ZOOM H1 starter tips?

J

Jonathan12uiz

Well-known member
Hey guys, I'm new to the audio game and wanted to start out simple. I recently purchased a ZOOM H1 with the intentions of using a wired lav, Rode Video Mic, and bare H1 for ambience and room tone.

Right now I just have the H1, I was wondering what some decent settings are for:

Youtube/Vimeo edits?
Broadcast edits?
Cinema edits?

Right now I'm shooting in mp3 192kbps for simplicity and storage.

Can someone explain to me wav's 48/16bit or 48/24 bit? what exactly is the difference and what should I use for maximum efficiency? Will this be more sensitive in the recording process and retain more sound?

My workflow usually involves a timeline in FCP of prores LT 1920x1080 24p 48k or 1280x720 30p 48k

Also, I hear some hissing in my headphones during recording, but I barely hear it on playback on my pc... is that how its suppose to be? or does the H1 have some AGC hissing problems I am unaware of?

When monitoring levels, how do I set the proper level input setting? just play with it till the audio is just about to PEAK?
 
First things first: avoid recording in MP3 format. That's a lossy compression, and it has to be converted for FCP to be happy with it. Your audio within FCP runs at 48kHz, 16-bit. that, at the very least, is what you should be using. I'll get to what all that means in a moment. About your mics...

You say you're using a wired lav and VideoMic. I assume that's what you use for on-camera dialog? Yet you say you're using the H1's built-in mics for room tone. You should always record room tone on the same mic(s) used for dialog, since room tone is what's used to smooth out dialog edits. As soon as you're done with a scene, or an interview, record :30 of everyone just sitting still in the room. This is the natural noise floor that was present under the dialog, and through the same microphone.

Now, back to what all the numbers mean:

Uncompressed audio can have a few different formats (WAV or BWF, AIFF, SDII, etc.), but has two numbers that matter: sample rate and bit depth. The higher the number, the higher the resolution, and the more information is being stored.

Sample rate has to do with how many times per second the device is recording sound data, with each sample taken being equal to one cycle per second, aka one Hertz (Hz). At 44.1kHz, the device is taking a sample 44,100 times per second. 48kHz=48,000 samples per second.

This is digital audio. Unlike analog, where information is constantly recorded in an unbroken stream, digital takes "snapshots" at regular intervals. If the audio were to be plotted on a graph (X and Y axis), it would be a dotted line instead of a solid one. When the audio is played back, the device then has to play "connect the dots" to fill in the missing information. The more information (or the higher the sample rate), the smaller the gaps between samples and the better the quality of reproduction. 44.1kHz is actually pretty darn good, and has been the native sample rate of CD audio since CDs came out. Video runs at 48kHz.

Every sample that is taken during recording is a word, and that word is made up of bits. This is all binumerics, or 1s and 0s. The higher the bit-rate, the longer the word and the more information is recorded at that moment. With 16 bit audio, each sample has a word that contains 16 digits: 0110010111001001, for example. If there's nothing there, it's all 0s: 0000000000000000. If the information is full, it's all 1s: 1111111111111111. At that point, we hit clipping (the 0dB threshold) and start to hear nasty square waves. That means the levels are too high and we cannot pack any more information into the signal.

Adding bits does not mean we can get higher. 0dB is 0dB and is the limit no matter what. What a higher bit rate does mean is that finer details can be recorded in softer levels. With 16-bit audio, recording too low means that you're only using up half the bits (or less). That makes for lowered quality on playback, and it also means that pushing the signal in post with too many plug-ins (compression, EQ, etc.) will make the audio warble (think of it as the digital sound equivalent of a low-rez digital picture, where it becomes blocky and pixellated). With 16-bit audio, you really don't want to drop below -24dB. At 24-bit, however, lower levels can be recorded cleanly. It's also very common for sound to be recorded in 24-bit, and then converted to 16-bit later before distribution. That allows sound designers and editors to do more with the audio in post without the fear of degrading the signal integrity.

MP3 and other compressed formats remove lots of the information because they are designed to fit more audio into a much smaller package. They use really tricky algorithms that are supposed to remove information the human ear typically won't notice as missing, but it's not that accurate. MP3 is also a distribution format and not an acquisition or editing format. In order to really work with it, the MP3 has to be "uprezzed" to a WAV file. Well, there's lots of missing information, and the computer has to try and fill it in... and the computer isn't so good at doing that. This results in artifacting in the audio.

So to answer your question, record in WAV format and at either 48/16 or 48/24.

As far as levels, do not set it to "just about to peak." That's a danger zone. For dialog, you should be averaging between -22dB and -12dB, with occasional peaks (sudden, loud noises like coughs and claps) at -6dB.

If you're noticing hiss in the headphones during recording, but it's not showing up in the actual recording, I'd say the headphone amp in the H1 is the culprit. I haven't handled the H1 personally yet, but that's probably a safe assumption.
 
C2V said:
First things first: avoid recording in MP3 format. That's a lossy compression, and it has to be converted for FCP to be happy with it. Your audio within FCP runs at 48kHz, 16-bit. that, at the very least, is what you should be using. I'll get to what all that means in a moment. About your mics...

You say you're using a wired lav and VideoMic. I assume that's what you use for on-camera dialog? Yet you say you're using the H1's built-in mics for room tone. You should always record room tone on the same mic(s) used for dialog, since room tone is what's used to smooth out dialog edits. As soon as you're done with a scene, or an interview, record :30 of everyone just sitting still in the room. This is the natural noise floor that was present under the dialog, and through the same microphone.

Now, back to what all the numbers mean:

Uncompressed audio can have a few different formats (WAV or BWF, AIFF, SDII, etc.), but has two numbers that matter: sample rate and bit depth. The higher the number, the higher the resolution, and the more information is being stored.

Sample rate has to do with how many times per second the device is recording sound data, with each sample taken being equal to one cycle per second, aka one Hertz (Hz). At 44.1kHz, the device is taking a sample 44,100 times per second. 48kHz=48,000 samples per second.

This is digital audio. Unlike analog, where information is constantly recorded in an unbroken stream, digital takes "snapshots" at regular intervals. If the audio were to be plotted on a graph (X and Y axis), it would be a dotted line instead of a solid one. When the audio is played back, the device then has to play "connect the dots" to fill in the missing information. The more information (or the higher the sample rate), the smaller the gaps between samples and the better the quality of reproduction. 44.1kHz is actually pretty darn good, and has been the native sample rate of CD audio since CDs came out. Video runs at 48kHz.

Every sample that is taken during recording is a word, and that word is made up of bits. This is all binumerics, or 1s and 0s. The higher the bit-rate, the longer the word and the more information is recorded at that moment. With 16 bit audio, each sample has a word that contains 16 digits: 0110010111001001, for example. If there's nothing there, it's all 0s: 0000000000000000. If the information is full, it's all 1s: 1111111111111111. At that point, we hit clipping (the 0dB threshold) and start to hear nasty square waves. That means the levels are too high and we cannot pack any more information into the signal.

Adding bits does not mean we can get higher. 0dB is 0dB and is the limit no matter what. What a higher bit rate does mean is that finer details can be recorded in softer levels. With 16-bit audio, recording too low means that you're only using up half the bits (or less). That makes for lowered quality on playback, and it also means that pushing the signal in post with too many plug-ins (compression, EQ, etc.) will make the audio warble (think of it as the digital sound equivalent of a low-rez digital picture, where it becomes blocky and pixellated). With 16-bit audio, you really don't want to drop below -24dB. At 24-bit, however, lower levels can be recorded cleanly. It's also very common for sound to be recorded in 24-bit, and then converted to 16-bit later before distribution. That allows sound designers and editors to do more with the audio in post without the fear of degrading the signal integrity.

MP3 and other compressed formats remove lots of the information because they are designed to fit more audio into a much smaller package. They use really tricky algorithms that are supposed to remove information the human ear typically won't notice as missing, but it's not that accurate. MP3 is also a distribution format and not an acquisition or editing format. In order to really work with it, the MP3 has to be "uprezzed" to a WAV file. Well, there's lots of missing information, and the computer has to try and fill it in... and the computer isn't so good at doing that. This results in artifacting in the audio.

So to answer your question, record in WAV format and at either 48/16 or 48/24.

As far as levels, do not set it to "just about to peak." That's a danger zone. For dialog, you should be averaging between -22dB and -12dB, with occasional peaks (sudden, loud noises like coughs and claps) at -6dB.

If you're noticing hiss in the headphones during recording, but it's not showing up in the actual recording, I'd say the headphone amp in the H1 is the culprit. I haven't handled the H1 personally yet, but that's probably a safe assumption.
Click to expand...

Thank you sir for your wealth of great knowledge.
 
You must log in or register to reply here.
Back
Top