[Bruce Watson]

[phil wang] “I have free access to a Rode NTG-3 mic which I know is a completely different mic than the AKG p220 but I heard people use it for voiceover work and it is much more expensive so maybe it’s just as good or better than the akg p220 even when it is used for studio recording?”

This is like comparing oranges to concrete blocks. The LDC cardioid is designed for studio music uses. The SDC shotgun is designed for dialog capture on a boom pole, mostly exteriors. There’s little overlap here.

Yes, you can use a shotgun like this for voiceover. It exhibits a lot of proximity effect, and you can use that to your advantage sometimes. But shotguns are also quite sensitive to early rapid reflections off nearby walls and ceilings, resulting in comb filtering artifacts that you aren’t likely going to be able to fix in post. So unless your studio is big, or very very dead, I suggest you rethink your ideas for using a shotgun in the studio.

[Bruce Watson]

Is this new behavior or has it been like this since you got the units? If it’s new behavior, what did you change that resulted in the new behavior? Anything at all — new mic mounting technique, new radio setting, mounting on someone you’ve miced before, mounting on someone new, etc.

What microphones are you using? Are all three of the microphones the same make/model (verify it with physical inspection of all three)?

What batteries are you using? How “fresh” are they? When in doubt try fresh alkalines (not rechargeables) and see what happens.

As Richard suggests, standard trouble shooting in these cases is to swap components to narrow down the trouble spot. If you swap mics between your troubled system and a working system, and the trouble stays put, it’s not the mic. If you then swap out the transmitters and the trouble stays, it’s not the transmitter. That leaves the receiver.

If the trouble follows the part when you swap something out, the problem is with the part you swapped out.

[Bruce Watson]

[Phil Parkinson] “So just to be clear, a shotgun mic is only useful at shorter distances (i.e. just out of frame), otherwise it’s lav mic all the way? A meter is too far for the shotgun mic to work.”

It depends on what you are trying to do. If you’re trying to capture dialog, for example a movie actor, or a sit down interview, then a shotgun mic at 1m is IMHO generally too far away. Being this far away lessens the level of the signal (actor talking), and therefore lessens the signal to noise ratio, so you end up with more ambient sound in your dialog capture, and if your mic preamps are noisy, more preamp noise in your capture also. If you want “good crisp dialog” and nothing else, that’s what the “close as you can get and still be just out of the frame” distance (45-60cm) is for.

That said, if you’re recording ambient sounds like birds, traffic, whatever, you can certainly do that with shotgun mics, and at quite a distance too. People often use shotguns at sporting events to capture crowd sounds for example. I’ve read of one guy using a pair of shotguns, widely spaced (he wrote something on the order of “about the width of a normal flatscreen TV” IIRC), the pair pointed at the stands, to pick up stereo ambient crowd noise at football games. He swore by the technique. I’ve never tried it however.

[Bruce Watson]

[Robert Withers] “??? I’ve used shotgun mics at 1-2+ meters in many situations for years. That’s what they’re for.”

Wherever did you get that idea? A shotgun mic should be boomed no farther out than 60cm. Closer is better. I aim for 45cm when possible — above and out in front of the speaker, pointing down at the speaker’s mouth at about a 45 degree angle. If you don’t dip into the frame every once in a while you aren’t really trying ? Seriously, this requires good communication between boom op and camera op.

The NTG4 has a rated sensitivity of -32 dBV/Pa. Compared to a Sennheiser MHK 8060 which is rated at -24 dBV/Pa, the NTG4 is around 8 dB quieter given the same source. That’s part of your problem. Another part is the less than stellar gain available from the Zoom H4.

So, you’re booming from too far out, using a mic that isn’t very sensitive, into a recorder with not a lot of clean gain available. Easiest thing to do is try booming closer. That should help some, but at best probably only around 6 to 8 dB. If that’s not enough, I’d look at a better recorder that has more clean gain available. If those two aren’t enough, I’d look at a more sensitive mic.

[Bruce Watson]

[Peter Holt] “”I don’t follow what your proposed workflow is or exactly what is a “standalone timecode recorder”?”

I would like to be able to drop an audio event onto the timeline and have it align with the video, where there is video.”

This requires four things. First, a timecode generator of some kind. The generator has to provide timecode to the video and audio recorders. Second, a video recorder that can record the provided timecode in some way, while it records video. Third, an audio recorder that can record the provided timecode in some way, while it records audio. Forth, an NLE that can read the timecodes recorded by the video and audio recorders, and use these timecodes to line up the video and audio when you drop the video and audio onto your timeline.

You may be able to get around not having specific time code functionality by recording the timecode onto an available audio track. You’d use something like Tentacle Sync maybe (I’ve never used it, only read about it, and there are undoubtedly competitors as well). As to NLEs that can use audio timecode directly, you’d have to research it.

Can you get away without dealing with timecode? For a few minutes, probably. But your video recorder is going to drift, as will your audio recorder, and there’s no telling if they’ll drift together or apart, or at what rates. Thus… SMPTE timecode was invented.

[Bruce Watson]

[Sue Howard] “https://files.creativecow.net/mp3-files”

I downloaded and listened. Very little voice content and what there is sounds highly distorted. I wouldn’t be able to extract anything usable from this. I doubt it’s salvageable. Sorry.

[Bruce Watson]

[Sue Howard] “I need help with a conversation recording that has alot of background noise and humming to get the dialogue in the back. Can anyone help or recommend a very good, easy to use software.”

As others have said, we’d need to hear a sample to have any idea what to recommend.

That said, this is sort of like saying you put too much salt in your cake, and want to know how to take some of it out now that the cake has been baked. Very difficult, easy way is to bake another cake with the correct levels of ingredients. IOW, restage and rerecord with the proper mic in the proper position this time. If there’s any way for you to have such a do-over, that’s what you should pursue because that’s what will get you the best results. Really.

As for recovery / repair of what you have, about the best you can hope for is to increase the signal to noise ratio (dialog to background) a bit, and roll off the top and bottom end that’s not in the voice frequencies. It’s likely not going to sound stellar, but may be just good enough to use. IOW, don’t get your hopes up. You can’t generally just drop the background noise out completely, because that background noise almost always overlaps the frequencies used by your dialog. Much depends on the nature of that background noise, which is why the requests for you to post a sample.

[Bruce Watson]

[Craig Alan] “setting the limiter below+20 to +12 makes sense to me.”

Notice from the SD docs on the 302’s input limiters: “+18 dBu threshold, 20:1 limiting ratio…” So SD ships them with the default limiter threshold at +18 dBu. That’s because of the massive 20:1 ratio. The 2 dB between activation of the limiter and the 302 actually clipping is huge. Not a brick wall, but you can see it from here. It’s highly unlikely you’ll encounter anything that will go through the limiting process and still make it to clipping. So setting the limiter threshold to +12 dBu, which is 6 dB lower, is effectively tossing out that 6 dB of gain. Yes, it’s technically headroom, but practically nothing short of an explosion will be able to make use of the upper range of that headroom. And if you’re recording explosions, please be careful!

That said, just because SD expects people to use the mixer this way doesn’t mean anyone has to. That’s why SD makes the limiter threshold adjustable. I’m just sayin’ that setting the limiters to +12, and making your average dialog peaks 0, that’ll work too. You’re not doing anything that’s wrong.

I only answered that I typically set my average dialog peaks at +8 because you asked the question in your original post:

[Craig Alan] “But why aim for a peak of 0 on the mixer if you can record at say +4 +8 or +12?

My answer is that I personally aim for a peak of +8 on the mixer. Because I can’t think of any reason not to, and SD couldn’t either when I asked them. Limiters at +18 dBu, calibration tone on the recorder at -20 dBFS, and average dialog peaks around +8 dBu on the mixer (which translates to -12 dBFS on the recorder) seem to be the way SD expects the mixer to be used in my understanding.

[Bruce Watson]

[Ty Ford] “I set the limiter threshold on the Sound Devices devices to -12. This gives me bump and shriek room…”

My understanding (that is, my interpretation of what SD tech. support told me years ago) is that the MixPre-D doesn’t clip until +20 dBu on the MixPre-D meter (second to the last LED). And the lowest you can set the limiter is +8 dBu. So maybe you mean +12 instead of -12? Or maybe I’m just misunderstanding what you’re saying.

[Bruce Watson]

[Craig Alan] “This allows the user to record with a full 12 to 20 dB of headroom while in the ne-resolution green-colored part of the meter. Additionally, the meter color changes to orange at 0 dBu and red at +8 dBu. These color changes correspond to -20 dBFS and -12 dBFS respectively, which are commonly used recording levels for today’s digital recording devices.”

Not a very elegant description. What they are saying, more or less, is that the 302’s meter is a Vu meter (so called dBVu units). A completely different thing, and a different scale, than digital dBFS.

What I do is use the calibration tone on the mixer (mine is a MixPre-D which is one step down from the 302). Use that calibration tone to set the recorder (camera, dedicated recorder, or whatever is recording) to -12 dBFS. Then completely ignore the recorder and do all adjustments from the mixer. I typically aim for a dialog peak of +8 dBVu on the mixer, which gives a corresponding -12 dBFS on the recorder. This gives me 10 dB of headroom before the mixer’s limiters kick in at +18 dBVu (which is 2 dB before the recorder’s 0 dBFS).

Done this way I’ve found my recording chain to be more or less “clip proof”. IOW, I’ve never had it clip, and seldom seen the limiters actuate. It takes something like a cooking instructor setting down a glass mixing bowl on a granite countertop considerably too hard to kick the limiters in; I’ve never had a voice do it.