Zoom H4N really low levels but still distorting - why?

T

tikigod19

Well-known member
hi all,
recently switched to DSLR for weddings and I bought a H4N to capture audio using my Senheisser ME66 and Senheisser G2 wireless system. (phantom set to on on the shotgun) I'm having an absolute nightmare with it. As the display goes up to 0dB and not beyond, I can never really tell if anything is peaking unless I have it set so that the levels go to -6dB or occasionally 0dB but then the levels recorded are so low when imported into my NLE (even with both input levels set to 100)

I find that if for example people clap in the audience then the sound distorts and clips, however in premiere pro the audio is still only showing at about -6dB.

When I used to plug the same mics into my DVX100 it was a lot simpler than this with no clipping (and also preferred how the level display went above 0dB)

somebody please make me love this unit before I bin it... it can't be all that bad as it seems everyone and their dog has one if shooting with DSLR...
 
The G2 needs to be setup properly, first and foremost. Peaking in the middle of the display indicator is considered good levels.

On the Zoom...can't help you there. -6dB is way too darn loud, especially since the Zoom is a 24-bit recorder. From product photos ( http://www.samsontech.com/images/productImages/h4n_page/4-channel-mode.jpg ) it appears you should be able to view recording levels below -6dB. They may not be labeled but those divider markers have to mean something. Maybe it's in the manual? http://www.zoom.co.jp/downloads/h4n/manual/ . Once you figure that out set your peaks to between -20dB and -10dB.

Also gain stage properly. If there is a "unity" gain/trim setting on the Zoom set your controls there. Then on the Sennheiser receiver set the AF out so that your levels are good (-20dB and -10dB peaks).

Edit: Just checked the manual. The notches/dividers are 6 dB steps. But like teaching a man to fish instead of handing him the fish please read the manual, especially starting at page 47. ;-)
 
Last edited:
tikigod19 said:
hi all,
recently switched to DSLR for weddings and I bought a H4N to capture audio using my Senheisser ME66 and Senheisser G2 wireless system. (phantom set to on on the shotgun) I'm having an absolute nightmare with it. As the display goes up to 0dB and not beyond, I can never really tell if anything is peaking unless I have it set so that the levels go to -6dB or occasionally 0dB but then the levels recorded are so low when imported into my NLE (even with both input levels set to 100)

I find that if for example people clap in the audience then the sound distorts and clips, however in premiere pro the audio is still only showing at about -6dB.

When I used to plug the same mics into my DVX100 it was a lot simpler than this with no clipping (and also preferred how the level display went above 0dB)

somebody please make me love this unit before I bin it... it can't be all that bad as it seems everyone and their dog has one if shooting with DSLR...
Click to expand...

You are recording way too hot. Meter readings going over 0dB make no sense on a digital peak reading meter. At 0dBFS, all the bits in the signal are ones and the signal can go no higher. That is the absolute point at which it clips. So if your meter is hitting the 0 mark, it's no wonder your signal is clipped and distorted, you should never even come close to 0, much less touch it or send it a signal so hot that would go over if the meter scale went higher. "Headroom" over 0dB is a concept from the analog recording world and does not apply to digital. The normal recording level for voice should be about -12dBFS while the alignment level should be -20dBFS when sending the recorder a 0VU sine wave tone. -3dBFS should be considered the 'never exceed under any circumstances' mark in order to give you a proper safety margin.

If you find the program sounds normal but the audience clapping is distorting, it may well be that there really is such a range of levels in the recording environment that you have to ride gain in order to get it all in ... a level setting that is good for the show is too high for the audience reaction. You haven't said how you're setting up on-site but if you're trying to set and forget the recording levels instead of devoting a full time skilled crew member to audio you're trying to do the impossible. Sound needs just as much constant attention as does camera.

It's strange you say the levels are too low when you get the file into your NLE. That just doesn't jive with what you said about clipping. Just how are you evaluating them? Are you looking at the editor's meters, the waveform envelope itself, or is it a subjective evaluation by listening? If by listening, are you using real monitors, headphones, or trying to get by with computer speakers? Are your monitors properly set up and calibrated so that a 0VU averaging, -12dBFS peaking, pink noise signal has a measured volume of 78 to 82dBSPL at the editing work position?
 
I can't speak for the zoom but I had a similar situation occur with the Tascam DR-100 after feeding it from my mixer. I used the XLRs on the tascam from my PSC Alphamix. While keeping the levels around -12 on the tascam, listening to the playback, and looking at the waveform shows heavy clipping of the waveform even though it never came close to 0db and the sound was distorted.

Once I moved the input from the XLR to the 1/8" line in of the Tascam, all was well. My only explaination was perhaps and impedence mismatch causing the low level waveform clipping.

However, I do agree with the others that gain staging may also have played a roll in this distortion, and yes, by all means, do whatever you have to in order to keep your levels below 0db on the meter. Unlike the analog world where tickling the red on meters means a good hot signal, hitting red in digital means you're screwed and blood as been spilled.
 
I will clarify something every one was talking about but you probably don't have a clue about. Meters are not nec. measuring the same thing. The reason that your old meter went above 0 was because they were analog meters measuring (I'm guessing here) VU. The Zoom is using a digital fs metering. In that kind of meter you count down from 0 so there is nothing ever above 0. Zero on your old meters should be about -20 on the Digital fs meters. That will have your peaks probably in the -12 to -6 range. The metering on the Zoom is, well not it's strong point. They are small short and hard to read.

There could be any number of reasons your files are "low" in the NLE. But first the question is are they low? The zoom (fisual in the NLE not the recorder) could be down and they look low but aren't. They will be low compared to pop music but that is because pop music (I'm using that term to cover just about everything but classical and some Jazz) tends to be compressed to with in an inch of it's life and crammed into the top few bits of the signal. You do NOT want your dialog to look like that. I'm assuming your transferring files in to your NLE not going analog out of the Zoom. If that is not the case then check your signal path, or better just transfer the files. And finally if you have been listening to music on your system then you probably have the volume too low on your outputs (remember pop music is WAY hot).

You could post a clip and people could tell you if it actually is low. I suspect it's not low.
 
If you're sending a line-in signal into your recorder, you HAVE to have it plugged into the line-in jack (on the DR-100, it's the 1/8" line only, not XLR).
 
i got stuck with having to use a zoom h4n recently. i was hired to do camera(something i still do quite a bit). production said they couldn't afford my kit for sound so they would have sound going direct to camera and said it was not going to be a dslr. (said would be af100, ex1/3 or similar camera). well i show up and low and behold, its a 7d and a zoom h4n and sennheiser g3 kit. all of the sound gear was brand new and had never even been opened. so i had like 10 minutes to get all that crap situated. i was not very happy with the recording quality or the usage of the h4n. i had an h4 a few years ago, but i remember it being much better and easier to use, but i also ran a mixer to it. i've said it once and i'll say it again, these recorders are not meant for working with audio for video… and i hope to never have to use one ever again! >:[
 
Digital audio has no "head-room" beyond 0 dB, which is why you should be setting audio levels to peak at either -12 dB or -20 dB. ( this is your "head-room" )

I would do a test to determine what the noise-floor of your set-up is. Record absolute silence with your set-up, and then use your editor to determine what the noise threshold of "silence" is, usually it's in the -60 dB or better range.

A really good set-up ( i.e. Sound Devices mixer + Sound Devices recorder ) can have a noise-floor lower than -100 dB ( your mics will probably have self-noise much louder than this ), so setting your audio to peak at -20, you would still have an audio range of 80+ dB.
 
So to sum it up:

Don't record so damn loud. Digital 0 (also known as 0dBFS or 0 decibels full scale) is equal to +18dB or +20dB (depending on who you ask) analog 0.

Record quieter and at 24-bits. The whole reason to have a 24-bit recorder is so you can record at lower levels and when you have audience applause or a screaming actor that throws your levels out of whack is so you can record quieter. Heck on a good recorder like the Sound Devices you can record stuff down to -30dB or -40dB with lots of headroom plus a limiter to catch anything before it hits digital 0dB. Then you just bring the level up in post.

When I started recording stuff quieter, mixing quieter and then bringing up levels in post as the very last step my recordings suddenly sounded better. Actually for video/DVD/Blu-ray my levels are around -18dBFS to -12dBFS. If I'm mixing for the web that's when levels are brought up to -0.1dBFS with a brickwall limiter on the end of the chain. This man put it better than me - http://www.massivemastering.com/blog/index_files/Proper_Audio_Recording_Levels.php . Yes, this applies to audio for video as well.
 
Steve House said:
"Headroom" over 0dB is a concept from the analog recording world and does not apply to digital.
Click to expand...

it appears that this may have been my issue then, I was not aware of this. I've spent years thinking 0dB is the ideal level.

When i import into premiere pro and the waveform is barely visible except for peaks, should I just apply audio gain to the signal then and there.. would this not reduce quality or add hiss in some way? Unless I'm wrong again, surely if my music is peaking just below 0db (but not touching it) then my dialogue should do the same (when not at the same time as music) in which case I need to apply audio gain in premiere?
 
Noiz2 said:
There could be any number of reasons your files are "low" in the NLE. But first the question is are they low? The zoom (fisual in the NLE not the recorder) could be down and they look low but aren't. They will be low compared to pop music but that is because pop music (I'm using that term to cover just about everything but classical and some Jazz) tends to be compressed to with in an inch of it's life and crammed into the top few bits of the signal. You do NOT want your dialog to look like that.
Click to expand...

As you've probably realised by now I know nothing about sound (I thought I did until this occurance) and I always assumed that your audio on a finished DVD should all be equal and peaking at 0dB or as close to it without any kind of distorting. When i watch a film I'm constantly turning volume down for large action scenes and up for quiet dialogue scenes and I hate it. its fine in a cinema but its a pain in the a$$ at home.

I have always followed a similar principle (or tried to) for my weddings, when the music before a ceremony fades out and speaking starts, i dont want my viewer to have to reach for the remote control.

Is this totally wrong?

I'm going to go back and listen to the raw files from the zoom now and play around with them. Are you guys saying it should be fine to add audio gain in my NLE due to the fact they were recorded in 24bit? Occasionally when (still with levels turned to max) somebody speaks very quietly, it barely shows anything on the waveform so in this situation I simply couldnt leave it without boosting the hell out of it.
 
tikigod19 said:
As you've probably realised by now I know nothing about sound (I thought I did until this occurance) and I always assumed that your audio on a finished DVD should all be equal and peaking at 0dB or as close to it without any kind of distorting. When i watch a film I'm constantly turning volume down for large action scenes and up for quiet dialogue scenes and I hate it. its fine in a cinema but its a pain in the a$$ at home.

I have always followed a similar principle (or tried to) for my weddings, when the music before a ceremony fades out and speaking starts, i dont want my viewer to have to reach for the remote control.

Is this totally wrong?

I'm going to go back and listen to the raw files from the zoom now and play around with them. Are you guys saying it should be fine to add audio gain in my NLE due to the fact they were recorded in 24bit? Occasionally when (still with levels turned to max) somebody speaks very quietly, it barely shows anything on the waveform so in this situation I simply couldnt leave it without boosting the hell out of it.
Click to expand...
It's not because they are recorded 24bit ... even if the originals are recorded in 16bit you can still make reasonable adjustments in the levels during editing. The noise floor differences between 16bit and 24bit really only enters the picture when you are mixing down a relatively large number of tracks into a single track because noise is additive when mixing. In fact the majority of dialog masters for theatrical motion pictures and network television are recorded in 16bit, 24 bit generally being the realm on music and FX recording.

You are absolutely correct in your desire to give the viewer consistent levels so they're not always reaching for the volume control ... to do this properly you need a controlled and calibrated environment from start to finish.

The visual height of the waveform envelope in your editor is a poor indication of actual signal levels since it depends on the zoom setting of the vertical axis in the display ... the same signal can be barely a squiggle or appear to swing all over the screen, depending on the zoom setting.
 
Last edited:
Digital 0 (also known as 0dBFS or 0 decibels full scale) is equal to +18dB or +20dB (depending on who you ask) analog 0.
Click to expand...

Kinda sorta, not really. 0dBfs is a fixed thing, that's all there is. Analog scales are based on a reference level. So 0dB on an analog meter can relate to just about anywhere depending on what your reference is. It used to be standard on Nagra's to use -18 as your zero reference because there is a nice tick mark at -18. When everything went digital meters started having the tick at -20 so that became the standard. Though in Europe I think -18 is still marked and often used. The only reason it matters these days is because there is still a lot of literature talks about zero (in the analog realm) and folks who don't know the background take it literally with digital meters and they get in trouble.

OK there are a lot of reasons it matters but for this discussion...

In the analog world you picked your zero and as long as you wrote it on the box so others knew how to calibrate the playback deck it was all good. In digital you got what you got and the most you can get is zero. So you need to build in the headroom by keeping your levels low enough that you don't hit the ceiling.



Record quieter and at 24-bits. The whole reason to have a 24-bit recorder is so you can record at lower levels and when you have audience applause or a screaming actor that throws your levels out of whack is so you can record quieter.
Click to expand...

Actually bit depth has nothing to do with headroom. Zero didn't change, you didn't get any more range. It's a good recommendation and your reasoning is sort of close. You get more resolution not more headroom. When you don't use all of the available bits your effectively recording at a lower bit depth. So at 16 bits you may effectively only be using 12 or so bits. By recording at 24 bits your using maybe 20 bits of resolution so it sounds better. You can record quite a bit lower and still have better resolution than 16 bit. If you record lower you give your self more headroom, but the bit depth didn't add the headroom you did by recording lower. The bit depth gave you a better sounding recording.

It may seem like splitting hairs but the distinction is important. Your trading resolution for headroom, 24 bit gives you more resolution to trade with.
 
so its just a case of recording quietly and then keyframing the audio line in premiere so that the quietly recorded speech is boosted and the loud applause is left alone? I wonder if there is some sort of way to do this automatically and just listen throgh to check it instead of having to fine tune it each and every time someone claps
 
tikigod19 said:
As you've probably realised by now I know nothing about sound (I thought I did until this occurance) and I always assumed that your audio on a finished DVD should all be equal and peaking at 0dB or as close to it without any kind of distorting. When i watch a film I'm constantly turning volume down for large action scenes and up for quiet dialogue scenes and I hate it. its fine in a cinema but its a pain in the a$$ at home.

I have always followed a similar principle (or tried to) for my weddings, when the music before a ceremony fades out and speaking starts, i dont want my viewer to have to reach for the remote control.

Is this totally wrong?

I'm going to go back and listen to the raw files from the zoom now and play around with them. Are you guys saying it should be fine to add audio gain in my NLE due to the fact they were recorded in 24bit? Occasionally when (still with levels turned to max) somebody speaks very quietly, it barely shows anything on the waveform so in this situation I simply couldnt leave it without boosting the hell out of it.
Click to expand...

Traditionally films also do a TV mix that is more compressed for this reason. I suspect that if you listen to the two track or Prologic version on the DVD it will be less dynamic and more oriented to TV watching. I think the 5.1 Dolby Digital tracks are closer to the theatrical mix. If your listening in "surround" then you should probably calibrate the center speaker a bit higher and that should even out the dialog level a bit with the big FX sounds.

And no you are not wrong and mixing for a TV listening environment is something you should be doing, especially with event videography.

Actually even if you want to squeeze as much level as possible out of your tracks you shouldn't peak above -3 dVfs. It's a bit technical but the converters "crest" above the max digital signal, so anything above -3 is likely to distort a touch.
 
tikigod19 said:
so its just a case of recording quietly and then keyframing the audio line in premiere so that the quietly recorded speech is boosted and the loud applause is left alone? I wonder if there is some sort of way to do this automatically and just listen throgh to check it instead of having to fine tune it each and every time someone claps
Click to expand...

I usually do this by compressing the clapping to bring it's volume closer to the volume of the rest of the audio.
 
TheDingo said:
I usually do this by compressing the clapping to bring it's volume closer to the volume of the rest of the audio.
Click to expand...

do you do that within premiere? What do you do use?
what sort of average peak levels would you say your project had within premiere?
 
tikigod19 said:
do you do that within premiere? What do you do use?
Click to expand...

I edit with Sony Vegas Pro and Sony Sound Forge. The compressor plug-in for Sound Forge is called "Wave Hammer" and I can apply it to sections of the audio track or the entire thing.

Normally when I edit, I export a .WAV copy of the video audio and work on it in Sound Forge. Vegas syncs everything, so when I jump back to Vegas my .WAV copy automatically replaces the original audio track from the video clip.

The process I normally go through when working with audio from a video clip is:

1- Export .WAV copy
2- Use Parametric EQ to tweak the overall sound of the audio
3- Use Compression to manually normalize the loud to quiet parts
4- Use a mild form of noise-reduction to remove any annoying background sound
5- Adjust the volume of the finished audio track to -0.1 dB
6- Save the .WAV copy then jump back to Vegas to finish the video edit ( my audio work is automatically synch'd with the video )
 
You must log in or register to reply here.
Back
Top