The No.1 Website for Pro Audio
 Search This Thread  Search This Forum  Search Reviews  Search Gear Database  Search Gear for sale  Search Gearslutz Go Advanced
Anatomy of a poor room / shotgun mic test Noise Reduction & Restoration Plugins
Old 16th September 2015
  #1
Gear Addict
 

Exclamation Anatomy of a poor room / shotgun mic test

If you have been following my occasional posts, you might know that much of my recording is done in a room that is pretty lousy. It is 15x12 meters and ceiling is about 5.5 meters. Lots of hard surfaces. What's worse, a very loud ventilation fan in the ceiling of which the "off" switch is currently broken, In addition the compressor of a skating rink is in the next room.

See attached pics for reference of the room and my test setup.

During most concerts I have to set up around 8 to 9 meters behind the conductor. In this acoustic that does not make for good recording.

I have been trying to find a way to make the best of what I have. Today I took the time and my trumpet to go there and do some testing with nobody else around (I have a key to the place).

I was playing my trumpet roughly where the gray chair is. The band would be there. Mics are visible in one of the pictures in the opposite corner where I need to be during a concert. Audience in front of the mics.

I set up a Sennheiser MKH805 (T powered, mid to late 1960's) long shotgun mic (I know you aren't really supposed to use shotgun mics indoors). Two Advanced Audio CM1084's with supercardioid capsules, 60cm spaced and 40 degrees angle. And for reference a CM1084 with omni capsule. The MKH805 and omni were pointed straight forward.

In the attached files I used only the left hand supercardioid of the pair. So strictly speaking this is at a slight angle from the omni and MKH805 shotgun. I don't think there is a significant difference, because the right hand cardioid sounds just about identical (to me).

These have been amplified so they read the same LUFS. Other than that no editing. I have also left in the sound of the ventilation (sounds like a jet engine, almost ).

Results.
The rumbling sound is the ventilation (loud, isn't it). This is, for the most part at least, not self noise of the mics. It is loudest in the omni, by far. Measuring a noise only section at the end, it comes in at -29.91 LUFS. No surpise there. The MKH805 (this is a 50 year old microphone, mark you) is -34.06 LUFS and the CM1084 supercardioid is even quieter at -38.88 LUFS.

I feel that the omni picks up all the nastiness of the acoustics here. The supercardioids do a lot better in that respect I think. It is my impression that the MKH805 is slightly drier than the supercardioids.

Opinions? (other than: "Find a new room"; I wish, but it just isn't an option for the band).

Regards, Christine
Attached Thumbnails
Anatomy of a poor room / shotgun mic test-p1040401.jpg   Anatomy of a poor room / shotgun mic test-p1040412.jpg   Anatomy of a poor room / shotgun mic test-p1040398.jpg  
Attached Files

CM1084 omni (V) Result 1.mp3 (3.18 MB, 958 views)

CM1084 super (V) Result 1.mp3 (3.18 MB, 934 views)

MKH805 Result 1.mp3 (3.18 MB, 955 views)

Old 16th September 2015
  #2
Lives for gear
Sure it's not a great room but it's not like you have any alternatives at this point, so might as well accept it and try to work around it.

An omni is going to reveal all the flaws in this place, including reflections and the extended bass in the A/C noise, so unless there is any reason to retain bass below 60 Hz (tuba plus bass drums are going to sound a little lighter in weight as a result) then a cardioid or supercardioid are a good direction to move in. However, the offaxis sounds those mics do pick up are going to be uglier than the on axis ones !

Why not try to live with the noise on location and simply do some noise reduction on the resulting audio files...here is a very quick attempt with Waves X Noise, and I'm sure programs like Rx Izotope and other big guns will give you an even better result.

Here I've simply tried to minimize the artifacts from a fairly 'enthusiastic' single pass, you may get better results by doing a spectral analysis of the noise and then doing several less damaging passes and thus retaining more of the harmonic content. It's not a hopeless case by any means...although it will never be a Concertgebouw !

With a whole band playing this level of background noise will be quite acceptable once you can get some effective noise reduction enacted on your recorded files ?
Attached Files

MKH805 noise reduced.mp3 (1.04 MB, 968 views)

Old 16th September 2015
  #3
Gear Addict
 

Quote:
Originally Posted by studer58 View Post
Sure it's not a great room but it's not like you have any alternatives at this point, so might as well accept it and try to work around it.
Agreed, and hence my test session this afternoon

Quote:
Originally Posted by studer58 View Post
Why not try to live with the noise on location and simply do some noise reduction on the resulting audio files...here is a very quick attempt with Waves X Noise, and I'm sure programs like Rx Izotope and other big guns will give you an even better result.

Here I've simply tried to minimize the artifacts from a fairly 'enthusiastic' single pass, you may get better results by doing a spectral analysis of the noise and then doing several less damaging passes and thus retaining more of the harmonic content. It's not a hopeless case by any means...although it will never be a Concertgebouw !

With a whole band playing this level of background noise will be quite acceptable once you can get some effective noise reduction enacted on your recorded files ?
I try not to invest too much in not-for-free software because those things tend to have a shorter lifespan than for example a microphone. However, knowing the noise characteristics of this room I did purchase a copy of Izotope RX4, for the purpose of noise reduction.

Three passes through RX4's noise reduction and a low cut at 70Hz makes a huge difference (this is the recording as above; music starts after 5 seconds). By the way, the banging and shuffling you might hear is in the hallway behind the black curtain. It is used by the restaurant next door for resupplying and such. And no, they can't hear the band practicing.

Edit: To my ears though, you can hear the ventilation when I am playing, and then when I breath it suddenly cuts off because of the noise reduction. Makes it sound a bit... artificial?

Regards, Christine
Attached Files

CM1084 super (V) Result 1 (RX4).mp3 (3.17 MB, 882 views)

MKH805 result 1 (RX4).mp3 (3.17 MB, 878 views)


Last edited by connloyalist; 16th September 2015 at 05:51 PM..
Old 16th September 2015
  #4
Lives for gear
 
Bruce Watson's Avatar
 

Quote:
Originally Posted by connloyalist View Post
I feel that the omni picks up all the nastiness of the acoustics here. The supercardioids do a lot better in that respect I think. It is my impression that the MKH805 is slightly drier than the supercardioids.

Opinions?
Oy. I think you might want to sacrifice some bottom end in exchange for some clarity. You're in a PAL country which runs in your favor since mains frequency is 50 Hz. That's where most of your motor and air handler rumble will likely be. Since you want to preserve as much musical information as you can, you'll have to experiment with your EQ. Use the steepest cut off you can (24 dB / octave if you have it), and slowly creep upward in frequency until you drop the rumble sufficiently. Hopefully you don't loose too much music in the process.

Thankfully there's usually not a lot of music information below around 50 Hz. Unless you've got a pipe organ in there, or a big Bosendorfer. And you're not playing the 1812 Overture with the cannons.

Another thing you can do, which is tedious and mind numbing, is do a little "site survey" around where you are forced to have your mics. These kinds of rumblings can vary quite a bit in intensity based on location. What you're looking for is a place where they cancel themselves out some. If you can find such a "sweet spot" (it's not guaranteed to exist, but I've found such spots in smaller rooms), and if you can record from there, it just gives you a somewhat better place to start.

There's also noise removal, using a noise print, and of course something like iZotope RX4, which isn't cheap and may not be worth it to you, IDK.

But... looking at that room, and the ceiling... won't they let you fly a rig? Getting out of the diffuse field and close enough where most of your signal is direct would be... helpful. But I'm sure you've been down that road with building management already.

As to microphones, I agree with your assessment. The hypers win.
Old 16th September 2015
  #5
Lives for gear
Well your RX4 certainly gets rid of the noise, however it also seems to remove the decay tail of the instrument in the room as well, so a case could be made for leaving (a bit of that) in...even at the expense of a residual of noise ? It's not a pretty reverb to be sure...but its complete absence seems a little odd to the listening brain (I guess you could always add some software 'wetness tail' later on)

I'm uncertain whether (from your description) the mics are typically located in the corner as shown...that would put them in the pressure zone for bass noise build-up, so if it's possible to move them away from walls and the corner that might be a good thing ?

You have good ambient noise removal at your fingertips, and with a little placement experimentation for the mics you will soon be at the end of the improvement road for this room. More expensive mics definitely won't improve the sound here...polar patterns, angling, height and wall distance plus post-RX4 will help though.
Old 16th September 2015
  #6
Gear Addict
 

Quote:
Originally Posted by studer58 View Post
Well your RX4 certainly gets rid of the noise, however it also seems to remove the decay tail of the instrument in the room as well, so a case could be made for leaving (a bit of that) in...even at the expense of a residual of noise ? It's not a pretty reverb to be sure...but its complete absence seems a little odd to the listening brain (I guess you could always add some software 'wetness tail' later on).
Agreed.

Quote:
Originally Posted by Bruce Watson View Post
Oy.
That's a good way of putting it

Quote:
Originally Posted by Bruce Watson View Post
But... looking at that room, and the ceiling... won't they let you fly a rig? Getting out of the diffuse field and close enough where most of your signal is direct would be... helpful. But I'm sure you've been down that road with building management already.
Actually, they might... I will have to give that some thought, discuss it with someone. That would be kind of an AB50 kind of thing with cardioids or supers?

Quote:
Originally Posted by Bruce Watson View Post
As to microphones, I agree with your assessment. The hypers win.
Just out of curiosity, does the unorthodox use of a (long) shotgun here give a plausible result?
Old 16th September 2015
  #7
Lives for gear
 
PatrickFaith's Avatar
 

great post, i have similar to worse room and noise issues. i've gone with schoeps ccm41 to cover the room ... basically almost mini shot gun mics which i position them to avoid problems, i love them and i think they are very musical. i then just do a few spot mics which i mix with the ccm41s.
Old 16th September 2015
  #8
Lives for gear
 
Bruce Watson's Avatar
 

Quote:
Originally Posted by connloyalist View Post
Actually, they might... I will have to give that some thought, discuss it with someone. That would be kind of an AB50 kind of thing with cardioids or supers?
That Grace space bar you're using can be flown more or less like it is (well, upside down -- click on the link below for a picture). So whatever mic setup you would normally use should work fine.

Assuming building management is amenable to the idea of flying mics: It looks like the part you need is SB-HB-30/66, which should be the hanging bar kit that matches your space bar. I've never used it, but it looks as if it only needs two mount points in your hall. But there are many ways to fly a stereo bar; work with building management and use what they give you. As long as you can get the bar at the height you want it, and pointed in the direction you want, you'll be good to go. Sounds so easy doesn't it? Running the cables might be... interesting.

There's a reason people would much rather use mic stands. But in this hall if flying the rig would get me out of that corner and into a proper position to record, I'd be willing to jump through all the hoops required. Because getting the mics into a better position trumps just about everything else when it comes to sound quality.
Old 17th September 2015
  #9
Gear Addict
 
voltronic's Avatar
 

Christine - I feel your pain. I'm a public school music teacher, and all of the auditoriums I record in have loud HVAC - my own building is far louder than that on your recording if you can believe it. I have to use RX on every school recording, even in our large high school auditorium which is relatively quiet as schools go, but when I've used omnis the noise is all over the place.

If I may, I have done lots of experimentation with RX and have arrived at what I think is a good starting point for reducing exactly this kind of noise before it becomes "gated" sounding as in the examples you posted:

1. Start with the "Rumble Reduction" preset, make sure it's set to Spectral / Manual, and grab your noise profile.
2. Activate the "Reduction Curve" (this step is quite important I've found).
3. Set quality to D (best) and leave all other settings alone.
4. Tweak the Reduction amount until it reduces the noise acceptably without starting to sound un-natural. I try not to go higher than 6-12, but for your MKH track I needed to bump it up to 16. In the choir recordings I do, I start to get artifacts on the consonants that high, so I'm usually at 6-9. I err on the side of slightly more noise in the results, if it means the music sounds more natural. To my ears the "processed" sound of too aggressive noise reduction is more distracting than the noise itself.
5. Alternate previewing as normal and through the "output noise only" setting while you work so you're not grabbing the parts of the audio you want to keep.

I've found RX to be the only tool that I can get acceptable results with this sort of thing. It's quite amazing what it can do, but for those of us who have to record in such places it's still a challenge. Good luck!
Attached Thumbnails
Anatomy of a poor room / shotgun mic test-rx4denoise.jpg  
Old 17th September 2015
  #10
Lives for gear
I agree with voltronic entirely, and he's made the case with far more precision than I was able to !

If you can reduce the background noise to convey a sense that 'there's some machinery in a nearby room which I can just hear' while leaving the natural decay of the instrument/band, you'll be achieving a worthy result.

A "gated anechoic chamber" won't serve/save the music...it's like Photoshop gone crazy !
Old 17th September 2015
  #11
Gear Addict
 

Voltronic, excellent advice. I had never used that reduction curve option, but the Rumble Reduction preset does indeed work very well. When listening to the noise only, I start to hear myself playing above 9, so that's where I left it. Which especially on the MKH805 file does leave some audible noise. But you are right, better that than the gated sound.

I followed it up with the "Remove all low energy" preset in the Equalizer section to take out everything below 44.8Hz (I wonder why they chose that number?).

For the curious, results attached.

Thank you!

Regards, Christine
Attached Files

CM1084 super (V) Result 1 (RX4-2).mp3 (3.17 MB, 749 views)

MKH805 result 1 (RX4-2).mp3 (3.17 MB, 758 views)

Old 17th September 2015
  #12
Lives for gear
 
Bruce Watson's Avatar
 

Quote:
Originally Posted by connloyalist View Post
Voltronic, excellent advice. I had never used that reduction curve option, but the Rumble Reduction preset does indeed work very well. When listening to the noise only, I start to hear myself playing above 9, so that's where I left it. Which especially on the MKH805 file does leave some audible noise. But you are right, better that than the gated sound.

I followed it up with the "Remove all low energy" preset in the Equalizer section to take out everything below 44.8Hz (I wonder why they chose that number?).

For the curious, results attached.
Results indeed. This is quite an improvement on where you started. Like the others, when using noise reduction software I always try to avoid any audible artifacts. If that means some low level of residual noise remains, so be it. I'm generally happy with less noise and don't need zero noise.
Old 17th September 2015
  #13
Gear Addict
 
voltronic's Avatar
 

Quote:
Originally Posted by connloyalist View Post
Voltronic, excellent advice. I had never used that reduction curve option, but the Rumble Reduction preset does indeed work very well. When listening to the noise only, I start to hear myself playing above 9, so that's where I left it. Which especially on the MKH805 file does leave some audible noise. But you are right, better that than the gated sound.

I followed it up with the "Remove all low energy" preset in the Equalizer section to take out everything below 44.8Hz (I wonder why they chose that number?).

For the curious, results attached.

Thank you!

Regards, Christine
Glad this worked for you. It's a definite improvement. In both the original and processed recordings, I strongly prefer that old Sennheiser shotgun. You should share that with the folks on Taperssection.

I can't remember how I came across this combination of settings in RX, but somehow it does the trick. Ironically, I can't get good results starting with the "remove hiss" or "remove rumble and hiss" presets - both give me artifacts even with manual adjustments and small amounts of reduction.

I've had mixed results with "remove all low energy" but it seems to have worked out here. Sometimes it's easier to just put on a highpass EQ and manually adjust cutoff and Q by ear to further get rid of rumbles. What made the biggest difference for me was upgrading my headphones to a set of sealed Sennheiser HD380s with very low bass extension. Now I hear all kinds of extreme-LF junk in old recordings that I get to go back and fix.

I have to ask: was your choice of "Misty" for this test an expression of how frustratingly difficult it is to try and make a decent recording in rooms like this?

Look at me...
I'm as helpless as a kitten
Up a tree...
Old 18th September 2015
  #14
Gear Addict
 

Quote:
Originally Posted by voltronic View Post
Glad this worked for you. It's a definite improvement. In both the original and processed recordings, I strongly prefer that old Sennheiser shotgun. You should share that with the folks on Taperssection.

I can't remember how I came across this combination of settings in RX, but somehow it does the trick. Ironically, I can't get good results starting with the "remove hiss" or "remove rumble and hiss" presets - both give me artifacts even with manual adjustments and small amounts of reduction.

I've had mixed results with "remove all low energy" but it seems to have worked out here. Sometimes it's easier to just put on a highpass EQ and manually adjust cutoff and Q by ear to further get rid of rumbles. What made the biggest difference for me was upgrading my headphones to a set of sealed Sennheiser HD380s with very low bass extension. Now I hear all kinds of extreme-LF junk in old recordings that I get to go back and fix.

I have to ask: was your choice of "Misty" for this test an expression of how frustratingly difficult it is to try and make a decent recording in rooms like this?

Look at me...
I'm as helpless as a kitten
Up a tree...
I too prefer the sound of that old Sennheiser shotgun in this test. That was a large part of the objective of this test, to see how it does in this room. I was able to pickup two of these on eBay for a good price (someone was selling three of these and I decided to get two, since if I liked it you could be sure I wouldn't have been able to find a second one). Not knowing whether or not it would work for me I bought one T-power adapter and shockmount. I have decided to go ahead and buy the stuff required to use the second one as well (the Rycote INV-BH shockmount might be a challenge, they seem to be somewhat difficult to find).

Good idea, I will post this on TS.

As for Misty... No, the choice of playing that has nothing to do with the song itself. It's just something I can play without sheet music. Whenever someone says: "Play something" or I just need something plausible I use that. It's my standard "filler".

Regards, Christine
Old 18th September 2015
  #15
Gear Addict
 
voltronic's Avatar
 

Quote:
Originally Posted by connloyalist View Post
I have decided to go ahead and buy the stuff required to use the second one as well (the Rycote INV-BH shockmount might be a challenge, they seem to be somewhat difficult to find).
If you're in the US, they are in stock at B&H and Adorama. Expensive, but my experience with the other Rycote shockmounts tells me they're probably worth it.

You may be interested in this BBC test report from 1966.
Old 18th September 2015
  #16
Gear Addict
 

Quote:
Originally Posted by voltronic View Post
If you're in the US, they are in stock at B&H and Adorama. Expensive, but my experience with the other Rycote shockmounts tells me they're probably worth it.

You may be interested in this BBC test report from 1966.
I had seen it at B&H, but I am in Europe. There ought to be a better way to find one of these than have it go from the UK to the US and then back to the Netherlands . I will either ask one of the places around here that sells Rycote if they can special order this, or else e-mail Rycote directly.

I agree, Rycote isn't the cheapest option but so far in my experience well worth the money.

Yes, I had found that BBC report. An interesting read.

Regards, Christine
Old 21st September 2015
  #17
Lives for gear
 
Earcatcher's Avatar
Quote:
Originally Posted by connloyalist View Post
I had seen it at B&H, but I am in Europe. There ought to be a better way to find one of these than have it go from the UK to the US and then back to the Netherlands . I will either ask one of the places around here that sells Rycote if they can special order this, or else e-mail Rycote directly.

I agree, Rycote isn't the cheapest option but so far in my experience well worth the money.

Yes, I had found that BBC report. An interesting read.

Regards, Christine
Noyzboyz.nl will be able to supply you with it pretty quickly.
Old 21st September 2015
  #18
Gear Addict
 

I found a place in the UK that has them in stock. One is one the way to me now. I did "discover" Noyzboyz recently when I was looking for something, and have put them on my list of audio webshops to check whenever I am looking for something. Thanks

Regards, Christine
Old 21st September 2015
  #19
Lives for gear
If you look on the BB List or EBay you will find MKH 805s complete with original suspension and basket for little money
They only work in big spaces and they do need a wind screen for external use
MKH 816 are superior and still cheap but are XLR rather than Tuchel.(and some P48)
Old 21st September 2015
  #20
Gear Addict
 

Quote:
Originally Posted by Rolo 46 View Post
[...] MKH 816 are superior and still cheap but are XLR rather than Tuchel.(and some P48)
Do you happen to know the difference between the MKH805 and MKH815? (I am assuming the 815 and 816 are identical except for the Tuchel/XLR connector).

Regards, Christine
Old 21st September 2015
  #21
Lives for gear
815 was tighter and quieter with TPower XLR
816 was matt black for tv/film and XLR P48 and T
I did many films with the 816, in difficult conditions it produced results similar to good radios, in the 80s it was difficult to run more than 4 radio mics together, so a god send.
Old 2nd October 2015
  #22
Gear Addict
 

The other day my second T power adapter and cable came in, so this afternoon I did a stereo shotgun mic test. Same location and basic setup as before. The MKH805 shotgun mics were spaced 50cm apart and straight ahead (no angle between them).

I moved around the place from center, to left and then over to right and back to center while I played.

As it turns out the 50cm spacing leaves it overwide and causes some negative numbers on Adobe Audition's phase indicator, especially when I move left. Probably because I am closer to the wall there (first file). So I went in with MSED and narrowed things a bit (second file). Next time I will try 40cm spacing. Files have no other processing except normalizing to -1 dB.

By the way, the background noise is NOT microphone self noise, but the ventilation system.

Regards, Christine
Attached Files

MKH805 stereo test.mp3 (2.59 MB, 465 views)

MKH805 stereo test (narrower).mp3 (2.59 MB, 432 views)

Old 2nd October 2015
  #23
Lives for gear
 
Bruce Watson's Avatar
 

Cool. Progress indeed. Good on ya for thinking out of the box. I probably would never have tried shotguns for this.
Old 2nd October 2015
  #24
Lives for gear
 
boojum's Avatar
For someone burdened with a concrete pit for a venue where all music is accompanied by the HVAC I'd say you are doing great. This never will be a good venue but you are making it a better venue against all comprehension. Congrats!
Old 2nd October 2015
  #25
Lives for gear
 
hbphotoav's Avatar
 

What Boojum said ^^^

Well done! I've made good use of my MKH8060 in similar situations... It's definitely not a one-trick-(on-camera)-pony.

HB
Topic:
Post Reply

Welcome to the Gearslutz Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Forum Jump
Forum Jump