Lav gain problem

[Greg Miller]

Dana,

If that was supposed to be an answer to my question, I'm afraid I didn't get it. You're confusing me by talking about various people, and apparently several Zooms.

Originally you asked about plugging one particular [Giant Squid] mic into one H2 (or so I thought). And you then plugged that Zoom into your camera. Is that what you were asking?

If so... then if you plug THAT mic into THAT Zoom, and set that Zoom to record in stereo (without having the camera connected)... how are the levels on the Zoom?

I realize you might need to set up a short test to do this, as I'm not sure whether you've ever tried this in the past without the camera. But the answer should be pretty simple. I would expect either (1.) the levels on the Zoom are fine in this case, or (2.) even without the camera, the levels on the Zoom are very low.

I would expect that the levels in FCP are the same as what you observe while recording, but I want to confirm that also.

If your answers are even a bit more specific, that wouldn't hurt. "The levels on the Zoom peak at around -6dB, and the levels in FCP are the same" or something like that.

Could you please perform a test (if necessary) and give me a specific answer? That will help greatly with a diagnosis of your problem.

[Dana Salsbury]

Thanks Greg. I will do that. I was just clarifying your question. Not sure why I rambled on about my setup. Posted too late at night.

[Dana Salsbury]

>I would expect either (1.) the levels on the Zoom are fine in this case, or (2.) even without the camera, the levels on the Zoom are very low.

2. Even without the camera, the level is lower than I feel it should be. I have several Zooms and several Squids, and get the same result. It's "acceptable", but seems odd to have to be on high gain on the Zoom just to hear a talking head with a lav. Am I right to believe that by having it on high gain means quality loss in the same way that increasing gain in Final Cut means quality loss?

[Greg Miller]

Dana,

If I understand this correctly, even without the camera, the level from the GS mic is lower than the level from your other mic(s). If that's the case, then there is something happening beyond what I had imagined.

I still don't rule out the problem that Steve House envisioned... I suspect the same thing. We can come back to that.

But I'm still curious about the GS mic -> stereo Zoom test. On your test track, roughly what was the level of the loudest peaks? (expressed either in % of full scale, or in dB below full scale)

High gain on the Zoom does not necessarily mean quality loss. The issue is signal/noise ratio. No easy way for me to say "yeah" or "nay" unless I can hear/see a bit of the file.

So... can you possibly post a small, unaltered section of the latest test file? Thanks for taking the extra trouble...

[Dana Salsbury]

Here it is:
https://www.dropbox.com/sh/arxwlkt1hcq1t16/l6kP1QulHu

It peaks at 100%, it just seems like it should peak more on high gain.

Thanks!

[Rick Reineke]

"It peaks at 100%, it just seems like it should peak more on high gain"

- I only see/hear handling and other (unknown) noise hits approaching or reaching full scale. Dialog is down around -25dBFS, w/ preamp and acoustical borne ambient noise around -45dB, which includes a LF hum.. ground loop; cable; power supply? Though a HP filter could attenuate that some.

[Dana Salsbury]

It seems like a quiet room. I've got a computer running in the closet with the door shut. The closet has a fan also.

Is the dialogue where you would imagine it to be?

What would you suggest?

[Greg Miller]

I concur with Rick Reineke about the levels and noise.

It's a strange file. I would guess that it's several different takes all scoped together into one file; certainly something changes drastically from time to time.

At the beginning there is just a lot of broad-spectrum noise. That could be a mix of preamp noise and air movement. (I note a peak around 3700 Hz, which might be an acoustical peak from the air noise source or from the mic.)

Beginning at 5:43 there is significant 60Hz power line hum in addition to the above noise. The power line noise disappears after all the banging and crashing at 9:20.

Then, starting at 18:22, there is a huge amount of low frequency noise... apparently acoustical noise (motors, refrigerator, etc.???) in addition to the ever-present broad-spectrum noise. And I note that the dialog at the end of the file (19:18 to 19:24) is about ten dB louder than the rest of the file. Peaks here are about -10 dB, the rest of the file is about -20 dB or lower.

I assume the lav is properly clipped onto the talent, is that correct?

So Dana: what's different about the last six seconds of the file? Why is the dialog much hotter here (and also more room noise)?

And one more question: Where is the H2 recording level set?

[Dana Salsbury]

I'm so grateful for you guys!

Here's the video after editing:

Vee Split Screen Interview - YouTube

It seems odd to need to filter such a quiet room, but I guess that's what I'll need to do. Once factor I didn't consider is that the Zoom is plugged in, whereas I usually use batteries. I also didn't use a broadcast loop since I was sitting still. I guess I'll need to be doing that. Turns out I do move!

I was doing several takes and left it going to have one less thing to remember. The switch setting was high gain. I don't use any of the menu compression features. I do use mono and select 48Khz at 16 bit. The zoom lay on the desk.

I'm not sure why the last six seconds are louder. It must have been closer.

[Greg Miller]

Originally Posted by Dana Salsbury

The switch setting was high gain. I don't use any of the menu compression features. I do use mono and select 48Khz at 16 bit. The zoom lay on the desk.

I'm not sure why the last six seconds are louder. It must have been closer.

I know you've already said the mic gain was set to "high." But what about the record level setting?

Was the Zoom on the desk for all the different takes? There had to be some change in what was turned on/off in the room, because the AC power hum is present only in parts of the file.

As to your comment about "closer"... what must have been closer to what? That take sounds drastically different from the rest. The dialog level is higher. The room noise is much louder, and has a much different frequency spectrum compared to the rest of the file.

I would suggest that someone ought to be monitoring on headphones whenever you do any recording, and you'd catch this stuff when it happens. But even so, don't you hear the difference in playback? Something changed drastically when you recorded that last, short take.

(BTW, not quite responsive to your question, but relevant to your YouTube file: how did you capture the harpist's end of the interview? I notice the level jumps up and down a lot on the edit cuts. Also, I seem to hear some artifacts on her voice. I'm just wondering how her end was recorded and how you received that audio.)

[Dana Salsbury]

I'm not sure what you mean by Record Level setting. The closest thing I have is Record Mode. (48/16)

The Zoom was in the same place. The only thing I can think of is that I had bumped the switch to the low position and caught it at the end. It just seems to happen too often. I definitely need monitoring -- both for audio and video.

The harpist had the same setup. I interviewed her at a wedding in a noisy place. I recorded my questions later and creatively edited it all together as a faux satellite interview.

BTW, in the last few seconds was the noise and ambience acceptable?

[Greg Miller]

Originally Posted by Dana Salsbury

I'm not sure what you mean by Record Level setting. The closest thing I have is Record Mode. (48/16)

I don't have an H2, but I've downloaded the online manual for reference.

P.15 talks about setting the mic gain switch (which you've said you always set on HI).

P.16 talks about setting the Recording Level (after you go into record/standby mode). The manual says there is a setting range of 0 to 127. That's what I'm asking: where do you have this set? What number?

Or do you have some variety of H2 that doesn't have this setting (which seems extremely unlikely to me).

[Dana Salsbury]

Oh my GOSH! I've been through the manual but I've never seen that! I'm set at 100, as is all my other Zooms.(I'm sure that's the default. That's huge, though, as I'm often recording PA speakers and low is not low enough, so I have to move the stand. Thank you!

So my ideas to improve sound:
1. Have it monitored, adjusting settings as needed to get the right level.
2. Shut down my Mac Pro, even though it's in the closet.
3. I'm thinking that with my lav being monitored it would make more sense to use my H2n for it's dial control.
4. Put up some acoustic foam on the wall I'm facing at least.

For what I'm doing, it sounds like you're saying I've got the right equipment, but that the thing is how I use it. Are there any post processing tricks I should employ beside hum removal?

[Greg Miller]

When all else fails, read the manual. !!!

If you need more gain when you're recording the GS lav, now you can try the Record Level setting.

Yes, make the room as quiet as possible.

You don't need to spring for acoustic foam unless you like the look. Moving blankets are better than nothing. Carpet on the floor can help somewhat. Any absorption in the room will lessen room reflections. But that's not the same as eliminating noise.

I still want to know why the last six seconds of the audio file you posted sounded so much different from the rest of the file... much higher voice level, much higher noise level, entirely different noise spectrum. (And why some takes had a little bit of AC power line hum, and other takes did not.) You need to identify and eliminate these variables, up front, before you do more recording. Connect a good set of headphones (e.g. Sennheiser HD-280 pro) to the H2, play with gain, play with low cut filter, move the recorder and wiring around looking for hum, etc. etc. etc. It can take some time, but it's well worth it. Your life will be a lot easier if you start with clean tracks.

Now, to come back to your original level question. If you feed the mic into the H2, and adjust the H2 for correct recording level (now that you know how), then feed the H2 output to the camera, can you adjust things so the level on the camera matches the level on the H2? Or is the level on the camera significantly lower than the level on the H2?

[Rick Reineke]

The H2 has a three-position low, med., high Gain switch (L-M-H) on the side and a recording volume control, via the front panel up/down' arrows" as I recall the record volume reads from 0 to 127, so in theory, a setting of of 70 to 100 should be a good initial setting to start, then set the 'gain switch' to achieve voice peaks of around -12dBFS or so on the H2's meter.
I'm not familiar with GS mic, if it needs external 'Plug-in power' that must be activated via the H2's software menu. If it has it's own battery, it should be disabled.