Actual Experience- Super CMIT- cicadas.... ?

[johngooch]

After a pretty significant cicada season here in the northeast-  i have the question, how has Super CMIT fared against this type of nasty BG noise....  Anybody have any positive reports?

[Philip Perkins]

OK--maybe time to fire up the Cedar DNS2

[Constantin]

Not cicadas, but a couple of months ago, we shot in some marshlands. And there were frogs there, lots and lots of frogs. They have the most amazing ability to stark and stop all as one. The must have some sort of conductor. Our AD got quite good at shouting at them, and sometimes they would actually stop. Of course, that's not how he tells that story ("the master of frogs"). But the SuperCMIT did not have a chance. At least not on the first setting. With that kind of background, setting 2 is unsuitable for a moving mic, so we never tried it. Besides, the actors stopped the take anyway, as it just got too distracting

[MartinTheMixer]

It makes the most beautiful, clear cicada recordings.

 

[johngooch]

That is what I thought .....

[MartinTheMixer]

I mean, if you turn the gain way down, put a sound blanket over there head and stick the mic in their mouth, you might be ok.

[Marc Wielage]

I seem to recall our host, Mr. Wexler, has some stories on how he dealt with very loud crickets on the film The Last Samurai. (The solution was more chemical than digital.)

I once worked on a commercial where the AD had a blank gun and just fired a few rounds up in the air (we were out in the woods), and the cicadas immediately stopped for about a minute or so each time. But this may not be possible in all areas.

[Constantin]

I once worked on a commercial where the AD had a blank gun and just fired a few rounds up in the air (we were out in the woods), and the cicadas immediately stopped for about a minute or so each time. But this may not be possible in all areas.

And if you're lucky, the shot frightens all the dogs in the area and they start barking

[johngooch]

Not always possible.  I have clapped two pancakes together in the past and r worked for a bit. But that was painfully loud and distracting.  

[MartinTheMixer]

5 minutes ago, johngooch said:

Not always possible.  I have clapped two pancakes together in the past and r worked for a bit. But that was painfully loud and distracting.  

Just throw the pancakes into the woods, the dogs will go in the woods to eat the pancakes. The dogs in the woods will make the crickets be quiet, and they won't bark (the dogs, not the crickets,) and voy luh.I think I would use waffles maybe, because the little holes would snag branches and keep the dogs busy for longer.

[jrd456]

This is no joke,low pounding frequencies scare the little creatures and they shut-up.Pound a 6"x6" post on the ground and they think it's a Mastedon [elephant] and they shut-up------it works !

                   I'm not crazy----J. Dehr

 

                            

[johngooch]

Got it.  So no actual experience with super cmit and these awfully noisy insects....   They are very much a moving target in the sense is that they change in volume and pitch almost constantly - it would not surprising if s cmit was not effective at all but I thought I would ask to confirm my suspicions.  

 

Thanks.

[Marc Wielage]

On 9/1/2016 at 11:21 PM, Constantin said:

And if you're lucky, the shot frightens all the dogs in the area and they start barking

There is that problem. And lawn mowers and leaf blowers.

The reality is that the Super-CMIT is not really different from using iZotope or the other noise-reduction devices that sample background noise then apply an inverse-phase notch against the main signal to try to filter it out. The problem with trying to do this with broadband stuff / variable-level like crickets and cicadas is that the artifacts will kill you.

I just did an iZotope job the other day where I was able to precisely slice out a "beep-beep" backup truck out of an interview for an editor I work with, and he was absolutely thrilled. That was a case where we could surgically remove the problem with no problem, since it was a very narrow frequency. But insects... that ain't gonna work. 

[berniebeaudry]

Izotope works great on loud bird chirps too!

 

[Philip Perkins]

Short pitched sounds that have a limited number of harmonics are fresh meat to RX.  Broadband noises like HVAC are ok but just ok, and noises like cicadas which are both broadband and pitched are tough for any system.  I was serious about Cedar.

[stevegrider]

I was on a shoot a month ago in Indiana and the cicadas were there and ready to party. I activated my brand new CEDAR DNS2 and gave it a try. As nice as it is, it cannot remove cicadas like a post production program like Izotope can. What I have been able to use it for thus far is reducing outdoor traffic low end rumble and indoor ac noise. I don't really think of it as a noise reducer per se, more like a really sophisticated digital EQ. I set it up on my cart so it only processes the boom channel, and I provide the boom clean as an iso and the boom effected as an iso. I used the unaffected boom track for the mono mix.  I know it is expensive, and there are many that say I shouldn't be monkeying around with noise reduction in the field, but if I have the extra track and post doesn't mind, why not? It is another nice tool to have in the tool kit. 

[Philip Perkins]

I meant Big Cedar, like in post.  I haven't had much luck with RX on things like cicadas--too artifacty.

[neeco13]

Hi Folks. I live and work in the south of France and cicadas are part of the soundscape all summer here. On stage, one of the better way to get rid of their sound is to have an assistant with a hose pipe. Between each take, he waters the tree from which cicadas are singing (in french we say they sing). As it cools down the temperature, cicadas stop singing for a while (they don't make noise when temparture is below 25° C). As a re-recording mixer, I'd rather use a post-production tool such as Izotope RX.

[Constantin]

Between each take, he waters the tree from which cicadas are singing

Won't that result in dripping water sounds?

[MartinTheMixer]

4 hours ago, Constantin said:

Won't that result in dripping water sounds?

I wouldn't think so, that Cicada sound carries 40 yards, I don't think you could hear drops of water dripping from more than 10, maybe 15 yards. Dripping little drops of water are pretty quiet.

[Constantin]

I wouldn't think so, that Cicada sound carries 40 yards, I don't think you could hear drops of water dripping from more than 10, maybe 15 yards. Dripping little drops of water are pretty quiet.

That's assuming that Cicadas are some distance away, but they might be right next to the set.
After some rain, it's usually trees which will shed water in large drops after the raining has stopped. That's not quiet at all and also depends on what they fall onto.
Nevertheless, it seems to work for Neeco, so...

[Glen Trew]

Single Cicada clicking sounds are very different from the pulsating drone of millions of synchronized cicadas in group-thinkj mode. My guess is that one is easier than the other to reduce with a digital algorithm device. The DNS-2 is a relevant comparison to the Super CMIT, in that the Super CMIT is basically a less sophisticated DNS-2 that is in constant learn mode, and with just two intensity options.

[cineli]

...not to forget that the SuperCMIT process is a very specific algorithm using two signals (the main capsule and the extra rear one). The goal is first to cancel the diffuse field (opposed to the direct one) which is totally not the concept of the DNS-2 !  A large group of far away cicadas can effectively create such a diffuse field...but cicadas are cicadas. Just remember the horrible vuvuzelas in South Africa world football cup: the very first 50 SuperCMIT microphones were very promising...but vuvuzelas are vuvuzelas !

[MartinTheMixer]

9 hours ago, Constantin said:

That's assuming that Cicadas are some distance away, but they might be right next to the set.
After some rain, it's usually trees which will shed water in large drops after the raining has stopped. That's not quiet at all and also depends on what they fall onto.
Nevertheless, it seems to work for Neeco, so...

I guess I was picturing the scene. I mean that if we are talking sound for picture, then if you have water dripping from trees in your scene on camera, I would think you would have to hear the drops the camera sees falling. Cicadas sound carry about 10 times or more, the distance that water drops do. 

[Constantin]

if you have water dripping from trees in your scene on camera, I would think you would have to hear the drops the camera sees falling. Cicadas sound carry about 10 times or more, the distance that water drops do. 

Yes, of course. If that tree is on camera. Usually though, whatever I do to improve the sound of a scene, is out of the frame and hopefully stays there