Become a Site Supporter and Never see Ads again!

Author Topic: Processing stereo channel difference  (Read 1032 times)

0 Members and 1 Guest are viewing this topic.

Offline dyneq

  • Trade Count: (12)
  • Taperssection Member
  • ***
  • Posts: 501
Processing stereo channel difference
« on: November 26, 2019, 11:30:56 AM »
My mics were DFC between the stacks about 25' back from the stage, and I used a rough PAS with a narrow spacing (~2" @ 120*)  >:D. At the time, I could tell that the right stack was louder. Looking at the waveforms in Audacity, you can see that the right channel looks more saturated vs. the left channel (no clipping, just capturing more sound energy). Listening back, the right side is, as you'd expect, dominant.

What's the best way to balance things out so listeners receive a more equal balance of the entire spectrum? I'm looking for ideas on how best to dial in the perceived balance. I've tried normalizing with the box checked to do each channel independently and amplifying each track separately to 0.0dB. Both still have a lot more sound energy from the right side.

I can upload a sample if anyone's interested in taking a closer look.


Offline morst

  • Trade Count: (2)
  • Needs to get out more...
  • *****
  • Posts: 3791
  • ISO ALL archival Cracker / CVB / Monks Of Doom etc
    • Soundscape Preservation Society
Re: Processing stereo channel difference
« Reply #1 on: November 26, 2019, 08:17:18 PM »
In audacity you can split the track pair to left and right (see example one), and apply negative or positive gain per channel (see example 2).
Audacity uses the ever non-mac-like HIDDEN TRIANGLE MENU on the tracks.

As long as the track is given enough vertical space, the left side of each track will have a slider marked cryptically with graphics (better shown in example 2) like my lousy ASCII imitation:
 "-   -----V-----   +"
That's the channel gain. I have experienced crashes trying to set gain to non-integers. YMMV, I use v 2.1.0
Teams: Neumann, Bay Area Tapers, Multitrack,  Mac Geeks, Cassette Masters, Poster Collectors, Pro Audio
Alumni of teams St Louis, Upper Midwest & Milwaukee / Southern Wisco, Pioneertown Tapers

Offline pohaku

  • Site Supporter
  • Trade Count: (35)
  • Taperssection All-Star
  • *
  • Posts: 1014
  • Gender: Male
Re: Processing stereo channel difference
« Reply #2 on: November 26, 2019, 10:08:55 PM »
Precisely.  Split the stereo track and then adjust gain to the track as necessary. 
Mics: akg c460 (ck61, ck63), c414buls, c568eb; at4049a, 4051a, 4053a, at853; josephson c42; neumann U87, km84i; beyer m130, m160, m500; aea r84; gefell m71, mt711s, m200, m201, um70S; sony c38; schoeps cmc6, CMBI (mk4, mk21, mk41, mk4v); sennheiser mkh30, mkh40, md421, md431, md541
Pres: API, a-designs, pendulum, purple, millennia TD-1 and HV-32P, gt, littlebox, tinybox, usbpre2, CA 9200, pipsqueak, grace V2, DAV BG1
Cables: KCY, CMR, Naiant AKG actives, PFAs, asst.  GAKables and Darktrain
Recorders/converters/monitors: dr680, m10, dr-2d, d50, zoom f8, 788T SSD CL-8, lynx aurora 8, Neumann KH20


Yeah, I'm an attorney, but everyone needs a day job!

Offline morst

  • Trade Count: (2)
  • Needs to get out more...
  • *****
  • Posts: 3791
  • ISO ALL archival Cracker / CVB / Monks Of Doom etc
    • Soundscape Preservation Society
Re: Processing stereo channel difference
« Reply #3 on: November 26, 2019, 11:31:47 PM »
Precisely.  Split the stereo track and then adjust gain to the track as necessary.
If you want to adjust by peaks, you can do it this way:
Once the tracks are split as in example 2, you can use the Amplify plugin to determine maximum peak for each track.
Select a single track (or any region you choose) and go to Effect > Amplify
The number shown in the "Amplification" box represents how much gain the plugin would require in order to make that signal be 0 dB full-scale.
The maximum gain for the amplify plugin is something like +40 dB, so if it reads that it would add +40 dB, it might need more than that, but that's REALLY REALLY low.
Anyhow, if you use this to measure each channel, but click CANCEL without invoking the Amplify function, you'll get a baseline as to how much you might try boosting.
Frankly, headphones are the real test.
Teams: Neumann, Bay Area Tapers, Multitrack,  Mac Geeks, Cassette Masters, Poster Collectors, Pro Audio
Alumni of teams St Louis, Upper Midwest & Milwaukee / Southern Wisco, Pioneertown Tapers

Offline dyneq

  • Trade Count: (12)
  • Taperssection Member
  • ***
  • Posts: 501
Re: Processing stereo channel difference
« Reply #4 on: November 27, 2019, 08:14:03 AM »
Thank you both, I will give this a try.

Online goodcooker

  • Trade Count: (36)
  • Needs to get out more...
  • *****
  • Posts: 3271
  • Gender: Male
  • goes to 11
Re: Processing stereo channel difference
« Reply #5 on: November 28, 2019, 09:21:08 AM »

Adding a little bit of gentle compression after getting the gain where you want it helps to smooth things out. I use Waves L3 Ultramaximizer with very conservative settings.
Joly Mod Oktava MK012 C,H,O > Tascam DR701d


http://www.archive.org/bookmarks/goodcooker

"Are you the Zman?" - fan at Panic 10-08-10 Kansas City

"I don't know who left this perfectly good inflatable wook doll here, but if I'm blowing her up, I'm keeping her." -  hoppedup

Offline pohaku

  • Site Supporter
  • Trade Count: (35)
  • Taperssection All-Star
  • *
  • Posts: 1014
  • Gender: Male
Re: Processing stereo channel difference
« Reply #6 on: November 28, 2019, 02:29:52 PM »
You can also tweak the channel panning to smooth things out.  Try 80% or 90% instead of 100% left and right.
Mics: akg c460 (ck61, ck63), c414buls, c568eb; at4049a, 4051a, 4053a, at853; josephson c42; neumann U87, km84i; beyer m130, m160, m500; aea r84; gefell m71, mt711s, m200, m201, um70S; sony c38; schoeps cmc6, CMBI (mk4, mk21, mk41, mk4v); sennheiser mkh30, mkh40, md421, md431, md541
Pres: API, a-designs, pendulum, purple, millennia TD-1 and HV-32P, gt, littlebox, tinybox, usbpre2, CA 9200, pipsqueak, grace V2, DAV BG1
Cables: KCY, CMR, Naiant AKG actives, PFAs, asst.  GAKables and Darktrain
Recorders/converters/monitors: dr680, m10, dr-2d, d50, zoom f8, 788T SSD CL-8, lynx aurora 8, Neumann KH20


Yeah, I'm an attorney, but everyone needs a day job!

Offline dyneq

  • Trade Count: (12)
  • Taperssection Member
  • ***
  • Posts: 501
Re: Processing stereo channel difference
« Reply #7 on: November 28, 2019, 04:20:06 PM »
I calculated average RMS power for both channels. Then I amplified the quieter channel to match the louder channel’s power. I am happy with the results and will add it to my workflow.

Thanks to all who responded.

Offline morst

  • Trade Count: (2)
  • Needs to get out more...
  • *****
  • Posts: 3791
  • ISO ALL archival Cracker / CVB / Monks Of Doom etc
    • Soundscape Preservation Society
Re: Processing stereo channel difference
« Reply #8 on: November 28, 2019, 05:59:27 PM »
I calculated average RMS power for both channels. Then I amplified the quieter channel to match the louder channel’s power. I am happy with the results and will add it to my workflow.
That's a super good idea. Set levels based on RMS, then gain & limit as normal!
 :headphones: :guitarist: :headphones:
Teams: Neumann, Bay Area Tapers, Multitrack,  Mac Geeks, Cassette Masters, Poster Collectors, Pro Audio
Alumni of teams St Louis, Upper Midwest & Milwaukee / Southern Wisco, Pioneertown Tapers

Offline checht

  • Trade Count: (5)
  • Taperssection Member
  • ***
  • Posts: 333
  • Gender: Male
  • Old and in the Way
Re: Processing stereo channel difference
« Reply #9 on: November 30, 2019, 03:20:22 PM »
Question:

What about using normalize, with normalize each channel independently turned on? Perhaps that'd only get the peaks equal, not RMS?
Schoeps MK41s > nbob KCY >
Naiant PFA 60v > Sound Devices MP-6  or  Naiant IPA > Roland R-07

Offline Gutbucket

  • record > listen > revise technique
  • Trade Count: (14)
  • Needs to get out more...
  • *****
  • Posts: 13631
  • Gender: Male
Re: Processing stereo channel difference
« Reply #10 on: November 30, 2019, 04:35:36 PM »
Question:

What about using normalize, with normalize each channel independently turned on? Perhaps that'd only get the peaks equal, not RMS?

Yes, that's correct, and why balancing channel gains by ear (or RMS, which is a roughly equivalent perceptual measure to "by ear") is the way to go to achieve good balance "as heard". Balancing by RMS will usually works well, unless there is supposed to be more energy to one side than the other and you wish to retain that original image balance- like maybe an on-stage recording with the bass or drums positioned over on one side rather than being in the center.

BTW, a good quick headphone double-check for balance is to take the phones off and flip them around so that the left-side driver is on the right ear and vice-versa, or flip L/R to your monitor speakers if you can do that easily.  If your source is well balanced the stereo image will be mirror-image flipped as one would expect, yet the center image stuff like vocals should remain in the center..   assuming the signal chain though the drivers, and your hearing sensitivity is evenly balanced from side to side.
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<<

Offline opsopcopolis

  • Trade Count: (3)
  • Taperssection All-Star
  • ****
  • Posts: 1730
Re: Processing stereo channel difference
« Reply #11 on: November 30, 2019, 08:38:15 PM »
Just pan it towards the left until it feels balanced. Essentially the same operation as splitting and changing channel gain
Mics: Berliner CM-33, CA-14 card, CA-11 card & omni, AT-853, Sony ECM-907
Recorders: Tascam DR-60D, Tascam DR-05, Sony Hi-MD

Offline morst

  • Trade Count: (2)
  • Needs to get out more...
  • *****
  • Posts: 3791
  • ISO ALL archival Cracker / CVB / Monks Of Doom etc
    • Soundscape Preservation Society
Re: Processing stereo channel difference
« Reply #12 on: December 01, 2019, 03:19:34 AM »
assuming ...your hearing sensitivity is evenly balanced from side to side.
^yup. (mine is not. Calibrate appropriately)
« Last Edit: December 01, 2019, 03:21:11 AM by morst »
Teams: Neumann, Bay Area Tapers, Multitrack,  Mac Geeks, Cassette Masters, Poster Collectors, Pro Audio
Alumni of teams St Louis, Upper Midwest & Milwaukee / Southern Wisco, Pioneertown Tapers

 

RSS | Mobile
Page created in 0.134 seconds with 55 queries.
© 2002-2020 Taperssection.com
Powered by SMF