounded good to me personally, maybe only very small stereo changes sounded good)
« Last Edit: March 30, 2022, 08:32:29 AM by kuba e »

Offline Gutbucket

  • record > listen > revise technique
  • Trade Count: (15)
  • Needs to get out more...
  • *****
  • Posts: 14692
  • Gender: Male
  • "and the rowers keep on rowing!"
Re: XY Greater than 90 Degrees?
« Reply #46 on: March 30, 2022, 10:17:53 AM »
I haven't used the Ozone Imager and am not certain how it works, but most stereo imaging tools convert L/R to M/S, make a Mid/Side ratio adjustment, then convert M/S back to L/R again.  Some of them are mulitband, spliting the frequency range into multiple bands prior to doing that with each band, allowing one to affect a different ratio modification for highs verses mids verses lows, for example.  In that way a multiband imaging plugin is doing something very similar to EQing Mid/Side, only across specific bands rather than as a continuous EQ curve.

Conversion from L/R to M/S and back again without any ratio change made is looses no information.   And if one changes the ratio by a reasonable amount, and leaves that ratio change in place (which is what we are talking about here for the most part), there is no effective loss of signal to noise ratio in the resulting recording..  However if one wanted to revert the modified recording to its original ratio again for some reason, and performs another conversion from L/R to M/S and back to make a ratio modification that is the inverse of the first one, doing so will return the ratio to what it originally was, however the noise floor in whichever channel was reduced will increase by the amount of that reduction.  In practical terms this will rarely matter.  For one thing we could just go back to the original files.. and even if we did want to do this it probably wouldn't be noticeable with reasonable ratio modifications. 

But, think of it this way- If you were to convert from L/R to M/S, then change the ratio to 100% Mid or 100% Side, you are completely eliminating the other channel and there is no getting it back when you convert to L/R again.  Same as if you were to pan a stereo recording all the way Left then record that output, there is then no way to pan back to center again later, you've thrown away all Right channel information by panning fully Left.  Changing M/S ratio or panning (which is changing L/R ratio) by lesser amounts which don't eliminate the other channel completely will reduce signal to noise ratio by the amount of the change if you later undo that change.
musical volition > vibrations > voltages > numeric values > voltages > vibrations> virtual teleportation time-machine experience
Better recording made easy - >>Improved PAS table<< | Made excellent- >>click here to download the Oddball Microphone Technique illustrated PDF booklet<< (note: This is a 1st draft, now several years old and in need of revision!  Stay tuned)

Offline checht

  • Trade Count: (5)
  • Taperssection Member
  • ***
  • Posts: 494
  • Gender: Male
  • Old and in the Way
Re: XY Greater than 90 Degrees?
« Reply #47 on: March 31, 2022, 12:26:34 AM »
Thanks all for the super informative thread. Learned a lot reviewing it.
I recorded many shows from '83-'92 using km84i's xy PAS and was pretty happy with the results. Biggest issue was that indoor large rooms often sounded distant and vocals got lost.
I've remastered a few using rx music rebalance and adding more of the vocal stem back into the mix. Good result.
Thinking to try converting to ms and adjusting ratio and then comparing.
Some of those recordings from outdoor venues like the Greek or Frost sound amazing 😀

Thanks again!
Schoeps MK41s > nbob KCY >
Naiant PFA 60v > Sound Devices MP-6 II  or  Naiant IPA > Roland R-07
Vanguard V1s matched pair
Recordings at LMA

Offline kuba e

  • Site Supporter
  • Trade Count: (1)
  • Taperssection Member
  • *
  • Posts: 445
  • Gender: Male
Re: XY Greater than 90 Degrees?
« Reply #48 on: March 31, 2022, 10:32:03 AM »
You are right, I was wrong. Ms is not lossless. I forgot the noise and also the high frequencies. At high frequencies, there will be a phase shift because small gap between the capsules in xy. Maybe it's in higher up 20kHz. I think Imager is widening stereo by phase shift. It can also turn mono into stereo with it's algorithm.

I am still thinking about changing the tonality with ms eq. It's not easy at all for me. I'm sorry, the following is just a theoretical consideration. I am only thinking loudly.

I've probably figured out what confused me. Ms eq can be analyzed using the basic relationship:
left = mid + side
right = mid - side

Eq of mid channel manifests itself equally in the left/right channel. We cant say how much, because left and right are composed of mid and side. Mid eq can be corrected by an additional left/right eq for overall tonal balance. (I would guess that the left/right eq correction will be the opposite of what we used for mid eq. It is because how behave the side eq.) The side eq is a different case. We add side channel to the left and subtract from the right. So the overall tonal balance should be maintained in left/right channel. This would then mean that if we equalize the side, we do not have to make additional adjustments. There is one exception. When the side is louder than the mid (mid - side is negative value, this does not happen in xy, only with a strong ms eq) then the right and left channels are in opposite phases. And overall tonal balance is not maintained.

I would be happy if somebody correct me. We can also look at ms eq from point of view of changing virtual directional patterns and angles. And patterns and angels shouldn't cause a change in tonality on their own.
« Last Edit: March 31, 2022, 10:40:05 AM by kuba e »

Offline Gutbucket

  • record > listen > revise technique
  • Trade Count: (15)
  • Needs to get out more...
  • *****
  • Posts: 14692
  • Gender: Male
  • "and the rowers keep on rowing!"
Re: XY Greater than 90 Degrees?
« Reply #49 on: March 31, 2022, 01:00:11 PM »
Easiest just to think of it in practical terms-

Use Mid/Side EQ as a more advanced way manipulating Mid/Side ratio of a coincident pair, by frequency range.  That affects image width and spaciousness.  It will also affect the frequency balance of the center of the image differently than the sides.  Those two aspects are interlinked.

Then use regular EQ to adjust the overall frequency balance (both center and sides together) to whatever you want.

You might go back and forth in an iterative way when adjusting the two EQ instances to home-in on exactly what you want.


To more specifically address your points [my comments below in italics]..
Eq of mid channel manifests itself equally in the left/right channel. [as does EQing the Side channel]

Mid eq can be corrected by an additional left/right eq for overall tonal balance. (I would guess that the left/right eq correction will be the opposite of what we used for mid eq.) [the EQ applied for tonality should be whatever sounds right. If using it specifically to 'undo' some unwanted tonal change imparted by whatever Mid/Side EQ was applied to manipulate the spatial qualities, it will address the tonal changes imparted by both Mid and Side EQ channels]

The side eq is a different case. We add side channel to the left and subtract from the right. So the overall tonal balance should be maintained in left/right channel. This would then mean that if we equalize the side, we do not have to make additional adjustments. There is one exception. When the side is louder than the mid (mid - side is negative value, this does not happen in xy, only with a strong ms eq) then the right and left channels are in opposite phases. And overall tonal balance is not maintained. [This is incorrect.  Side channel has opposite polarity in Left and Right channels, but that does not affect its tonal qualities.  There can be more Side information than Mid from an X/Y pair, which is likely to happen as you increase the angle of Blumlein to wider than 90 degrees for instance.. if you were to use a 180 degree  X/Y angle with fig-8's the result would be 0% Mid /100% Side]

We can also look at ms eq from point of view of changing virtual directional patterns and angles. And patterns and angels shouldn't cause a change in tonality on their own. [For all practical purposes, as soon as you start EQing things you are going to create tonal changes.  As mentioned earlier in the thread, with M/S a lot of variables are interrelated. Ambisonics breaks those interrelations, making it in some ways easier to mentally relate to theory and the mathematic concepts]
musical volition > vibrations > voltages > numeric values > voltages > vibrations> virtual teleportation time-machine experience
Better recording made easy - >>Improved PAS table<< | Made excellent- >>click here to download the Oddball Microphone Technique illustrated PDF booklet<< (note: This is a 1st draft, now several years old and in need of revision!  Stay tuned)

Offline kuba e

  • Site Supporter
  • Trade Count: (1)
  • Taperssection Member
  • *
  • Posts: 445
  • Gender: Male
Re: XY Greater than 90 Degrees?
« Reply #50 on: April 01, 2022, 03:00:41 PM »
Thank you very much Gutbucket. It is very kind from you to comment my thinking. I need some time to understand it deep. This thread is very helpful.

Yes, I totally forgot that there is negative lobe from hypercardiod  to fig 8. I was wrong, xy can have opposite signals.

 

RSS | Mobile
Page created in 0.034 seconds with 28 queries.
© 2002-2022 Taperssection.com
Powered by SMF