Boosting H4N waveforms

H

hcproductions

Member
I'm currently using a boom mic set up for a few interviews that I've shot. I'm using an older AKG-c568 eb shotgun mic with a wind screen on it and the bass roll off turned on plugged into my zoom H4n. I have it set up for one input-stereo recording. My recording level is usually about 86, and when I look at my meters on the H4N it usually falls around -24 to -12, but never really much higher. I had the shotgun mic on a stand about 3 feet away from the talent.

My question is, when I drag the files into something like FCPX or Soundtrack pro, I always get small waveforms and a pretty low volume off the bat. I could normalize them and boost them with some gain, but it seems to me that they should be much larger and louder than what they are in the first place. I've used the H4N on a variety of projects, all giving me similar results. Is this normal, or is there something I can do to help this out? The audio itself is usually ok even when I have to boost it (assuming there's no air handler noise) but it just seems like I should be getting a stronger signal into the post software.
 
Mic 3 feet is really too far away. Getting it closer will help a few dB's.
Other than that a preamp or hotter mic...
 
Even in applications where the mic was close, like recording a loud trumpet with an SM58 I had laying around, the waveforms still read low. Granted, these are different applications and uses, but it sounds like it may be a preamp thing then. Any closer on the shotgun and it's in the shot. Is using an external preamp the norm with an H4N? Or could it be something faulty with mine?
 
Most of the budget recorders preamps do not offer a lot of gain. A high output mic would help some, but good quality external preamp(s) would be much better. But beware, if using an external preamp or mixer with the H4n, it does not support +4dB (nominal) line-in.. only -10dB (and instrument level) via the unbalanced 1/4" input.

When peaking record levels around -12dBFS, the waveform on most NLEs or DAWs WOULD appear low. Most allow normalizing a clip, which would make it easy to achieve desired post levels... the 'quality' of the audio would not change though, only louder.. good or bad.

"Any closer on the shotgun and it's in the shot"
- If it's more than 18 inches or so, you should re-think your mic options. A lav for instance.
 
Last edited:
I too use an AKG shotgun and a ZOOM H4N since I began shooting with a DSLR. And I too experience low waveform levels similar to yours. But since I got Magic Lantern on my 5D I tried to plug the mic straight to the cam. I fixed the low levels from the mic by using a cheap pre amp called ART Dual pre. Cost me only $34 used from guitar center. I screwed a rail block onto it and now it attaches to my rails.

I now have good levels and I ditched the external recorder.
 
Where are you recording level wise? It may be you have your headphones up too high, and therefore hearing "false levels" when you monitor. To fix this.... make sure you are hitting somewhere around -12db maybe a touch hotter. Compare the stupid (LOL...hate that thing) external speaker to what you are hearing in the cans.

To me it sounds like your gain staging and headphone gains are not in sync for levels.
 
From a software point of view you'll need RMS normalization (limiting) not peak normalization. RMS value represents the loudness. Opinions vary but somewhere in between -15 -12dB RMS should be fine.
 
Yeiks, You don't as a rule ever want to normalize.

There are a couple of reasons. First you don't want everything louder, you just want what you want to hear louder, so MIX, it is what you should be doing in post. Second it can actually make your dialog lower. Say you have some quiet dialog and someone slams a door. If you normalize to some appropriate level for the dialog it may well lower the track because it's looking at the door slam also. If you use the RMS as suggested the dialog may come out OK but your door slam will distort. Yes you can cut it all up and process in small bits but then your levels are going to jump all over the place and you have made more work than just level automating in the forst place.

All level changes affect the "quality" a bit. Not a lot but you still want to change them as few times as possible. Even if you normalize you will still need to mix the levels so you just added another layer of level change that you didn't need to.

If you levels are so low that you have to add more than one layer of boost you recorded much too low. You will take a slight hit but you have no choice so...

Three feet is way too far away, and your levels attest to that. So you either need to rethink the shots (or the director does) or you need to pick a different tool ala lav, plant mic, ?.

And the last point nobody has brought up. The "waveform" has nothing really to do with the signal level. You can zoom the waveform and it will look huge, but the level didn't change. You can zoom out and the waveform will look like a flat line and you levels could be peaking out. The waveform is mostly so you can find where there is sound and not for editing. With some experience you can kind of read the waveform and know what it sounds like, but unless you have set up a reference you can't really tell much about the levels. And they are a separate file (or separate part of the file) and they can get out of sync with the actual sound. So use them as a visual aid but don't count on them for "real" information.
 
Before doing any processing on the waveform you can see the dynamic differences. A door slamming, breaking a glass...etc. will be really visible transients. Those must be handled if going for bigger volume otherwise they will distort. I recommend the Universal Audio 1176 or LA-2A. Probably the best out there...
 
I recommend the Universal Audio 1176 or LA-2A.
-- Yeah, they're nice, but most everyone works in the digital realm these days. The plug-in versions or a brick-wall limiter plug-in would work OK as well... for a lot cheaper
 
Good information. i think i'll go with some kind of preamp just to have around, but I'm going to try and work the mic in closer via a lav or something. It shouldn't be too difficult for my interview set ups, it's an easy standard to see a lav on the talent, but what do people do for feature films on wider shots? I was going for more of a medium shot on the interviews as well, I could probably tighten up and get just shoulders/head to bring a boom closer. I know getting the mic closer is the most important, but are there booms designed for a little farther away than others? Or do they hide a lav somewhere? Thanks for all the info.
 
Aside from level, an external pre' won't help much if the mic is not close to the talent. poor sound, just louder.
Normally on movies lavs are hidden. A B6 can sometimes be 'hidden' in plain sight, in hair, on glasses or behind buttons, or otherwise hidden in clothes and such. This is a whole other skill set, acquired though practice, experience and 'accessories' one has had success with.
 
PPeter said:
I recommend the Universal Audio 1176 or LA-2A. Probably the best out there...
Click to expand...
Let me guess.......you have focused on music and are now starting in film. So did I. And Rick is absolutely correct. These days, many people don't have the resources to route a isolated signal in and out of any outboard gear. Most people here don't even have a decent mixer, so the suggestion, though good, will not work.

Film audio is a totally different animal from "in the studio" audio.
 
Noiz2 said:
Yeiks, You don't as a rule ever want to normalize. [..] First you don't want everything louder, you just want what you want to hear louder, so MIX, it is what you should be doing in post. [..] Even if you normalize you will still need to mix the levels so you just added another layer of level change that you didn't need to.
Click to expand...

Thanks for the information. I've learned to normalize in post last year and I've been able to mix all my soundtrack by metering the levels and adjusting the sounds. It can introduce more work, but it's been working fine so far. But as the sounds aren't recorded in the same environments and using the same settings (in my case though), mixing is inevitable, hence normalization and levels adjustment.
 
Rick R said:
I recommend the Universal Audio 1176 or LA-2A.
-- Yeah, they're nice, but most everyone works in the digital realm these days. The plug-in versions or a brick-wall limiter plug-in would work OK as well... for a lot cheaper
Click to expand...

Once we tested the hardware 1176 to the plugin version which is running on the UAD DSP card. We were shocked how small the differences were when using it for general purposes (what a compressor is for). When it comes to the "all buttone mode" which is basically for distortion there the hardware is better but I dislike that sound. This test was a good lesson.

Doc Bernard said:
Let me guess.......you have focused on music and are now starting in film. So did I.
Click to expand...

Yes! :D I spent like 15 years on music/studio work. I got tired of it, I've needed something new, more...

Doc Bernard said:
...Film audio is a totally different animal from "in the studio" audio.
Click to expand...

Do you mean the layers : ambience, sound effects, dialoge, music ?

ADD:

I usually look for downloadable videos on vimeo and watch them withouth the sound (muted). If I find something good I download and pull it into DAW. Switching the measures from beats to time and the sample rate to either 48 or 96khz I can start thinking of what sound would fit to that picture using the same work flow. It's really (!!!) interesting. Checking the final render without the picture I always think that I couldn't have composed this had I not had the film in front of me in the video window during playback. I love this it makes me more creative. Next step will be recording sounds when my Tascam DR-40 arrives. I would be interested in how you guys working on the audio tracks.
 
Last edited:
drapeama said:
Thanks for the information. I've learned to normalize in post last year and I've been able to mix all my soundtrack by metering the levels and adjusting the sounds. It can introduce more work, but it's been working fine so far. But as the sounds aren't recorded in the same environments and using the same settings (in my case though), mixing is inevitable, hence normalization and levels adjustment.
Click to expand...

I think we may be using the same word for two different things. In a true sense in mixing you are "normalizing" the levels. That sounds like what you are doing.

But what folks are almost always talking about is a "normalizing" plugin. The plug in looks at a clip or group of clips and destructively re-renders them to a new level based on the peak levels (someone mentioned one that will also normalize based on an RMS level but I actually havent seen that).

For the various reasons I mentioned this is not usually a good choice. It used to be a much worse choice since "normalizing" plugins were designed originally mostly for creating samples for samplers and always normalized to a peak value of 0dBfs (which is at least 3dB higher than you want your peaks) and their math was les than great so there tended to artifacting.

These days almost all will normalize to a level you pick and the math is MUCH better.

It is really useful for somethings, just not usually in doing dialog.
 
PPeter,

I think a lot of post workflow is actually covered in the stickies so you may want to cruse that.

I clean and mix dialog first. If your working on a narrative film dialog is really king (and this from an FX editor!).

Next usually I will lay in backgrounds and ambiences. Then hard FX, then Foley, and last music.

On a big film those "food groups" are premixed separately usually by different mixers, on low budget one man crews you still need to work a bit that way, though I don't "pre mix" anything.

Things the will change the flow are some big FX that will take a ton of time, some specific sounds that picture needs (as in sometimes you can make a film cut work that is awkward by putting the right sound on it to mask it). And if the music is already chosen or there is one section that is going to be 90% score, then I will get that music in first so I can put FX in that wont fight it. Even with my "standard" workflow I am usually going back into FX and BG's to make room for score, so it's a bit interactive always.

If you have done something close to a "real" dialog edit, it's fun to play back everything but dialog and music. If you have done a really good job it ought to sound fully filled out.
 
"I think we may be using the same word for two different things. In a true sense in mixing you are "normalizing" the levels. That sounds like what you are doing."

Yes, when I 'peak' normalize a track or clip, it's pre-fader within the DAW, the final level/mix is by ear on a calibrated PB system.
 
You must log in or register to reply here.
Back
Top