Become a Site Supporter and Never see Ads again!

Author Topic: Oddball microphone techniques  (Read 55822 times)

0 Members and 1 Guest are viewing this topic.

Offline Gutbucket

  • record > listen > revise technique
  • Trade Count: (13)
  • Needs to get out more...
  • *****
  • Posts: 11537
  • Gender: Male
Re: Oddball microphone techniques
« Reply #330 on: November 06, 2017, 05:42:17 PM »
Yes, that's exactly what I'd like to try. In the previous post, you mentioned that by mixing the rear microphone, we can create a similar effect like playback with halfler rear speaker. I tried Halfler's rear speaker and I was very surprised how it changed the sound compared to the classic stereo playback. I felt like I was in the middle of the audience, the stereo has spread a lot. If I could get this feeling at least a bit into the recording by mid/side mixdown of rear mic, it would be great.

What the Halfler playback technique does is send the difference signal (L-R) to the additional rear speaker or to a pair of rear speakers.  That difference signal is the Side in Mid/Side stereo and tends to be made up of the reverberant/ambient content in the recording.  So what you are doing is a Mid/Side stereo playback modification, spreading the Side information out around the sides and back.  BTW, using two speakers works better than just one as that spreads out that stuff more diffusely in the playback space.  You reproduce the Mono sum of both channels (Mid) as a phantom image between the Left and Right speakers, Left and Right signals eminate from Left and Right speakers, and the difference sum from the rear speaker(s) with a 6dB difference between each adjacent quadrant.  In essence you have 4 channel LRCS reproduction of a 2-channel L/R stereo signal.  This is similar to the basis on which matrix surround decoding works, such as Dolby ProLogic/PL2/PLX and all the other matrix surround techniques, but without the additional tricks intended to forcibly steer surround information, so Halfler can sometimes work better for straight ambiance retrieval and doesn't mess with the purity of the left and right signals sent directly to those speakers.

What comes out of that rear Hafler speaker(s) is whatever content in the recording is different on each side.  So anything panned full left or right will not only be reproduced in either the Left or Right speaker but also the rear speaker(s).  More importantly for live recordings, besides level differences, phase and polarity differences are also reproduced from the rear channels.  So any content with inverted or randomized phase between Left and Right stereo channels gets more diffusely reproduced throughout the room, and much of that tends to be the audience and ambient content in the recording.

One of the things spacing the omnis far apart enough does, which is further enhanced by having the front/rear mics non-coincident and located in different positions in space, is randomize the phase of the diffuse ambient/reverberant content which ends up in the recording.  When that stuff gets reproduced in such a way such that it is spread more evenly throughout the room it leads to the open and immersive listening experience you've noted of feeling like you are in the middle of the audience. 

This is a direct tie-in to the low diffuse field correlation stuff I was talking about a week or so back when I revived this thread.  The Halfler technique is a method of leveraging this on playback more than 2-speaker playback can do on it's own.  The oddball microphone techniques help in leveraging it on the recording side more so than two channel recording typically does alone, in addition to also leveraging other things.  The Mid/Side mixing technique for the rear facing microphone routes reproduction of the rear facing microphone content to the far sides and to the Hafler speaker(s) in back if it is in use.  Without that Mid/Side mixing technique, that rear facing mic content is still good and useful in the mix, but will tend to be reproduced from the front of the room between the Left and Right speakers along with the stuff the forward facing microphone picked up rather than from the far sides and rear.
volition > vibrations > voltages > numeric values | numeric values > voltages > vibrations > virtual teleportation time-machine experience

Offline Gutbucket

  • record > listen > revise technique
  • Trade Count: (13)
  • Needs to get out more...
  • *****
  • Posts: 11537
  • Gender: Male
Decorellation of Audio Signals and Its Impact on Spatial Imagery
« Reply #331 on: November 09, 2017, 02:38:37 PM »
Before we moved on to other stuff, I'd revived this thread with a discussion of decorrelation, starting with this post about 4 pages back - https://taperssection.com/index.php?topic=96009.msg2244172#msg2244172

Here is an excellent paper on the topic I recommend to anyone interested in digging deeper into the effects of decorrelation on stereo signals - The Decorrelation of Audio Signals and Its Impact on Spatial Imagery - Gary Kendall (1995) - http://www.garykendall.net/papers/Decorrelation1995.pdf
volition > vibrations > voltages > numeric values | numeric values > voltages > vibrations > virtual teleportation time-machine experience

Offline kuba e

  • Site Supporter
  • Trade Count: (1)
  • Taperssection Regular
  • *
  • Posts: 79
  • Gender: Male
Re: Oddball microphone techniques
« Reply #332 on: November 13, 2017, 05:01:53 AM »
I tried the oddball technique for the first time. It's great technique, it gives us plenty of options to influence the final sound of the recording. The ability to change the ratio of direct sound and ambience, and add depth is perfect.I'm sure more people will start using this configuration. I'm not good at post processing, but I see this mics configuration is big help to achieve great results. Last weekend I made a trip to Munich in Germany for the Gov't Mule concert. The music was great and there were six tapers too. It was great to meet them. They were interested in this technique too. I sended them a link to this thread.

Unfortunately I came late to the club, the lights were off, no photo went well. At least I took the picture of the club at the end of the concert when they turned on the lights. And I took the picture of the microphone stand at home later. We were setup at sbd about 50 feet from the podium. The mics configuration was:
the forward & rear cards: nak300, spacing 13 inch
spaced omni: ca11, spacing 4 feet.
https://drive.google.com/drive/folders/1MKdygDhUAO1tSdNMz6pkwXxuCXRUsrBO?usp=sharing

Post processing is very difficult for me. I'm not sure at all. It was difficult to mix the recording, but I believe it will be better and easier next time. What surprised me very much was when I mixed the front and rear microphones in M/S. The result is very good. M/S is dominant in the final mix, omni are only slightly added there. When I added more omni, it did not sound good. Maybe I spaced them too much. Could I add more omni in the mix if they are less spaced?

I'm sending a sample. In the end, I did not choose it well, omni stereo is not well balanced there. (In the first set it was better balanced, maybe the stand was moved little. I balanced the second set omni by the gain and short delay of the left channel when mixing the final mix.) But I hope it would be enough to illustrate. There is sample of Hans's recording too. He had the stand right next to me with MBHO cards, din. It's for comparsion how it sounds single pair.
https://drive.google.com/drive/folders/1MLNWhfUh9eTgUgE22V2rcJH6QsG_Zpch?usp=sharing

Gutbucket, thank you for your explanations and for reference to the theory. I love to read it. It will take me a while, but then I will write back. I was thinking about the little thing relating to the theory when I was processing M/S. When there is a large part of S and M is weak, then much of the signal in the right and left channels is in the opposite phase. Is the partial opposite phase in the left and right signal what makes special sound in M/S (maybe blumlein too)?

Medeski and those guys should have no problems with recording AUD as long as the room allows it you should be AOK!
I'm just a bit jealous as I haven't seen Medeski in a couple of years and always enjoy his side projects (sometimes more than MMW)
I also like very much their side projects. It would be difficult to make decisions when I could choose. But I would probably go to Medeski and friends, Wicked Knee or Wood Brothers than MMW. Perhaps this is because their side projects can be seen in small clubs and in a very relaxed atmosphere.
« Last Edit: November 13, 2017, 06:28:16 AM by kuba e »

Offline Gutbucket

  • record > listen > revise technique
  • Trade Count: (13)
  • Needs to get out more...
  • *****
  • Posts: 11537
  • Gender: Male
Re: Oddball microphone techniques
« Reply #333 on: November 13, 2017, 10:54:37 AM »
Glad you got a chance to try these techniques and are enjoying doing so.  And thanks for the report, photos and samples.

Post processing is very difficult for me. I'm not sure at all. It was difficult to mix the recording, but I believe it will be better and easier next time. What surprised me very much was when I mixed the front and rear microphones in M/S. The result is very good. M/S is dominant in the final mix, omni are only slightly added there. When I added more omni, it did not sound good. Maybe I spaced them too much. Could I add more omni in the mix if they are less spaced?

I've downloaded your samples and will play with them myself when I get the chance. [question- in the front/back pair recording which is in the left and which is in the left channel?] I did a quick listen to each source on this computer and I don't think less omni spacing would make for an improvement or allow you to use more of them in the mix.  Listening to the omni pair alone I don't hear any problem with excessive spacing, but do want to make some significant EQ corrections.  I suspect that's mostly what you are hearing more than an effect from the microphone spacing. 

I typically start mixing that way with the omnis- EQ and level balance the omni pair alone to best effect, then add the center pair.  But you can also do it the other way around if your recording is stronger that way- make the Front/Back, Mid/Side pair primary and bring in the omnis in support of that.  The EQ on the omnis might then be somewhat different, possibly with less mid emphasis (getting more of that from the M/S pair), rather than boosted lows and mids with a carefully cut "mud and boom" region in-between, typically centered somewhere around 300Hz. Omnis can be tricky and commonly need some EQ manipulation to work around the room resonances in that bass region when indoors in a strongly PA amplified room. The top end EQ often depends more on the characteristic of mics and how smooth they sound up there.  I'd have to play around with your samples before commenting there and expect your omnis may want only a bottom and mid emphasis with some top end reduction and the aforementioned cut around 300Hz.

Quote
I was thinking about the little thing relating to the theory when I was processing M/S. When there is a large part of S and M is weak, then much of the signal in the right and left channels is in the opposite phase. Is the partial opposite phase in the left and right signal what makes special sound in M/S (maybe blumlein too)?

Yes, there are definite parallels there.  One of the interesting things to me is how sufficiently randomized phase of the diffuse room sound (via sufficient mic spacing) sounds similar to having inverted polarity in each channel (via Blumlein or Mid/Side with sufficient Side content), even though one relies on spacing and the other is coincident.

Thanks for the post!
volition > vibrations > voltages > numeric values | numeric values > voltages > vibrations > virtual teleportation time-machine experience

Offline Gutbucket

  • record > listen > revise technique
  • Trade Count: (13)
  • Needs to get out more...
  • *****
  • Posts: 11537
  • Gender: Male
Re: Oddball microphone techniques
« Reply #334 on: November 13, 2017, 12:10:49 PM »
Embedding kuba e's mic setup photos in the thread-


volition > vibrations > voltages > numeric values | numeric values > voltages > vibrations > virtual teleportation time-machine experience

Offline Gutbucket

  • record > listen > revise technique
  • Trade Count: (13)
  • Needs to get out more...
  • *****
  • Posts: 11537
  • Gender: Male
Re: Oddball microphone techniques
« Reply #335 on: November 13, 2017, 12:11:41 PM »
Front/back pair detail-

volition > vibrations > voltages > numeric values | numeric values > voltages > vibrations > virtual teleportation time-machine experience

Offline rocksuitcase

  • Trade Count: (0)
  • Needs to get out more...
  • *****
  • Posts: 4063
  • Gender: Male
    • RockSuitcase: stage photography
Re: Oddball microphone techniques
« Reply #336 on: November 13, 2017, 12:42:00 PM »
kuba e: my input regarding mixdown:
Mostly follows Gut's input. I usually load all four channels into the Processing software (I use Audacity). then I focus on the stereo pairs. With the Omni pair, recorded in an indoor room, typically do a lot of listening (speakers, headphones, car speakers) then decide on a EQ scheme. Often, the boominess inherent with the Omni needs some low frequency reduction. that 300Hz region typically needs reduction, I tend to start the reduction at 400Hz then down all the way to 20 Hz. The main idea here is to make the Omni pair sound good with less low frequency boominess. I typically don't worry about anything above 3KHz with the Omni pair as those frequencies get emphasized with the middle card pair.
I then change the card channels to two channel mono labeling the front and rear channels properly (I am NOT dealing with MS option here). I then take the rear mic and lower it by 6-10 dB as compared to the front facing mic. Then I start blending the front and rear into the mix, setting the balance between omni sides and front/rear using the solo or mute buttons. You then have a 4 channel mix with the omni pair EQ'd and can either apply EQ to the front/rear channels as you see fit, OR mix down to two channels then EQ the resulting mix.
I don't do the MS mixing I leave that explanation to Gutbucket.
« Last Edit: November 13, 2017, 03:32:36 PM by rocksuitcase »
music IS love

When you get confused, listen to the music play!

Mics:         AKG460|CK61|CK1|CK3|CK8|Beyer M 201E
Recorders:Marantz PMD661 OADE Concert mod; Tascam DR680 MKI

Offline morst

  • Archivist: Camper Van Beethoven & Cracker
  • Trade Count: (1)
  • Needs to get out more...
  • *****
  • Posts: 2477
  • Get in touch if you wanna record Cracker or CVB!
    • Soundscape Preservation Society
Re: Oddball microphone techniques
« Reply #337 on: November 13, 2017, 02:11:37 PM »
The term "MID" is used for two different things in this conversation - the M in M/S and also Midrange. Please be sure to specify, if you don't mind, these new techniques are a bit confusing to read about anyhow!

If I get into this, I'll want more than 4 channels soon enough... may have to fire up the laptop and MOTU 828 again!?

.....I typically start mixing that way with the omnis- EQ and level balance the omni pair alone to best effect, then add the center pair.  But you can also do it the other way around if your recording is stronger that way- make the Front/Back, Mid/Side pair primary and bring in the omnis in support of that.  The EQ on the omnis might then be somewhat different, possibly with less mid emphasis (getting more of that from the M/S pair), rather than boosted lows and mids with a carefully cut "mud and boom" region in-between, typically centered somewhere around 300Hz. .....
Teams: Neumann, Bay Area Tapers, Multitrack, Pioneertown Tapers, Mac Geeks, Cassette Masters, Poster Collectors, Alumni of teams St Louis, Upper Midwest & Milwaukee / Southern Wisco

Offline kuba e

  • Site Supporter
  • Trade Count: (1)
  • Taperssection Regular
  • *
  • Posts: 79
  • Gender: Male
Re: Oddball microphone techniques
« Reply #338 on: November 13, 2017, 03:09:07 PM »
Thank you for help with eq and mix, it's very useful for me. Post processing and eq are very demanding, and there is also a need for sound sensitivity. The recording sounds bad in the speakers but I could not fix it. At least it sounds good in the headphones. I have to learn step by step. I'll try another try.

[question- in the front/back pair recording which is in the left and which is in the left channel?]
I think front mic is in the left channel. I'm not sure about one hundred percent. When I set my microphones, I did not mark it. I'll mark it next time.
« Last Edit: November 13, 2017, 03:11:22 PM by kuba e »

Offline Gutbucket

  • record > listen > revise technique
  • Trade Count: (13)
  • Needs to get out more...
  • *****
  • Posts: 11537
  • Gender: Male
Re: Oddball microphone techniques
« Reply #339 on: November 13, 2017, 03:36:28 PM »
The term "MID" is used for two different things in this conversation - the M in M/S and also Midrange. Please be sure to specify, if you don't mind, these new techniques are a bit confusing to read about anyhow!

Yes, good point! I'd try to say midrange when referring to frequency.

.....I typically start mixing that way with the omnis- EQ and level balance the omni pair alone to best effect, then add the center pair.  But you can also do it the other way around if your recording is stronger that way- make the Front/Back, Mid/Side pair primary and bring in the omnis in support of that.  The EQ on the omnis might then be somewhat different, possibly with less midrange emphasis (getting more of that from the M/S pair), rather than boosted lows and mids with a carefully cut "mud and boom" region in-between, typically centered somewhere around 300Hz. ..... [edited]

And agreed with rocksuitcase, except that when dialing in that upper bass/lower midrange cut (which is often significant indoors and can range anywhere from 120 to 500Hz), I almost always carefully boost the lows below that range to restore the deep bass which is a big part of the realism I feel the omnis uniquely provide.  I may cut the very lowest frequencies if its only non-musical noise, but I'm probably boosting between there and the upper bass/lower midrange reduction, else the overall solid bass which the omnis are supposed to contribute gets anemic.   With a cut on either side, that range gets pulled down along with its neighboring regions and needs a boost to bring it back to where it should be.

Of all the raw tracks you posted, that the Mid/Side mixed front/back pair sounds best to me prior to any other manipulation, so using that as your starting point here seems apt. I often bump the 2-3kHz area of my omnis for clarity and that seems relatively common for most omnis.  Its the frequency range above 3kHz where different mics seem to behave quite differently and its harder to make general statements about what to expect there.  Quite frequently I'm reducing the omni content in that region if the center mics are taking care of it nicely, yet bring the omni level back up in the high treble around 14kHz or so to get some nice open diffuse ambient air.  But that really depends on the mics and situation.  Might make sense to smoothly reduce all the highs from the omnis above 3kHz if they are bright sounding to begin with or if that content conflicts with the same frequency range provided by the center pair.

Quote
If I get into this, I'll want more than 4 channels soon enough... may have to fire up the laptop and MOTU 828 again!?

One of the keys to not going crazy with complications is keeping things manageable.  If you can do that with a laptop and interface, great.  Everyone needs to find their own comfort level, and even 4 channels is more than many tapers want to have to manage.  Nothing wrong with that.  The only way I can do 6, 8 or more channels is to set most of it up beforehand so I can pretty much just roll at the event.  For the last two open-setup recordings I posted about above, the first using the 4 mics took me about 4 times as long to arrange and setup at the show, and also break-down afterwards, compared to the 6-channel pre-wired miniature mic setup I ran a couple days later.  For that one I just popped up the stand and pointed the mics, fired up the recorder, adjusted levels and rolled.  Quick and easy on the recording side of things.  More to manage on the back side once recorded, but that's the nature of these setups.

Quote
I think front mic is in the left channel. I'm not sure about one hundred percent. When I set my microphones, I did not mark it. I'll mark it next time.
 
Thanks.  I may be able to tell by listening.  You'd think it would be obvious, but it can be amazing how similar the different channels can sound in isolation. It's the interaction between them where identity arises.  And yes, always keeping the forward facing mic in the left channel and rear-facing in the right is good practice, and it makes Mid/Side decoding straight-forward.

I've found that as the mic count increases, its extremely helpful to do a mic check either just before or just after the performance, all setup in place, snapping fingers or speaking directly into each mic to ID which channel in a short separate recorded file-set.  More often than I would have expected or care to admit, I've found accidental left/right channel pair swaps or even messier channel mis-routings I did not intend and might not have noticed otherwise.  That's another complication pre-wiring everything helps me avoid, yet I still like to do the check.  I actually keep a dog-clicker in my bag which helps with those mic IDs in loud environments.
volition > vibrations > voltages > numeric values | numeric values > voltages > vibrations > virtual teleportation time-machine experience

 

RSS | Mobile
Page created in 0.196 seconds with 32 queries.
© 2002-2017 Taperssection.com
Powered by SMF
Website Design by Foxtrot Media, Inc., a Baltimore Website Company