Skip to main content

Recording instruction video where teacher is talking and demonstrating on guitar. Is there any way to eliminate pick click, that is, the sound of the guitar pick plucking on the strings? It is very loud on the voice mic, and makes the sound awful.

Comments

bouldersound Fri, 02/28/2020 - 13:25

MC208, post: 463474, member: 49667 wrote: I have a RODE NTG-1. It's placed just out of the frame above the subject.
The odd thing is that I was looking at recordings made years and years ago (with the same pick and same mic) and the noise isn't there.
The only difference is the camera being used, and the location.

It sounds like you're pointing the mic at the talker's head and at the guitar, and there isn't a huge difference in distance between them. I would expect picking noised to get captured by the mic in that situation. The odd thing is the earlier video where it doesn't happen. How is he hearing his guitar? Is it by headphone/IEM? Maybe turn it up in his ears or increase the amp/sim gain so he doesn't pick as hard.

MC208 Sun, 03/01/2020 - 08:57

Well the guitar is recorded direct, it's not mic'd.
It wouldn't be possible to mute the voice mic when he's playing because he's talking while playing (teaching)
I guess the only solution is the SM58, as close as possible to his mouth without being in the shot. I can't see any other way ?

Boswell Sun, 03/01/2020 - 10:32

Although the NTG-1 has a hypercardioid response pattern, it does not help much in your situation, as it is still wide out to at least 45 degrees. The big rejection is at 90 degrees, so it would be worth trying to set up a mic position that could take advantage of that.

Keep the polar picture in your head when experimenting. For example, it just might be that placing the NTG-1 on a short floor stand out of shot in front of and below (or level with) the guitar but pointing up past it at the performer's mouth would reject enough of the instrument to make it work.

Boswell Sun, 03/01/2020 - 10:51

Think of it as though the microphone were mounted horizontally on a stand and you were above it looking vertically downwards. The capsule of the mic is at at the centre of the diagram, pointing towards the top of the pattern (0 degrees), and the XLR cable exits at the rear (180 degrees). The nulls are off to the sides, and for this mic are very sharp (in angular terms). The sound source you are wanting to record would normally be positioned following the 0 degree line out from the top of the diagram.

bouldersound Sun, 03/01/2020 - 11:25

MC208, post: 463489, member: 49667 wrote: I don't fully understand how to view the polar response pattern image. I understand what it's for, but from what perspective is the image supposed to be viewed? Am I looking at the mic itself as if I were talking into it? It's confusing to me.

Here's a photo that might help. This pattern is conventional cardioid, but it shows how to relate a polar pattern to the mic. It's a three dimensional pattern so it applies when viewing the mic from the side or top etc. The distance from the center of the grid to the line represents the strength of the response of the mic at that angle. A cardioid pattern has the null at 180° from the axis of the mic. The polar pattern for your mic puts the nulls at about 105° off axis, forming a "cone of silence" as I call it. Then there's a smaller lobe of response centered on 180°. If the mic were placed low and very close to the instructor's mouth angled steeply upward you would be making the most use of that polar pattern.

MC208 Sun, 03/01/2020 - 13:21

bouldersound, post: 463494, member: 38959 wrote: Here's a photo that might help. This pattern is conventional cardioid, but it shows how to relate a polar pattern to the mic. It's a three dimensional pattern so it applies when viewing the mic from the side or top etc. The distance from the center of the grid to the line represents the strength of the response of the mic at that angle. A cardioid pattern has the null at 180° from the axis of the mic. The polar pattern for your mic puts the nulls at about 105° off axis, forming a "cone of silence" as I call it. Then there's a smaller lobe of response centered on 180°. If the mic were placed low and very close to the instructor's mouth angled steeply upward you would be making the most use of that polar pattern.

Thank you, that photo matches up exactly with what Boswell described so I do understand how it works now.
I'm not sure if placing the mic low and angled upwards would work because then I believe it would probably be visible in the picture, and worse yet, it could be blocking something important. I'll try to play around with placing it somewhere low that isn't in the way and see if it helps any. I guess probably the mic is also picking up what's coming out of the speakers too, so it's a combination of the pick sound, the acoustic guitar sound, and the sound of the guitar coming out of the speakers that the mic is picking up. Quite a mess.

bouldersound Sun, 03/01/2020 - 13:41

MC208, post: 463495, member: 49667 wrote: I'm not sure if placing the mic low and angled upwards would work because then I believe it would probably be visible in the picture, and worse yet, it could be blocking something important.

I know, I was just trying to explain the general case of how the polar pattern could be exploited. I was deliberately disregarding the visibility issue. You'd place the mic just below mouth level, quite close, pointing up. The idea is to aim the axis at the desired source and the null at the undesired source, which would be impossible with the mic out of the camera's view. Also, that polar pattern changes with frequency. At low frequencies the mic will become omnidirectional.

bouldersound Sun, 03/01/2020 - 14:33

With the grille of the mic just below the person's mouth and the mic oriented near vertically, the null of the mic would be pointing right at the guitar. In addition, you would have relative distance working in your favor with the diaphragm much closer to the mouth than the guitar.

In a live performance situation with stage monitors, the null in your picture would be aimed at the monitor in front of the performer to minimize the potential for feedback.

A figure-8 pattern mic is a pretty fair option for capturing voice while isolating a guitar, but that would need to be close which would put it in the frame.

paulears Sun, 03/01/2020 - 15:18

You've got to get to grips with how patterns work for you - so the on axis to the sound source with the unwanted in the nulls. Anything else is less than ideal. Video or audio? As in which is more important. It may be you cannot have no mic visible because of the odd playing technique. It's why many resort to headsets.

MC208 Sun, 03/01/2020 - 15:34

We'll just set aside some time and I'll listen on the headphones as he plays, and move the mic around and see if it makes any difference. As for the speakers, if I place them in the null area, should they be pointed at the mic or in the opposite direction? And how do you even find exactly 150 degrees? It seems like that Rode mic picks up everything no matter where we put the speakers. Headset would be nice, but they're not an option unfortunately.

bouldersound Sun, 03/01/2020 - 15:48

MC208, post: 463501, member: 49667 wrote: We'll just set aside some time and I'll listen on the headphones as he plays, and move the mic around and see if it makes any difference. As for the speakers, if I place them in the null area, should they be pointed at the mic or in the opposite direction? And how do you even find exactly 150 degrees? It seems like that Rode mic picks up everything no matter where we put the speakers. Headset would be nice, but they're not an option unfortunately.

One other contributing factor we haven't discussed is the acoustics of the room. If a sound is reflecting off a wall/floor/ceiling then that direction also has to be considered. In a live room sound will be reflecting off all the surfaces and arriving at the mic from all directions, with many variations in timing, level and filtering effects from the nature of the reflecting surfaces. Reverberation can be managed with positioning, but that means putting the mic closer to the source and probably in the camera's frame.

Boswell Mon, 03/02/2020 - 07:08

MC208, post: 463501, member: 49667 wrote: We'll just set aside some time and I'll listen on the headphones as he plays, and move the mic around and see if it makes any difference. As for the speakers, if I place them in the null area, should they be pointed at the mic or in the opposite direction? And how do you even find exactly 150 degrees? It seems like that Rode mic picks up everything no matter where we put the speakers. Headset would be nice, but they're not an option unfortunately.

Take care to use the NTG-1 polar plot that I posted earlier, not the one for a totally different type of Shure mic. The nulls on the NTG-1 are at 90 degrees - easy to estimate.

If you must use monitor speakers, they should be placed along the null line. They should face the performer and not outwards so they can be run at the minimum level needed without undue muddying of the mix by reflections.

I'm sorry to say that I found the tone of your recorded pickup track from the guitar not very pleasant to listen to. If there is no better combination of pickup switches, you could try re-amping the track during the mixing phase, and then using a combination of the direct and re-amped sound to try to sweeten it a little. The re-amping could be via a live amplifier-microphone combination, or by using one of the many guitar cabinet simulation plugins that are now available.

Boswell Tue, 03/03/2020 - 01:01

MC208, post: 463506, member: 49667 wrote: Boswell This is what it shows on their site. Regarding the sound, it's just raw, there's no fx applied so it doesn't sound good.

I would believe the polar diagram rather than the words. Note that "hypercardioid" is often used (incorrectly) by technical copywriters to mean a pattern that is other than true cardioid rather than a specific shape.

I wasn't being critical about the guitar sound, simply wondering how you were going to treat it before it was printed to the video sound track.

MC208 Sat, 02/06/2021 - 15:36

So I've made a discovery regarding this ongoing problem I've been having with the mic picking up stuff other than voice. All this time I've been setting the input level on the mic so that it peaks jus below 0 on the meter. But I discovered that although I can't see it on the meter, I can still hear the voice on the mic clearly when the input level is much less and this also eliminates, or almost eliminates the bleed over from everything else. I tested this with the SM58 (which I did have to turn up pretty high since the mic is out of frame), and the NTG-1.. So is this ok to do? Does it raelly matter if I can't see the input level where it "should" be as long as I can hear the speaker and it fixes the other problem that's been a PITA?

bouldersound Sat, 02/06/2021 - 15:51

MC208, post: 467324, member: 49667 wrote:
So I've made a discovery regarding this ongoing problem I've been having with the mic picking up stuff other than voice. All this time I've been setting the input level on the mic so that it peaks jus below 0 on the meter. But I discovered that although I can't see it on the meter, I can still hear the voice on the mic clearly when the input level is much less and this also eliminates, or almost eliminates the bleed over from everything else. I tested this with the SM58 (which I did have to turn up pretty high since the mic is out of frame), and the NTG-1.. So is this ok to do? Does it raelly matter if I can't see the input level where it "should" be as long as I can hear the speaker and it fixes the other problem that's been a PITA?

Changing the gain doesn't affect the difference in level between different sounds being picked up by the mic. If an unwanted sound is 10dB below the wanted sound in the mic, that relationship stays constant regardless of gain. If you're going to have a mic out of the camera frame, you might want something with a narrower pattern, perhaps even a shotgun.

That said, if the meter is reading dBFS (full scale), 0 is the absolute maximum allowable level. You definitely should be peaking lower than that, perhaps -12 to -6. For most signals, they should average around -18 dBFS (more or less, depending on the converters). That is, they should regularly cross the -18 mark. If the meter is reading dBVU, that becomes your target and the signal should regularly cross, or at least reach, the 0 mark.

MC208 Sat, 02/06/2021 - 16:11

The meters on the mixer go from -30 to +16 but it doesn't say dBFS or dBVU so I don't know about that. What I do know is what I said. The unwanted sounds don't get picked up by the mic when I set the input gain to just enough that I can hear the voice. I can hear the voice well before the signal shows up on the meters. If I set the input level to -3 on the mic, then the mic picks up all the other sounds I don't want. -30 to 0 are green, +3 to +9 are yellow, and +16 is red. So I think the input level is normally supposed to be a bit less than 0, I understand that part. But bringing the mic up to the level where it's supposed to be doesn't seem to be necessary is what I'm saying. It seems to be what's been causing all these other sounds to be picked up.

bouldersound Sat, 02/06/2021 - 16:18

If it goes to +16 it's probably VU, in which case 0 is the target and the level can swing above and below it.

Unless there's some kind of noise gate being used, the differences between loud and quiet parts will remain constant with different amounts of gain. I think what you're hearing is a "sonic illusion." If the part you don't want is 10dB quieter than the part you want, that won't change.

kmetal Sat, 02/06/2021 - 17:36

It is possible that non-linearities in the circuitry can be picking up a bit more pick noise than at lower levels. Ive found cheaper pre amps and gear to get harsh/overly bright as gain gets increased especially when cranked up. I would not expect the effect to be extreme. Before i learned about gain staging i used to set the trim knob based on where it sounded best to me.

So it could be a bit of placebo effect and some could be the gears response.

You can just do 2 recordings with the two different gain settings, and match their level in the daw to anylize the difference.

bouldersound Sat, 02/06/2021 - 17:41

kmetal, post: 467329, member: 37533 wrote:
You can just do 2 recordings with the two different gain settings, and match their level in the daw to anylize the difference.

Ideally, you'd split the mic signal into two inputs with an XLR Y-cable and record the exact same performance at different gain levels onto two separate tracks.