Never put mic on camera?

[jcs]

I think everyone is getting a little stressed on this thread. It's time for one of JCS's videos to relax everyone. From his Cosmic Flow series: https://www.youtube.com/watch?v=M5J9N1xLHRg. I wonder how it was mic'ed. Was that a parabolic or a mic on the camera?

Of course that was mic on camera (Schoeps CMIT5U), don't have a parabolic dish though now I want to try one
Primary noise source was camera left, so perpendicular to mic axis- max rejection. It was really loud, so even close mic in this case wouldn't have made a huge difference in signal to noise ratio.
I did try to reduce the noise in post using PP CC's built-in adaptive NR etc. (also did a nice 4K upscale (1080p 12-bit 444 RGB source) which greatly improved video quality watching on YouTube):

The best it gets is at around 15s onward, however I prefer the natural noise better. If sound needed to really sound better, I would have used Audition and possibly Izotope plugins etc. which are much much better than the builtin PP CC filters.

As far as people getting stressed, what is stress, what is anger? Does it have something to do with ego and controlling yourself or others? Instead of getting stressed or angry when challenged, why not let go and laugh instead if you disagree with someone, then politely state your view? Try it and see if you feel better!

Does close-mic Schoeps CMC641 boomed overhead in a quiet studio sound better, sure it does, signal to noise ratio is much higher! (plus maybe this'll cheer some folks up):

The outdoor video Paul F posted was good enough- no one on YouTube or Instagram had any issues with sound quality. Have you ever listened to PewDiePie's audio haha? As was discussed earlier in this thread, the purpose of the recording is important, not religiously following rules stated by others without thinking for yourself. And if it doesn't matter, why get your nickers in a twist?

 

[Noiz2]

...
A simpler, more helpful comment for beginners would be: as distance from microphone source increases the signal to noise ratio decreases. In other words, as we move the mic closer to the source, the speaker's voice will sound increasingly louder compared to all other sounds. This isn't "better" or "worse", any more than changing treble or bass is good or bad, it's what's the purpose of the recording? For some shots we want background noises to be louder when the speaker's voice is supposed to be farther way.

Sound quality doesn't magically decrease with distance: only signal to noise ratio changes. Many focused on the NFL examples possibly using lavs, but didn't pay attention to the animal recordings with parabolic mics: confirmation bias holding on to "mic on camera is 'bad'".

THere are a couple of "errors" here. One that you slightly mention is that it's not just signal to noise, the "quality" of the sound changes significantly. You can in post sort of make close sound distant but you can not make distant sound close. So if you record at a distance it is always going to sound distant. That maybe OK but you should know going in and a LOT of new filmmakers don't.

Second, yes if you want a more distant sound then a more distant mic makes some sense. But only IF your story takes place on a no budget film set. Because that is the extra "ambient "sound you will be picking up.

Which is why a film with a budget will always close mic actors. You can cheat it more distant and you get to put in the ambience that fits the story rather than being stuck with the ambience of the set.

 

[Noiz2]

What's the point of this thread? Don't listen to so-called experts or pros. Think for yourself, and if 'experts' don't make sense, do what's right for you. Don't blindly listen to me either (except think for yourself)! I don't mind being challenged, and there were many challenges in this thread. I learned new things. Hopefully others did too and enjoyed the process.

This needs some singling out.

I can roll with the "so called pros" but on this list most of the "pros" are in fact pros. There is a great quote that relates."If you don't learn from history you are doomed to repeat it". Sound pros have mostly learned their trade in the field. They don't say "don't do X" because some textbook says so they say it because they tried it the other way and it failed. People should definitely listen to their advice.

It is WAY too easy to start a "credit" war, but I would say to people take a look at what posters have actually done and rate advice by how their work fits what you want to do.

There are people who want to make infomercials. That is valid and if you find a poster who does infomercials you like then listen to them. If you want to make features then listen more to posters who have done features. There are posters from just about every walk of life on this forum so add some weight to the ones that do what you want to do.

I tried to be really neutral there because the reality is that IF you want to do infomercials it is not necessarily in your best interest to listen to folks who do features, and visa versa.
But it can be a bit hard to tell who is who. So check them out. Usually the name is in the profile and IMDB is your friend. Or they have links in their signature.

Personally I do a lot of narrative features and a lot of narrative shorts, some documentaries and a ton of ads. So that is where I come from. I don't do a lot of production sound or much "reality" or ENG, and almost no Infomercial. And I have done a ton of theatre. I am an actual professional, BUT there are whole areas where I have very little experience. Like infomercial and event videography.

 

[jcs]

THere are a couple of "errors" here. One that you slightly mention is that it's not just signal to noise, the "quality" of the sound changes significantly. You can in post sort of make close sound distant but you can not make distant sound close. So if you record at a distance it is always going to sound distant. That maybe OK but you should know going in and a LOT of new filmmakers don't.

Second, yes if you want a more distant sound then a more distant mic makes some sense. But only IF your story takes place on a no budget film set. Because that is the extra "ambient "sound you will be picking up.

Which is why a film with a budget will always close mic actors. You can cheat it more distant and you get to put in the ambience that fits the story rather than being stuck with the ambience of the set.

We had a long discussion about this, physics etc., remember? It was also verified by both combatentropy's real-world experience shooting outside and getting perfect audio from a camera-mounted shotgun and parabolics, remember? Parabolic mics also prove that sound does not magically degrade with distance. Other than losses due to friction, however at distances useful to a parabolic mic, it's not significant. Did you listen to the animal recordings? (since the NFL clips probably have lavs in there). Anyone with access to an anechoic room (or perhaps even a booth) can verify this as well. The issue is signal to noise ratio, which is related to other (unwanted) sounds and reflections. That's it. By what physical property does sound degrade by distance (other than friction), and what are the spectral characteristics of degradation with distance? Again, excluding reflections and the near-field effect. I'm open to learning new things, but you're going to need to use math and physics or real-world recordings to make your point.

In the future we'll have even better technology than simple parabolic dishes to record excellent quality at distance. For example, for cameras we're coming up on computational photography, which allows for any lenses and DOF to be simulated in post. Already it's possible to take an ambisonic recording and simulate basic microphone sound fields in post. Along with machine learning and other technologies it will be possible to filter out the reflections and other sounds not in the desired sound field to provide much better quality than parabolics can do today.

 

[jcs]

This needs some singling out.

I can roll with the "so called pros" but on this list most of the "pros" are in fact pros. There is a great quote that relates."If you don't learn from history you are doomed to repeat it". Sound pros have mostly learned their trade in the field. They don't say "don't do X" because some textbook says so they say it because they tried it the other way and it failed. People should definitely listen to their advice.

It is WAY too easy to start a "credit" war, but I would say to people take a look at what posters have actually done and rate advice by how their work fits what you want to do.

There are people who want to make infomercials. That is valid and if you find a poster who does infomercials you like then listen to them. If you want to make features then listen more to posters who have done features. There are posters from just about every walk of life on this forum so add some weight to the ones that do what you want to do.

I tried to be really neutral there because the reality is that IF you want to do infomercials it is not necessarily in your best interest to listen to folks who do features, and visa versa.
But it can be a bit hard to tell who is who. So check them out. Usually the name is in the profile and IMDB is your friend. Or they have links in their signature.

Personally I do a lot of narrative features and a lot of narrative shorts, some documentaries and a ton of ads. So that is where I come from. I don't do a lot of production sound or much "reality" or ENG, and almost no Infomercial. And I have done a ton of theatre. I am an actual professional, BUT there are whole areas where I have very little experience. Like infomercial and event videography.

What's wrong with questioning someone if what they are saying doesn't make sense, regardless of their credentials? What's wrong with thinking for yourself, and trying new things? Sometimes you'll discover something new, sometimes you'll understand why someone else said something, and truly understand the principle behind the advice, vs. wrote memorization with no ability to generalize to other forms of problem solving? If man didn't question so-called "experts" science would never move forward! History has shown ego is a huge problem in academia, slowing down progress.

Regarding "ALL sound is 3D"- I'm not sure what you meant by that. By 3D, I meant 3D spatialized audio where the listener can perceive sound cues in 3D space, around one's head. For example surround sound playback or 3D HRTF playback on headphones. There's a significant perceptual difference in mono, stereo, surround sound and spatialized 3D audio (HRTF for headphones).

By repeating list item (7) "There are no rules" that meant don't blindly listen to me either- challenge away if it doesn't make sense to you.

 

[Noiz2]

You didn't "prove" anything but to yourself. Parabolic mics make a rather large change in the sound so not sure what you are talking about on that count. It is why they are NEVER used for dialog. I completely agree with the not "blindly" listening to someone. However if I ask advice and I get one answer from someone with no experience and one from someone who has decades of experience and has maybe done the exact thing I want to do then I'm probably going with the person who knows what they are talking about.

As to 3D sound? I didn't say anything about that in what you quoted so? If you are listening with out headphones you are listening in 3D space. More speakers will let you place sound with more specificity but you are always listening in 3D.

The whole HRTF and the rest is really much more of a video game thing that a film thing, though the atmos system may change that.
The reason is simple physics (I know you don't believe so called science either, but for the rest of us)... With out a huge number of speakers and a lot of work you end up with too big a difference in audience experience. Truthfully I doubt that in my life that will actually get solved but? Anything you place will change it's location depending on where you are sitting.

Now with headphones everyone is sitting in the same seat so it can be really cool, but in a theatre it generally doesn't work well. You can get a "wow" factor but if you can't control it the same for every seat then it becomes a lot of work for very little bang.

BTW "no rules", literally, does not mean "don't blindly listen to me either- challenge away if it doesn't make sense to you". If you had said the later you would not have gotten pushback, because pretty much everyone would agree with that.

 

[jcs]

You didn't "prove" anything but to yourself. Parabolic mics make a rather large change in the sound so not sure what you are talking about on that count. It is why they are NEVER used for dialog. I completely agree with the not "blindly" listening to someone. However if I ask advice and I get one answer from someone with no experience and one from someone who has decades of experience and has maybe done the exact thing I want to do then I'm probably going with the person who knows what they are talking about.

As to 3D sound? I didn't say anything about that in what you quoted so? If you are listening with out headphones you are listening in 3D space. More speakers will let you place sound with more specificity but you are always listening in 3D.

The whole HRTF and the rest is really much more of a video game thing that a film thing, though the atmos system may change that.
The reason is simple physics (I know you don't believe so called science either, but for the rest of us)... With out a huge number of speakers and a lot of work you end up with too big a difference in audience experience. Truthfully I doubt that in my life that will actually get solved but? Anything you place will change it's location depending on where you are sitting.

Now with headphones everyone is sitting in the same seat so it can be really cool, but in a theatre it generally doesn't work well. You can get a "wow" factor but if you can't control it the same for every seat then it becomes a lot of work for very little bang.

BTW "no rules", literally, does not mean "don't blindly listen to me either- challenge away if it doesn't make sense to you". If you had said the later you would not have gotten pushback, because pretty much everyone would agree with that.

Some statements were made (paraphrased):

1. Sound degrades with distance
2. ALL sound is 3D

Note how I am focusing on the statements only and not the source of the statements. Now since you asked:

"Great summation of why so many new filmmakers films sound so bad.
Success in challenging rules requires understanding those rules and why they are rules.
But of course there never were any "rules", just best practices. If you swap those two you see how ridiculous #1 and #7 are.
#2 ALL audio is 3D
#3 when someone actually accomplished this let us know.
#4 practical experience trumps both
#5???
#6 I question the point of the thread."

The primary disagreement in this thread has been challenging the very common statement: "never put mic on camera!!". We covered the pros and cons and there was no basic disagreement that the sound does change with distance in many situations. The fundamental concept as to why this happens in reality is related to the inverse square law and signal to noise ratio, which typically doesn't get covered by the blanket statement "never put mic on camera!!".

Another statement was made that "sound degrades with distance". I have a math and physics background, I wrote that racing simulator I posted earlier in this thread. I wrote the physics engine, from scratch (did not use any libraries). This includes computational fluid dynamics for airflow, as well as particle systems. I've written 3D audio simulation systems from scratch as well. I've posted productions samples of my work, both mic on camera as well as close boomed. The quality in both cases is acceptable for the target audiences.

After thinking about the statement "sound degrades with distance" (note how I'm focusing on the statement and not who said it or their credentials), it didn't make sense to me given my background in 3D audio and physics simulations. I made the conjecture that sound doesn't automatically degrade with distance, and that if reflective surfaces are removed, sound will not significantly degrade with distance spectrally, only amplitude due to the inverse square 'law' (spherical wavefront growing with distance on the surface of the sphere, which is a 2D area, and works out to an inverse square function). Air is composed of particles (air molecules) and nothing in physics predicts any kind of spectral decay with distance. I realize computer simulations and the Navier Stokes fluid dynamics equations are beyond most people's experience. However the videos I linked with fluid simulations of waves was to show that sound doesn't fundamentally spectrally decay with distance. This can be tested in an anechoic room or sound booth by simply moving the mic away from the source and normalizing in post etc. With a clean preamp and low-noise mic, the prediction is distance won't significantly change spectral quality. combatentropy reported a use case outside in the real world with a single primary reflector: the ground. And sure enough, mic on camera at distance sounded pristine. So we don't even need an anechoic room or booth to test this theory.

I was curious if parabolic mics could also show that sound can be captured at distance with usable quality by effectively focusing on the source sound to eliminate capture reflections. Sure enough, they do just that. I first found the animal examples, however I felt spoken voice would be more compelling, so I also provided the football videos. Whiles we all know parabolic mics are used for football, it's now common practice to use lavs too, so the animal examples still provide real-world examples beyond combatentropy's report that sound does not fundamentally spectrally degrade at distance. In the future, computational audio will further improve the ability to capture quality sound at a distance. You can get a taste by experimenting with post production of ambisonic 3D audio recordings, etc.

Regarding 3D audio- it's been around for a very long time, however recently it's become more popular with the resurgence of VR. The highest quality 3D audio relies on psychoacoustics, so the brain perceives audio not just left/right (stereo) or around the head in a plane (surround sound, a 2D circle), but also above/below and front/behind: a full 3D sphere. HRTF stands for Head Related Transfer Function. In simple language this is just a special audio filter (transfer function), which spectrally filters audio based on the shape of the ear, ear canal, and sound masking/absorbing properties of the ears and head. This effects audio spectrally (multi frequency amplitudes), as well a time delay between the ears (phase). Here's an example recorded with the Sennheiser Ambeo into a Zoom F4 and processed with the free Ambisonic Toolkit and Reaper:

While this recording was output for 2 channel headphones using an HRTF, it still provides 3D spatial cues on 2 channel monitors (near field) and speakers (such as living room distance).

Regarding "no rules", I've repeated that I welcome challenges to what I say as well: that was not the first time I said that in this thread. Many "experts" or "pros" get angry when challenged, and many times they are wrong. This is in general, and not singling anyone out. The anger comes from the ego, and introspection into anger can deliver one to freedom from suffering, which manifests in many ways, including physical illness.

I brought up 3D audio in this thread as that's a perfect and exactly correct use for mic on camera for 3D 360 degree VR videos.

I encourage anyone who loves audio to learn about and experiment with 3D audio- it's very cool!

 

[combatentropy]

combatentropy's real-world experience shooting outside and getting perfect audio from a camera-mounted shotgun

I'm sorry if I made you think it was perfect. That's not what I meant. When I thought we were still talking about whether a hobby video could use an on-camera mic, I just meant I was surprised how good it sounded.

Even then I tried to qualify it by saying it had to be outside, and it had to be in a quiet neighborhood.

Now that I see you are trying to use it as scientific evidence, I feel I should tell you that happened 15 years ago. I don't still have the tape, I am retelling it all from memory, and I am not a professional sound guy. I don't have golden ears now, and I was even stupider back then.

 

[jcs]

I'm sorry if I made you think it was perfect. That's not what I meant. When I thought we were still talking about whether a hobby video could use an on-camera mic, I just meant I was surprised how good it sounded.

Even then I tried to qualify it by saying it had to be outside, and it had to be in a quiet neighborhood.

Now that I see you are trying to use it as scientific evidence, I feel I should tell you that happened 15 years ago. I don't still have the tape, I am retelling it all from memory, and I am not a professional sound guy. I don't have golden ears now, and I was even stupider back then.

It's all good man. I understood your point and the conditions, and it makes perfect sense based on physics. The parabolic mic examples of far away animals sound excellent for anyone with an open mind who cares to listen. Thanks for sharing your experience.

 

[Greg_E]

Parabolic mics make a rather large change in the sound so not sure what you are talking about on that count. /QUOTE]

Having used many different parabs, in different situations, made from different materials, and with different curves, and different microphones in and around the focus point, I have to strongly agree with this statement. It again comes down to physics and the changing focus point of each wavelength. They work great for hearing "stuff" from football players clashing at the line, but suck for any dialog that isn't just a surveillance level where you just need to hear what is said. Also in order to get decent gain, the dish get's big.

 

[Noiz2]

It's all good man. I understood your point and the conditions, and it makes perfect sense based on physics. The parabolic mic examples of far away animals sound excellent for anyone with an open mind who cares to listen. Thanks for sharing your experience.

A couple of things.

I am not arguing over simulations. For those to be really useful in the "real world" you would have had to model way more things like random wind, the guy who starts up his lawnmower three blocks away etc.

So in an idealized perfect mathematical world you are probably right. Next time I'm recording in the Matrix I will remember that.

In the actual world where folks work close is better 99.99999% of the time.

Now you made an important points as kind of a toss away.

I've posted productions samples of my work, both mic on camera as well as close boomed. The quality in both cases is acceptable for the target audiences.

The "acceptable for the target audiences" being the operative words, because for many uses including narrative film work the sound of your examples is not at all "acceptable for the target audiences". And I have said this over and over, that it is very important to know what you need to end up at to know where you have to start.

For your audience camera mounted may be fine, no argument from me - you know your audience.

Air is composed of particles (air molecules) and nothing in physics predicts any kind of spectral decay with distance.

What are you using "rules"! Air is composed of many things and it changes. Depending on what your "air" consists of it affects the sound more or less. That change is compounded by the amount of that "air" that the sound wave travels through. Very damp air rolls off the highs a bit and the more air you travel through the more that roll off affects sound.

And sure enough, mic on camera at distance sounded pristine.

I have no idea what you think "pristine" sounds like. Nothing you have posted would I call "pristine" so?

Regarding "no rules", I've repeated that I welcome challenges to what I say as well: that was not the first time I said that in this thread. Many "experts" or "pros" get angry when challenged, and many times they are wrong. This is in general, and not singling anyone out. The anger comes from the ego, and introspection into anger can deliver one to freedom from suffering, which manifests in many ways, including physical illness.

Wow I grew up just out side Berkeley CA and that sounds "SO California" even to me.

Professionals, and I don't think you will find any of the professionals here calling them selves "experts", don't angry at challenges. They get challenges all the time. They get impatient with folks who have little or no practical experience telling people the world works like X because I read some stuff and played with the numbers. Professionals don't like that because they have years of actual experience and despite the numbers "real life" is a lot messier than a computer simulation.

Nobody who has any amount of experience hasn't had to change their thinking and understanding of how things work a ton of times. Recording, mixing, editing, has all changed vastly in the last twenty years.

 

[combatentropy]

They work great for hearing "stuff" from football players clashing at the line, but suck for any dialog that isn't just a surveillance level where you just need to hear what is said.

Thank you, this is something I wanted to say, but left off for the sake of brevity. The requirements for dialog are a 100 times more stringent than sound effects, birds chirping, etc.

For example, this sample from GaryNattrass in post #2 works, with crowd noise recorded by a mic at the sidelines, https://youtu.be/66oTdtBrAxE
But your opening example with the nice lady talking actually bothers my ears, because it is dialog.

This is psychology now, not physics. You opened with an example that was Youtube, not professional work, so of course everyone can only say, "Well, I guess you can just use the on-camera mic." However, even in non-professional circumstances like that, I would recommend you try to use a lav or boom. Here are good reasons not to: (1) I am a student and just want to get started practicing, (2) My cat was being funny, and I had to whip out my phone, (3) I am a reporter on the battlefield.

To me, this is a bad reason: (1) I didn't feel like it.
I think if there's anyway you can, boom that nice shotgun or hyper 1.5 ft. above head.
Else if you can't do that, clip on a lav.
Else if you can't do that, reconsider whether you have to shoot it now.

I'm just saying this for the sake of your audience --- your general, non-pro audience. They will appreciate it, even if only subsconciously. They will be distracted when not, again even if only subconsciously.