About 10 years ago, I threw together this post about how to choose a recorder for recording frogs. Well, a lot has changed in the intervening 10 years, so I guess it is time to update my thoughts on choosing a recorder and talk about some of my recommendations of recorders that are good enough to help you capture your local frogs and "frogscapes". There is a lot of terminology associated with audio recording that you may not know, so there is a brief (well, not very brief really) "glossary" at the end of my list as well.
NB - this post is not complete yet, I want to add some discussion about several other recorders and will do so soon.
My recommendations are not based on some extensive knowledge of electrical engineering, sound editing or any other technical expertise. They are based on
- having owned too many recorders - GAS (Gear Acquisition Syndrome) is a serious disease 😜,
- having stood out in the field all over the world, at night, in the rain trying to get recordings for the last 20 years, and
- having done a lot of reading, discussing and listening on various nature recordist groups online.
These are my opinions based on my experience. Many will disagree. My only goal is to give an updated (for 2026) list of recorders you might consider if you are setting out into the world of field recording.
Talking about recorders does require a fair bit of jargon, so I have included a "glossary of terms" at the end for those new to recording jargon.
The following recorder suggestions are not ranked, they are just "recorders you should consider" if you are starting out in this hobby. I will cover the recorders I have used first, then add a few I am familiar with from my online investigations.
Recorders with built in microphones
Zoom M3
 |
| Zoom M3 Recorder/Microphone image from Zoom Website |
Price - $199 USD at the time of writing
Pros:
- very small, light and field hardy
- "32 bit recording" (dual ADC)*
- shotgun microphone* plus mid-side recording*
Cons:
- not the quietest microphone/recorder on the market (particularly the side mics)
- button system takes some getting used to and isn't easy to see in the dark
- can't play back recordings in the field (except for the last recording)
- plastic case can lead to handling noise (true of any plastic recorder)
I started my list with this one because it is probably the most controversial and because I love this little recorder. The Zoom M3 is an "all-in-one" recorder than includes a short shotgun microphone and a figure eight microphone (or maybe two cardioid? I'm not sure what's inside it?). This combination allows you to capture the call of an individual singing in a noisy environment using the shotgun mic or to capture a stereo ambience of the whole scene using mid-side recording. Once you get the recording home, you can adjust the gain (loudness) of each of these channels to bring out more of the ambience or more of the 'soloist'. So you don't have to choose which type of recording you want while you are in the field. You just point the M3 towards your frog and press record.
Part of the "problem" the M3 has is not really a problem anymore. When it was first released, some of the original M3s had significant problems with picking up interference from electromagnetic sources such as cell phones, WiFi antennas and even some radio waves in very densely populated areas. This was apparently due to insufficient electrical shielding in some units. However, Zoom recalled and replaced all the units with this fault at no charge and the newer units do not have this problem. So the M3 available now is at least on par with any other plastic mic/recorder combinations on the market. So if you go online or onto YouTube, many of the reviews by "professional" reviewers were about these faulty units. But most of those reviewers failed to evaluate the fixed versions. They didn't like the faulty ones, gave a bad review and never looked back. Such is the way of the "professional" reviewer on YouTube, I guess.
As for the handling noise, I get around that simply by mounting it on a short "table top" tripod (I like the qubo mini tripods on Amazon) or a rubber camera grip handle with a 1/4-20 tripod screw. They are available on Amazon for less than 20 dollars. With the recorder on a grip/tripod, I can still tuck the recorder into a pocket for carrying in the field. It only takes a few seconds to pull the recorder out, turn it on, point it at my target and start recording.
It does have a headphone out if you wish to monitor your recording live or to help you point directly at a single individual.
So, my overall thoughts on this recorder? I think this is a great recorder for someone starting out with recording amphibians. Are there better recorder/shotgun microphone or mid-side combinations out there? Yes, but not at this price point and not in this small, field friendly form.
Here are some example recordings I've made using the Zoom M3:
Red-eyed Treefrog and Rosenberg's Gladiator Frog from Costa Rica
Jabiru (Fat) Toadlet from Kakadu National Park, Northern Territory, Australia
Zoom H2essential
 |
| Zoom H2essential image from Zoom website |
Price = $179USD at the time of writing (+$49 for the Zoom APH-2e custom windscreen)
Pros:
- very small, light and field hardy
- 32 bit recording (single ADC?)*
- mid-side recording
- capable of multiple styles of stereo recording
- quiet (can't find a listed EIN value), but my tests say it is very quiet
Cons:
- side mics aren't as quiet as front/back microphones (true of most built in M/S systems)
- button system takes some getting used to
- plastic case can lead to handling noise (true of any plastic recorder)
I started using this recorder recently and used it a lot on my 2025 trip to Costa Rica. For an all around "ambience" recorder or to record a large chorus of calling frogs, this is one of my new favorites. It is a tiny, all-in-one recorder with good quiet mics and preamps. (NB - the predecessor, the Zoom H2 and H2n were not as quiet as the H2e nor did they offer 32 bit recording - you want the H2e aka H2essential, not the H2 or H2n).
One of the things I like about this recorder is its small size and great sound. It allows you to record with just the front microphone, the front and side microphones (Mid/Side recording) or the front, back and side microphones for double Mid/Side recording. This gives you a lot of flexibility in the type of recording you want to produce out in the field.
It does need wind protection, but there is an available custom windscreen that comes in the APH-2e accessory pack you can buy (and you should!). It has a 1/4-20 thread on the flat bottom so you can stand it up on a flat surface or attach it to a short (or tall) tripod or even hand-hold it on a 1/4-20 threaded handle.
The only thing this recorder is not optimal for is selecting out one individual in a chorus. The front microphone is cardioid so it can focus in somewhat, but it is not a shotgun mic and lacks its side rejection. But for stereo recording this is one of my new favorites. You don't need extra microphones or cables, just this small recorder (it fits in a pocket easily) and something to stand it on and you can be out recording.
It runs on 2x AAA batteries which don't last a long time, but you can also power it by USB-C using a powerbank and get days worth of recording out of it if needed.
Roland R-07
Price = $250 USD at the time of writing (but can be found much cheaper on sale)
 |
| Roland R-07 image from Roland website |
Pros:
- very small, light and field hardy
- nice stereo field from a small pocket-sized recorder
- quieter than most pocket recorders
- 3.5mm stereo input for using with better microphones
- comes in colors other than black!
Cons:
- difficult to protect from wind
- 24 bit only
- fidgety buttons
- flimsy cover for micro SD card
- overpriced at full retail price
- micro-USB connector
- quiet-ish (A weighted EIN of -117dBu - or maybe less?*)
I originally bought the R-07 because I wanted a small, pocket recorder that I could have with me anytime I am in the field that could quickly capture some calls or ambience without any complicated setup. I had the Zoom H1-n and although that was a decent recorder, I didn't like some of its features (the position of the gain dial up by the microphones, for example).
I bought this recorder when it was on sale on Amazon for $99USD (for the red or white ones, the black one was still full price). I compared it side by side with my Zoom H1N and found the Roland to have noticeably less input noise. It was also a better form factor for fitting in a pocket than the longer Zoom H1n. I actually ended up buying a second one about a year later after I misplaced my original. I found it again so now I have two?
Overall, this is a great little pocket recorder. Simple to use with reasonably quiet inputs, I think it was one of the best options in its class when it first came out. It is still available even though it is an older unit.
It comes in black, red or white. The reason I like this is that the red or white versions are a lot easier to find if you set them down in the field! That might not be to everyone's taste though and so there is a traditional black version available.
A couple of things I don't like about this recorder are its buttons (they feel like they are going to break....but haven't yet?) and the position of the microphones. While the microphones are positioned nicely to give you a good X/Y stereo image, they are very difficult to protect from wind. Yes, you can get windscreens for cell phones, etc. that will slide over the top of this recorder, but those will slide off easily if bumped and they cover up part of the LCD screen.
Also, the cover of the SD card slot is a very flimsy plastic thing that won't stay closed and seems destined to fall off and be lost (one of my two units is missing this).
In spite of those negatives, I do like the convenience and sound quality of this little pocket recorder. It has an A weighted EIN of -117 which is right at the threshold that most purists regard as "quiet enough" for good field recording. But if you are actually recording frogs and not just trying to capture the silence of a gentle breeze blowing through grass, this recorder is quiet enough to get you excellent recordings.
* While this recorder is pretty quiet at max gain (-117dBU), some users have reported that you can significantly decrease the input noise by not turning up the inputs all the way. Keep the inputs at a lower gain level and the relative noise will decrease. What the input level is at those gains I do not know, but quieter than -117dBu puts it into the pretty damn quiet category?
So in 2026 do I still recommend this recorder.....yes, if you can pick one up for $100 or less (often possible). But if you are paying its full retail price of $250, the Zoom H2essential is a better recorder and cheaper (less than $230 with it accessory windscreen).
Recorders with no microphones (external microphones necessary)
 |
Left to Right: Zoom H1-XLR, Zoom F3, Tascam FR-AV2 |
Zoom H1-XLR
 |
| Zoom H1-XLR image from Zoom website |
Price = $169 USD at the time of writing
Pros:
- very small, light and field hardy
- 32 bit recording (dual ADC?)*
- can be used with a variety of 3.5mm and XLR microphones
- can be attached directly to most shotgun microphones
- quiet (reported EIN -122 dBu A weighted)
Cons:
- small screen and buttons (you get used to them)
- plastic case can lead to handling noise (true of any plastic recorder)
- no built-in microphones
Honestly, I purchased this recorder because of its tiny size and interesting specs. This recorder is capable of capturing two XLR microphone inputs (or 3.5mm stereo input) at 32bit in a tiny recorder that can easily fit in a pocket.
There are several older iterations of the Zoom H1, but the Zoom H1-XLR is the only one with XLR type inputs.
I have used this recorder attached to a small tripod with two XLR microphones attached and posititoned in such a way to create a nice stereo field. However, I have also used the Zoom H1-XLR as a "stand alone" recorder attached directly to my shotgun microphone with no cables for a one-handed long shotgun field recording kit. I wouldn't try that with any other recorders because it would put too much torque on the XLR plug to mount a heavier recorder. But because this recorder only weighs 164 grams (5.7 oz) with its two AA batteries, I don't think that weight represents a threat to the XLR mount on my shotgun microphone. I use a Movo shock mount to help support the XLR connection and I had to put a short XLR female to male extension in the middle because the microphone diameter is a bit thick for the XLR receptacle on the recorder. Here's what the whole rig looks like (I have a furry windscreen over the microphone in use).
Here is a recording I made using the above setup in Panama. I was walking along the Pipeline road one night listening for frogs and I happened to capture this Crested Owl calling near the road (the ray-gun "peow" you hear is the Tungara Frogs calling nearby).
Zoom F3
 |
| Zoom F3 image from Zoom Website |
Price = ~$300 USD at the time of writing (prices vary a lot between $299 and $340)
Pros:
- small and field hardy
- tough metal body
- true 32 bit recording (dual ADC)
- two XLR inputs
- 'set and forget' recording
- extremely quiet (EIN = -127 dBu A weighted)
Cons:
- small buttons
- small monochrome screen
- heavy for its size
- no built-in microphones
- no 3.5mm microphone input (only XLR)
- unusual record switch takes some getting used to
When a new member joins on of the forums or social media groups about nature recording and asks for recommendations for a good, affordable starting field recorder many respondents will quickly jump in and suggest the Zoom F3. And they aren't wrong.
The Zoom F3 is often regarded as one of the best field recorders on the market in its price range. It certainly is built like a tank and has very quiet pre-amplifiers. It was one of the first quiet, affordable 32bit recorders on the market and has proven itself over and over again in field recording all across the world.
For some experienced recordists, the Zoom F3 takes some getting used to. It relies on its 32 bit float and dual AD converters to be able to capture whatever sounds you try to record without clipping or excess noise. Therefore, there really isn't any input gain control. To start recording with the F3, you plug in your mics, put them in position and press record. No adjustments are necessary in the field. That takes some trust and getting used to if you have been using older 24bit recorders in the past. But rest assured, it really works. Any adjustments can be safely made in post.
Unlike many recorders which use a "moving bar graph" to indicate input strength, the Zoom F3 uses a waveform viewer instead. This takes some getting used to, but once you learn to trust this unit's excellent pre-amps and dual 32bit A/D converters, you don't really need to visually monitor your recording. Just press record.
Although you will have to BYO microphones to use this recorder, it is an excellent recorder and there really isn't a better one on my list. There are some that are probably equivalent (the Tascam FR-AV2, for example), but not any that will produce better quality recordings.
The only thing that prevents this from being the obvious first choice for everyone starting out in nature/frog recording is the fact that you have to get your own microphones and they have to have XLR plugs, not 3.5mm plugs. But the best microphones for field recording come with XLR plugs anyway, so this isn't a huge deficit.
A couple of idiosyncrasies of this recorder that are worth mentioning would include the numbering of the XLR inputs and the record switch. On most recorders I have used, the tradition is that the XLR plugs are numbered from the left to the right. So the leftmost input is track 1, the next one to the right is track 2. But on this recorder they aren't exactly like that. I think the XLR inputs are meant to be visualized from the point of view of the recorder being flat with the screen right side up. The XLR inputs are then at the top of the recorder and the one on the left side is input 1 as expected. But because of this positioning, when the inputs are facing you as you are plugging mics in (if you are holding the recorder in your hand), input 1 is on the right and input 2 is on the left. This is only a minor issue for people who are used to plugging left into 1 and right into 2. I have had to switch the tracks on a few of my recordings because I plugged my mics into the "wrong" inputs. Easy fix and you get used to it, but it can be confusing at first if you don't notice the numbering.
The other minor niggle I have about this recorder is the way you start recording. There is a hold switch on the right upper side of the recorder that you slide forward to start the recording. This works fine, it just feels like something that will break? I've had my F3 for years and it hasn't, it just "feels" like it might break.
Zoom also make a larger "version" of this recorder with 6 inputs called the Zoom F6. It is a bit larger and heavier, but has some extra features like gain knobs and will take 6 XLR inputs at once should you need that. It runs around $750.
Tascam FR-AV2
Tascam FR-AV2 image from Tascam website
Price = $399 USD at time of writing
Pros:
- small and field hardy
- tough plastic body
- true 32 bit recording (dual ADC)
- two XLR inputs and 3.5mm stereo input
- small, but bright, colorful screen
- function buttons with different functions on different screens
- large record button
- traditional gain bars visible on screen to monitor input
- very quiet pre-amps (EIN -127 dBu A weighted)
- timecode capable (for syncing with video)
Cons:
- small buttons
- no built-in microphones
- bright color screen (might frighten animals?)
I believe the Tascam FR-AV2 was Tascam's answer to the very popular Zoom F3. When the Zoom F3 came out, there really was no competitor with such high quality pre-amps and 32 bit recording anywhere near the price range of the Zoom. But then along came the FR-AV2.
Like the Zoom F3, it is a pocket-sized recorder that is capable of capturing nature sounds with almost zero input noise from the recorder. But the Tascam FR-AV2 has a few new tricks up its sleeve that are lacking in the F3.
The Tascam has the capability to generate Timecode which can be used to synchronize an audio and video recording. If you don't need to do this, this isn't really much of a benefit. But it also has a 3.5mm stereo input jack so you can use 3.5mm terminated microphones as well as XLR terminated microphones. This was always a deficit in the Zoom F3.
The Tascam is a little bit larger than the Zoom when seen from above but the Zoom F3 is noticeable thicker, so they are probably both about the same volume (photo above shows them side by side).
As a reflection of the direct competition these two recorders, a few months before the Tascam came out Zoom lowered the price of the F3 from ~$400 to ~$300 to make it less than the Tascam. If they were the same price, I think the Tascam's extra features would make it the better choice, but for 25% less, the Zoom is still a good choice.
I should add here that I am not one of those Luddites who believe everything made of plastic is cheap junk and metal is better. I think that is a perception that stopped being true some 50 years ago with the development of modern plastics. Today's plastics are as strong as many metals and good plastic recorders are just as field hardy as any metal ones in my experience. Yes, the cheapest plastic ones are not, but this unit is not one of those. This is an excellent quality recorder that happens to have a plastic body to reduce weight.
I haven't had my Tascam as long as my Zoom F3, but they are both great recorders. I can't say I prefer one over the other as they are both capable and trustworthy. The Tascam does take three AA batteries rather than the Zoom's two, but it is likely the Tascam's larger color LCD uses more power so the battery life is a wash. And I generally use external power through the USB-C ports on these recorders most of the time anyway. The Tascam is a little bit lighter overall as well.
One thing I like and dislike about the Tascam is its record button. Unlike the Zoom F3, this recorder has a large, round record button. I prefer that. However, the Tascam requires you to press record twice to start a recording. The first press arms the system and starts sending power to the microphones and the recording light blinks. But to start recording you have to press it a second time. Now this is fairly normal in many audio recorders (but not the Zoom F3). But it is something you have to watch out for. You could press record and walk away and then come back an hour later to find out nothing had been recorded! Just something else to remember. It is always a good idea to check the time indicator to make sure it is moving/counting up!
In order to get this post up, I have decided to stop for now and post it incomplete. There are a number of other good recorders I would like to discuss, but if I go through them all, I may never get done. So for now, if you are in the market for a new recorder in 2026, you should also include the following recorders in your investigation! I will expand on them later.
Other good recorders I own and use (and recommend) include:
- Sound Devices Mix-Pre 3 (version ii is better, but version i is great)
And some good recorders I don't own but are worth investigating
- Tascam Portacapture X6 and X8
- Zoom H4Essential
- Zoom H5 Studio and H6 Studio
I will post more on these and update this post soon.
Glossary:
Signal and Noise in recordings - Whenever you start investigating recorders or recording, you quickly come across the term Signal to Noise Ratio. So what does that mean, and why is it important?
Generally, the signal is what you are trying to record (the frog's call or the chorus). The noise is the part you didn't intend to record. The noise can interfere with your ability to hear the signal. If you are recording next to a busy highway, the sound of the traffic (noise) might drown out your signal (frog calls). I know the pedants will raise objection to my simple definition, but it works for what we are talking about. So the goal of our recording is to maximize the ratio of the signal (wanted sounds) to the noise (unwanted sounds).
In order to maximize our signal to noise ratio (SNR), we have to think about where the noise is coming from. There are two general categories of noise in recordings - ambient noise and system noise.
Ambient Noise
Ambient noise is things in the environment that you pick up with your recorder that you didn't intend to pick up. This could be traffic noise, airplanes flying over, people talking, dogs barking, etc. It can even be other frogs. If you are trying to record species A and species B is closer and its calls are really loud and drowning out your target species, you could regard the calls of species B as noise over your desired signal species. Insects drive me crazy this way! You set out your gear to get a nice frogscape ambience and some stupid cricket lands right in front of your microphone and "audio bombs" your recording.
This happened to me while I was trying to record the wonderful "doops" of Ornate Burrowing Frogs (Platyplectrum ornatum) in Western Australia. I set my stereo rig up to capture the "doop party", but I didn't see the Desert Treefrog (Colleeneremia rubella) sitting under the grass just next to the left microphone. So for my recording purposes, this was "noise" I had to remove in post.
So how do you reduce ambient noise in a recording? Mostly that is a matter of technique. Choosing a microphone with the right polar pattern (shotgun, omni, cardioid) for the situation is a great starting place. But then mic placement can make a big difference. If you have a somewhat directional microphone (cardioid or shotgun), position the microphone so the noise is off axis. This would have been the solution in my example above. If the frogs you are recording have a busy road behind them, move over to another position to put the busy road behind or perpendicular to your microphone to reduce how much of it is being picked up.
Also consider the noise you are contributing to the recording. Omni mics will pick up your breathing, as will shotguns if you are directly behind them. Stand further away or hold the shotgun in such a way that you aren't directly behind it. Don't move since the microphones will pick up your footsteps. Don't talk (getting the people with you to shut up is one of the perennial curses of the field recordist!). Don't handle the microphone or recorder, or at least, hold them very still. The sounds of your hands rubbing against the recorder or microphone will be picked up. Be careful of overhead electric wires and electric fences. They can sometimes cause interference sounds on your recording. Turn off your cell phone or keep it away from your recorder/mic. Some equipment can pick up noise from cell phones or wifi signals.
By using good technique, you can significantly reduce the ambient signal that is interfering with your ability to record your target.
But there is some noise inherent in the recording process that isn't coming from the environment, it is coming from the recording system itself.
System Noise
System Noise is noise created by the "system", i.e. by the microphones and recorders.; This is noise that was not present in the environment you were recording in. This includes sounds like the wind buffeting your microphone but also more technically generated noise like the noise that comes from amplifying the microphone signal as it comes into the recorder.
Microphones produce an electrical signal in response to sound vibrations. That electrical signal is what travels up the cable and is eventually recorded in a digital format by the recorder. But the electrical signal coming out of the microphone is very weak, so it must be amplified before it can be recorded. This is called "pre-amplification" and is carried out by part of the system called the pre-amplifier, or pre-amp. Then the amplified electrical signal has to be converted from analog to digital format so that it can be written to the SD card or other device.
These two steps, preamplification and analog to digital conversion, are imperfect. Sometimes, some extra electrical noise is added to the original recording during the process. So the microphone, preamps and A-D conversion can add extra "sound" to your recording. This usually shows up as hiss or even a buzz in a recording. Higher quality microphones and recorders generally have lower noise levels (i.e. they add less system noise to your recording).
The problem this system noise presents depends on a signal to noise ratio (SNR). If you have a loud signal (loud frog) and quiet gear, your SNR will be high. That is good - more signal, less noise or more of what you were trying to record, less system noise. Most of what you record will be what was actually going on in the environment.
But if you have noisy gear and a weaker signal (like in quiet nature ambience recording), the noise becomes more obvious.
Reducing System Noise
Once the system noise is in your recording, it is hard to get out. Recording a distant frog call and then amplifying it when you play it back (increasing the volume) will make the noise louder as well, so it doesn't improve the recording. But if you can increase the signal strength by getting your microphone closer to the frog, you will have more signal without increasing the system noise. Therefore, your SNR will be higher - more signal for the same amount of system noise. Then when you turn up the volume while listening, the signal will still be louder relative to the noise. This is why getting close to the subject is so important!
So even if your gear isn't the quietest, you can still keep your SNR high by getting close to your subject, thereby increasing the signal level. This is why people recording rock concerts don't worry too much about SNR, people recording the sound of a gentle breeze blowing through grass have to worry about it a lot.
Having said all that, a lot of nature recordists obsess about system noise. But for recording frogs, you can relax a little bit. Frogs aren't quiet. And if you can get reasonably close to the frog/chorus, you can generate enough signal to overcome the system noise and keep your SNR low. And if you have a good/interesting recording and a reasonably low noise level, no one will notice (except the audio snobs 😛).
If you are trying to compare system noise of different recorders, the best comparison parameter is the Equivalent Input Noise (EIN). For accurate comparisons you want to compare the A-weighted dBu value for the EIN. (Do some googling if you want to know what A-weighted and dBu are - but fair warning, it is a deep rabbit hole).
In general, the lower the value (more negative) is better. For nature recording, most people seem to feel anything with an A-weighted EIN below -118dBu is fine for field recording. Some of the quietest recorders have values of -130dBu although anything less than -125dBu is REALLY quiet. Again, this matters when you are trying to record silence. When you are recording a normal chorus of frogs, any EIN less than -110 is probably just fine. There are a lot of recorders that were considered "great" for field recording 20 years ago with EIN values of -112dBu. It is really only with the development of affordable recorders with EIN values in the -125 or less range that has made people start to look less favorably at these -112 to -115dBu recorders.
It reminds me of the pixel wars with cameras. Previously pro-level cameras had pixel counts that are silly by today's standards, yet were good enough for professionals back then? In 2004, Canon upgraded its top of the line, pro-level EOS-1D from 4 megapixels to a massive 8 megapixels. Now that is considered way too low even though most people wouldn't be able tell the difference looking at the images on screen.
I think the same is true of recorders. If a recorder with an A-weighted EIN of -112dBu was good enough to get nice nature recordings in 2005, why is it too noisy now? Is it just because there are quieter options available? So buy the quietest recorder (and microphone) you can get within your budget range and don't worry that there are recorders that are theoretically quieter. Get out there recording!
Gain vs. Volume - When you first start out recording, you come to the hobby with an understanding of what the volume of a recording means. It is the perceived loudness to your ear of the output of the recording. Gain is slightly different. Gain is the measure of the strength of the electrical signal coming in to the recorder's processing system. You increase the gain as you are recording, you can increase the volume as you are playing it back.
So if you increase the gain during recording, you will get more volume in the output (assuming you leave everything else the same). But if you have a weak signal in your recording (it is too quiet), you can't make it "better" by increasing the volume in playback. That is because you will increase both the signal AND the noise. You can reduce this problem by increasing the gain of the input into the recording. Then you are increasing the signal without amplifying the noise to the same degree. More on signal:noise ratios above.
It might be useful to think of the analogy of the exposure of a photograph vs what it looks like on screen/print. You can't recover details in the whites of an overexposed photograph in printing, you have to do it in the actual exposure (while taking the photo) itself. Adjusting the exposure while taking the photo is analogous to adjusting the gain. Adjusting the exposure while printing (or on-screen) is analogous to volume.
Decibel Scale - The loudness or intensity of a sound is measured by the decibel scale. Briefly, on the decibel scale, 0 decibels represents absolute silence and a gun firing at close range is 140 dB or higher. A soft whisper is around 30 dB while a lunar rocket launch is usually near 180 dB. Any sounds over 130dB are painful to the ears and can cause ear damage. Lower decibel levels can cause ear damage as well, depending on exposure time.
The decibel scale is logarithmic and relative. For example, if you double the sound energy of a source, the sound increases by 6dB. But to make it be perceived as twice as loud, you have to increase it by 10dB (as you would expect in a log scale). Confused? Yep, me too. This is because a decibel can be used to represent several different measurements. There is a lot of complex math involved in the quantification of sound energy and if you want that answer, feel free to search online to go down that rabbit hole. 😉 There is actually a pretty good (confusing) discussion of decibel scale and math here.
I just remember that something around 20 decibels is barely audible and 120 is REALLY loud. And every 10dB you increase the sound makes it sound approximately twice as loud.
32 bit recording - the bit depth of a recording is a way of describing the dynamic range of that recording. This is a measure of how much detail is captured between the quietest parts and the loudest parts of the recording.
For those familiar with photography, it is analogous to the dynamic range of a photo, i.e. how much detail there is the darkest and lightest areas. If something is brighter than the dynamic range can capture it shows up as pure white in your photo. If something is darker than the dynamic range can capture, it shows up as pure black (even if your eye could see detail there in the field).
Dynamic range in recording is similar. If something is too quiet to be picked up within the dynamic range of the gear, it will show up as silence. If it is too loud, it will be distorted or clip. Once that sound clips in your recording, it can not be recovered by turning down the volume/gain. The recorded sound is permanently clipped. Again, this is similar to how you can't make the details come out in a photo that is too far overexposed by decreasing the exposure after the picture is captured, there is no information there to recover. Clipped recordings cannot be unclipped by turning them down.
With older 16bit or 24bit recorders, it is possible to have your recording clip if what you are recording has too much dynamic range because it is too loud and too quiet. If you set the recording gain high to pick up the quiet sounds, the loud sounds will be lost to clipping. If you set the recorder gain lower to successfully capture those loud sounds, the quieter sounds will be lost as silence. A 32bit recording (or more correctly a 32bit float recording) has so much dynamic range, you can't really lose information like this, assuming your microphone is capable of capturing that sound.
For a 16 bit recording, the recording is capable of capturing 96dB of dynamic range between the quietest and loudest parts. But if any sound exceeds that range, it will be lost or clipped. Even if you reduce the gain (volume) on your computer in software, the lost signal cannot be recovered because the recorder was not able to capture it. This is not to say that 16bit recording is bad, just that it is limited to 96dB of dynamic range. That is still a big range! That is the difference between a whisper and a gunshot, so you could successfully capture both in the same recording at 16bit. But the problem is if you come across unanticipated sounds beyond that 96dB range. Those will be lost in your recording if they are too quiet or will clip if they are too loud.
With a 24 bit recording, you can capture 144dB of dynamic range. That is enough for most recording. With 144dB of dynamic range, you can pretty much capture the full range of human hearing (in regards to loudness). The problem is setting the gain perfectly in advance of pressing the record button (part of the process recordists call gain staging). So if you are trying to capture a quiet nature scene and set your gain to capture all the quiet sounds, a really loud sound close to microphone might clip. Cricket Frog clicks can do this as can any hidden frog that is too close to your microphone. If you put your microphone/recorder too close to a really loud frog, you could still get clipping.
In a 32bit float recording, you can capture a theoretical 1528dB of dynamic range. In effect this means no matter how loud the sound is, you can capture it without clipping. You may need to reduce the gain (volume) after recording, but the signal will be there. You can also bring up really quiet areas of a 32 bit recording without introducing extra noise to your recording. BUT.....
There are a couple of important caveats to add here:
1. No recorder is actually capturing a 32bit recording. For mathematical/electrical engineering reasons I don't fully understand, this is an electrical impossibility for commercial recorders at this time. What the recorders are doing is capturing two 24bit tracks simultaneously using two separate recording pathways (two Analog to Digital Converters - ADCs). One records the quiet sounds and the other records the loudest sounds at a lower level. Therefore between the two tracks, you have captured the loudest and quietest parts of the sound. The recorder then combines these two tracks in a single 32bit float file which can then deal with the incredible dynamic range you have captured.
2. Not all 32bit recorders even record 32bit in the advertised sense because many of the lower end "32 bit" recorders have only a single ADC. This means they can either capture the loudest sounds or the quietest sounds, but not both. These recorders lose the benefit of having the extra track to use for one extreme. Therefore, they are really just creating a 24bit recording and placing it into a 32bit file. You get larger file sizes without the benefit of the extra dynamic range. Of course, a 24 bit recording does have 144dB of dynamic range which is plenty for what we record, and results in smaller file sizes. Now there are clever technological workarounds for this, but it seems unlikely that these manufacturers would be putting their more sophisticated ADCs in the low end products only?
However, all of this depends on whether your microphone can capture a signal that loud without clipping. Some microphones are great with loud sounds, others are not. Here's an example where I had clipping take place even though I was using a 32bit recorder with dual ADC. In this case, the Tatayo's Glassfrog (Hyalinobatrachium tatayoi) was so loud that the microphone clipped the signal before the 32bit recorder captured it. (The microphone used was only capable of recording sounds less than 133 dB - these little frogs are LOUD!) And on top of that, the M3 isn't a real 32bit recorder. The clipping gives the peep a "buzzy" quality that isn't present when heard in person. Notice that the more distant Glassfrogs peeps are not "buzzy", because they were far enough away not to clip the microphone.The only solution here would have been to get further away from the signal source (frog). But in that case, the stream noise would have been increased in the recording? So I needed to be close to the frog to isolate it, but the frog was so loud it clipped anyway. Some microphones have "pad switches" that allow you to decrease the sensitivity of the microphone to prevent this, but I didn't use one in this case unfortunately.
Mid-Side recording - mid-side recording is a way of capturing a stereo field of sound source by using a directional (shotgun/cardioid microphone) in combination with a figure-of-8 (bidirectional) microphone. The beauty of this system is it allows the recordist to change the relative amount of directional (subject) and ambient signal in the recording after recording. By increasing the gain of the directional channel (cardioid/shotgun) microphone, you enhance the sound of the soloist/subject. By increasing the gain of the bidirectional microphone, you increase the perceived stereo width of the field. And you can do this in multiple ways to make different types of recordings after you get home from the field. I will post more on this interesting technique later.
