Four Field Recorders v Soft Sound

TL;DR

My primary interest in field recording is capturing soft nature soundscapes. So the microphone sensitivity and noise characteristics of the recording equipment play a significant role in the outcome. There are plenty of online resources providing equivalent input noise numbers (EIN) and recorders’ and mic combos’ numerical comparisons. But how different does a premium low noise recorder connected to a good pair of mics sound vs. a hundred dollar unit with its built-in mic capsules? I devised a controlled experiment and simultaneously recorded the sound of dripping water using four rigs: Zoom H1n, Zoom H5, Sony PCM-D100, and MixPre 6 II + Sennheiser MKH 8040. Below are the four samples of the same 30-second time period for you to listen to. Headphones are highly recommended:

Note: the samples above are in 128kbps AAC lossy format. There is a link to download lossless versions of the same samples at the end of the article.

If you are interested in the details, please keep reading.

Contenders

Zoom H1n

It is a popular first device for many starting in field recording. It is small and light. Buttons, switches, and the gain knob are on the smaller side, making it somewhat fiddly to operate, but the interface is straightforward. The screen could have been larger too.

I do not have much experience with the Zoom H1n. I got it for a future project while it was on sale for under $80 (and that also explains why I ended up with the white version). The Internet consensus seems to be that H1n has equivalent input noise (EIN) of -112dBu. Having said that, if you own one, I recommend watching Julian Krause's exploration called "Zoom H1n Optimal Gain Level" on his YouTube channel, because this little device’s behavior is not as simple as one might think.

Zoom H5

This is the first dedicated recorder I ever purchased. I still like and use my Zoom H5. The build quality is pretty good; controls and interface are logical. Notice those metal bars over the gain dials, they do work protecting against bumping the knobs and changing gain unintentionally.

Zoom H5 can record up to four tracks in two pairs of stereo or separately using the interchangeable mic capsules plus two mics connected via the XLR inputs. It is an excellent unit for a beginner eager to experiment and learn various stereo arrangements and recording techniques. The Zoom H5's EIN is -121dBu, which is suitable for the majority of recording situations. But then, when I attempted to record birds' dawn chorus, its noise became quite apparent and distracting.

Sony PCM-D100

The D100 is my favorite and most used recorder. Its build quality is excellent; batteries last a long time; built-in mics are super sensitive, with the pre-amps introducing respectably low -127dBu of noise. Because it is quite popular with hobbyists and professionals, there are many in-depth reviews available online, so I would rather avoid going over the points here. As for downsides, I can think of the following three:

High price
Lack of XLR inputs
Placements of switches, indicator LEDs, and ports too close to the mic capsules

The last point results in difficulties when using windshields as some controls become either completely obstructed or hard to access. And you have to use a windjammer with the D100 because of its susceptibility to picking up even the slightest air movements.

Sound Devices MixPre 6 II + Sennheiser MKH 8040

Sound Devices MixPre 6 II and Sennheiser MKH 8040's in X/Y Arrangement

MixPre 6 II is my latest addition. I have not had a chance to put it through the paces just yet. But based on the few experiments I’ve run so far, I am very impressed with this recorder. The same is true for a pair of Sennheiser MKH 8040 Cardioids, which I plan to use in the ORTF configuration enclosed within Rycote stereo windshield and connected to the MixPre. For this experiment, however, I put the matched pair of MKH 8040’s in X/Y arrangement.

Setup

I chose dripping water as the source because upon impact, water droplets produce low levels of sound pressure, and it is non-constant, making it easier also to hear the room and pre-amps introduced noise. Three dripping stations were arranged at 10, 12, and 2 o'clock positions with the recorders in the middle of an imaginary clock dial. I kept the distances between mics and the dripping stations at about 4 feet. The whole setup was constructed in the quietest room of the house with carpeted floors and surrounded by sound blankets (I used moving blankets for that) on three sides, as pictured.

The three recorders and the Sennheiser mics were placed close together in a two by two grid:

Four recorders assembled in a two by two grid: Sony PCM-D100, Sennheiser MKH 8040’s, Zoom H1n and Zoom H5

Recording

Before the actual recording session, gain was adjusted by playing white noise through a portable speaker. I targeted the levels on all four devices to -12dB, using the recorders' own meters. That was the point of my first discovery. Zoom H5 with its standard X/Y stereo mic capsule appeared to be the least sensitive. It required the maximum level of gain dialed in: 10 out of 10. Zoom H1n came in second at 8 (out of 10). I did not expect H1n to place ahead of the H5. Sony PCM-D100 ended up registering the -12dB at just below 6 out of 10. And MixPre 6 II with MKH 8040's ended up set to about 1:30 position. All four devices recorded 24 bit / 48 kHz WAVE.

The session took place at 4 AM on a Saturday to ensure the lowest possible level of sound pollution coming from the outside. Additionally, I turned off the furnace, fridge and other noisy appliances, including the lights in that room.

Post Processing

Setting gains via the recorders' meters produced about a 7dB discrepancy in the recorded levels. I could have used loudness as the measurement to bring them in line but decided against it. The considerable presence of noise energy would have equalized the overall loudness, while I was interested in making "useful" water droplets signal the same levels. So instead, I used the readings of true peak values. Even then, I had to apply the high pass filter at 150Hz before touching up recorded gains. That was due to the Zooms having enough low-end noise to affect the true peak measurements by almost 2dB. All four samples were treated the same way, and the gain of each recording ended up being exactly -30dB. I also added fade-ins and fade-outs to the first and last seconds.

Results

To my ear, MixPre 6 II + Sennheisers is the clear (pun intended) winner, with Sony PCM-D100 taking second place and sounding the best out of the handheld devices. Both Zooms are struggling, and their pre-amps’ noise is noticeably more distracting. The spectrograms below supports the findings.

Side note… You can clearly hear a short electronic beep (with base frequency of 480Hz) occurring roughly every four seconds on the MixPre's sample. It turned out to be my uninterruptible power supply "chiming in" from about 10 feet away. Honestly, I did not even hear the beeps while recording and discovered them only while auditioning the samples. My first reaction was to turn it off and re-run the experiment clean. But then I realized that this beep is not audible on the Zooms' samples. It effectively got drowned in the Zooms' noise. Again, the spectrograms below give visual illustration to this auditory observation:

Conclusions

In the context of this experiment, I find the results from MixPre 6 II with Sennheiser MKH 8040 to be exceptional. Sony PCM-D100 held its own, and I will continue using it with confidence, even for the quietest of situations.

Zoom H5, as nice of a device as it is, will be limited to recording in louder environments.

I will use Zoom H1n in my upcoming project, then it will be kept around as a backup only.

Lossless Versions Of The Samples

Click to download

4 Field Recorders v Soft Sound