Comments Please?

[ric8138]

I did my first indie film shoot and would appreciate it if any of the members would take a listen to this short (60 sec) sample of dialog and music and give me some feedback. I am hearing some weirdness on the S's in the word "witness" and think the mic might have been over driven (Countryman B3, Sennheiser G3 wireless system). I recorded it at -6 and -18 on my Zoom H6, and the problem seems to be on both tracks. The gangster speaking with the Russian accent was recorded on both a lav and an MKH50 on a boom. This is the boom track, which I think sounds really great, and I want to thank everyone that urged that mic purchase, definitely worth the $$.

https://vimeo.com/179338498

 

[TheDingo]

>>>I recorded it at -6 and -18 on my Zoom H6

I normally use -18 dB when using a LAV mic with my Sennheiser G3 transmitter ( the pack the LAV plugs into ) and -6 dB on the receiver pack.

 

[ric8138]

>>>I recorded it at -6 and -18 on my Zoom H6

I normally use -18 dB when using a LAV mic with my Sennheiser G3 transmitter ( the pack the LAV plugs into ) and -6 dB on the receiver pack.

Yes I do as well, I was referring to the actual VU meter settings on the H6, I recorded the main track a bit "hot" at -6 db, and used the Zoom "backup" file feature that records a separate file at 12db less for safety.

 

[Rick R]

The compander in the G2/3 system is known to affect HFs. Dialog is 'usually' ok, but in the presence of other HF sounds, artifacts/distortion can be quite audible.

 

[Alex H.]

The compander in the G2/3 system is known to affect HFs. Dialog is 'usually' ok, but in the presence of other HF sounds, artifacts/distortion can be quite audible.

Yeah. Try jingling car keys into the mic and see what happens.

I heard a little distortion on the very first word ("Drop"). The rest of it is loud, and yes you do have sibilance standing out a little. I'm confused about the original question, though: you start by asking about sibilance distorting in the Sennheiser wireless system, but then at the end of your post you say that we're listening to the boomed MKH50? Which are we listening to?

Back to the Sennheiser G3 system, there's also a proper gain staging process that needs to be completed in order to get the best signal out of the system. It starts with setting the sensitivity in the transmitter to be full but not clipping (the meter should average 2/3-3/4 of the way up). Leave enough headroom that a sudden shout won't overmodulate. Then, adjust the AF out from the transmitter to match the input impedence on the H6. You can distort the signal at this point as well, if the AF out is too hot for the input of the recorder. Then set your record levels from there.

Typically, dialog should ride between -24dB and -10dB, with occasional peaks at -6dB (and maybe -3dB if it's a really loud peak... a yelp, scream, etc.). Recording hot like you are minimizes headroom.

Also, have you done any kind of post-processing on this audio? Compression of any kind?

 

[David W. Richardson]

Maybe I'm deaf, but I think it all sounds pretty great. I think when it's paired with the picture, nobody's going to even notice. That's just my 2 pesos.

 

[ric8138]

Yeah. Try jingling car keys into the mic and see what happens.

I heard a little distortion on the very first word ("Drop"). The rest of it is loud, and yes you do have sibilance standing out a little. I'm confused about the original question, though: you start by asking about sibilance distorting in the Sennheiser wireless system, but then at the end of your post you say that we're listening to the boomed MKH50? Which are we listening to?

Back to the Sennheiser G3 system, there's also a proper gain staging process that needs to be completed in order to get the best signal out of the system. It starts with setting the sensitivity in the transmitter to be full but not clipping (the meter should average 2/3-3/4 of the way up). Leave enough headroom that a sudden shout won't overmodulate. Then, adjust the AF out from the transmitter to match the input impedence on the H6. You can distort the signal at this point as well, if the AF out is too hot for the input of the recorder. Then set your record levels from there.

Typically, dialog should ride between -24dB and -10dB, with occasional peaks at -6dB (and maybe -3dB if it's a really loud peak... a yelp, scream, etc.). Recording hot like you are minimizes headroom.

Also, have you done any kind of post-processing on this audio? Compression of any kind?

First a more complete description of the scenario. 3 Actors, but actors 1&2 have most of the lines, actor 3 (George) just interrupts once.
I have lav's on all 3 actors, with each one going into a separate channel on the H6. I am using the "Backup File" function on Actor 1&3 which records a second version of the same file at 12db less than the main version which is why I run the main file a bit "hot".
The boom is the MKH50, and I run that through a splitter to get that into 2 channels as well, one hot and one not. These are primarily stage actors, and they shout their lines on occasion out of habit.

When we start the scene with the wide take, the boom often can't do much due to framing. When we do the coverage shots of each actor's face I put the boom over the actor being featured in the CU. Of course the wired boom sounds better than the wireless lavs. The sample I posted consists of the wireless lav's on actor 1&3 and the boom take on actor 2 (fake russian accent). I suspect the distortion in actor 1's track ("Drop the gun" and " we know what happens to the witness") is due to overdriving the wireless circuit, as it appears in both the hot and cold recorded track. I had the sensitivity on the transmitter at -9. I have lowered it to -18.
I could turn to the CU take of actor 1 (with boom mkh50) to remedy this scenario. My gain on the H6 inputs runs at half to 3/4 (5-7) on the dials.

 

[ric8138]

Maybe I'm deaf, but I think it all sounds pretty great. I think when it's paired with the picture, nobody's going to even notice. That's just my 2 pesos.

Thank you David! I am not displeased, and it is certainly better than any effort I have achieved in the past plugging into camera sound. I may be overthinking all this with the backup tracks and all. The sound circuits on my Sony FS700 have no where near the subtlety or range I am getting from this standalone setup. We always did the "hot and cold" track because it was pretty easy to exceed the abilities of the Sony in camera circuits.

 

[ric8138]

Also, have you done any kind of post-processing on this audio? Compression of any kind?

I do a "voice gate" that drops out everything below 125, and above about 12Khz, as that gets rid of most air conditioner/ road noise/ room noise rumble and doesn't affect the voice to any great degree. No compression. A light limiter on the final mix to knock off any peaks.

 

[Noiz2]

I didn't hear any distortion that would bother me, and it sounds really clean which is really rare on low budget shoots.

One thing I would probably try is to sweep the mids because I think you are getting some resonance peaks. I find a lot of times there is a slight muddiness that is caused by resonances in the mid range. Kind of like the system is just about overloaded, or someone speaking too close to the mic. It doesn't come off as distortion but a kind of over fullness. Hard to describe. But if you take a really sharp notch filter and crank it up + (whatever you can get to, usually 6-12dB) and sweep it around in the mids I bet you will find a few spots where it resonates. You will easily be able to hear it, instead of just sounding weird because of the narrow boost it will sound a bit like PA feedback. When you find them you reverse the filter so it's notching out a bunch of the resonance. You have to do it voice by voice since it's a factor of mic placement and actors voice quality. In the end you will get rid of the over full mid range and the voices will sound much clearer.

Good job.