Microphone/Audio Recorder Tests

mic tests

Mic shootout comparing the Sennheiser 8040s in ORTF vs. the Schoeps MK4-MK8 MS configuration.

Louis Cyr - Sennheiser MKH-8040 ORTF

city night rumble with train horns close to shore of Mississippi River.  Distant church bells.

NEW ORLEANS NIGHT ST B SCHOEPS MS 01

NEW ORLEANS NIGHT ST C DPA BIN 01

NEW ORLEANS NIGHT ST A MKH8040 ORTF 01

Another test of the BP4025 vs Primo EM172 (FEL Clippy XLR). This time in a more real-world setting. 

I have noticed how bright the EM172s are in comparison to the BP4025 (and AT4022 which I tested earlier). This led me to think they can stand a little EQ adjustment to try to reduce the noise, which is naturally higher than the BP4025.

1. Primo EM172 stereo pair
2. BP4025
3. Primo EM172 with EQ applied (-4db reduction in the 5-15khz range)

1 and 2 were recorded simultaneously to a Zoom F6 and later normalised. 3 is a copy of 1 with EQ applied.

Conclusion 1: BP4025 is naturally more quiet than the EM172 but has a more narrow stereo image. However, applying some EQ to the EM172 brings the noise level even lower, and without any negative impact to the overall sound. To my ears, it sounds even more natural now. 

Conclusion 2: I didn't need to buy the BP4025 to achieve lower noise recordings. Damn!

EM172 VS BP4025 VS EM172 With EQ

2 x Clippy microphones in a cupboard recording to an Olympus LS-P4. Same source recorded at 7 different gain settings (1, 5, 10, 15, 20, 25, 30) and normalised in post (adding 70db to the lowest gain setting, 5db to the highest). Recorded at 24/48.

This shows a very clear linear progression in the reduction of noise. I often record at level 15 in quiet ambiences but this shows there's something to be gained (in terms of HF noise reduction) by cranking it up to 20 so long as there's no likely to be any loud sound sources that will clip the recorder.

Olympus LS - P4 Self - Noise Test

Test of the baffled omni array I have been working on with two Primo EM172 capsules (FEL Clippy). Microphones were pointing out from a first floor window up towards the sky (at some birds clattering about in the gutters).

Baffled Omni Array Test

This test is comparing the self-noise of the Sony PCM-A10 and the Olympus LS-P4 at various gain settings using external microphones (Clippy EM172). Both recorders allow gain settings from 1-30. I recorded the same sound source (quiet ticking clock) at levels 1, 5, 10, 15, 20, 25 and then 30 simultaneously on both recorders and then normalised the levels in Reaper. I then had to do quite a bit of manual correction of the levels because there were some low frequency sounds that were louder than the ticking clock and which were messing with the normalisation process. Recordings were made at 24/48 WAV.

In this recording you first hear the Sony at gain level 1, followed by the Olympus at gain level 1 and so on up to level 30.

As shown in an earlier test the Olympus exhibits much higher levels of self noise at lower gain settings, with a noticeable improvement coming at level 20. This means that when recording quiet scenes, it is worth taking the risk of setting the gain high to minimise the recorder self noise.

The Sony behaves quite differently. At gain level 1 the noise is quite terrible  but at level 5 there is a sudden improvement and in fact there is almost no difference between the noise at level 5 and level 30. This seems to suggest that you can be much more conservative with setting the gain on the Sony to avoid clipping, and increase the levels later in post with no apparent self-noise penalty.

Sony PCM - A10 VS Olympus LS - P4

Another test of the Sony PCM-A10 at different gain settings using external microphones (Clippy EM172 stereo pair). Recordings of two different subjects at low and high gain to compare the recorder self-noise. Tracks normalised in post afterwards. Recorded at 24/48.

Clip 1: Indoors with windows shut. Birds audible through the windows. Recorded at input level 5.
Clip 2: Same as above recorded at input level 25
Clip 3: Outdoor ambience recorded at input level 5.
Clip 4: Same as above at input level 20 (highest I could go without the sparrows clipping the recording!) 

A gain setting of 5 on the PCM-A10 is very, very low in case it isn't clear. There was virtually nothing registering on the meters while I was recording.

I hear zero difference in noise for the recordings at very low and very high gain settings. This is incredible as it means you can set your levels very conservatively to avoid clipping and increase levels in post later with no fear of losing the signal in the noise.

Sony PCM - A10 Gain Test

More tests of the Sony PCM-A10 at extremely low gain levels. I recorded a few seconds of test tone followed by 20+ seconds of a quiet room at gain levels 1, 2, 3, 4 and 5, and then level 25 as a high gain level for comparison (highest gain should have the best noise performance).  The recordings were then normalised in Reaper, resulting in gain being added to each clip of: +40.8dB, +39.1dB, +37.7dB, +35.7dB, +34.6dB and finally +2.87dB (gain level 25).   

Hopefully what you can hear in this is that levels 1 and 2 have a high level of noise, but then it suddenly drops off at level 3, and levels 3, 4, 5 and 25 all have nearly identical levels of audible noise.

My conclusion is that when recording a dynamic source such as a thunderstorm, I can safely go all the way down to gain level 3 to minimise the chance of clipping, while not adding significant noise when I later raise the levels in post.

Sony PCM - A10 Gain Test (Extremely Low Levels)

I'm always looking to shrink the size and weight of my rig. I just love carrying the minimum amount of equipment!  I did this test to see how small I could go with microphone separation without sacrificing quality. All configurations were Clippy EM172 stereo pair to Sony PCM-A10.

My conclusion is that AB40cm rig with simple Rycote lav wind protection gives me a great sound with a very small and light footprint but not great wind protection. The quasi-binaural rig gives perhaps a more accurate stereo field with the advantage of excellent wind protection (it is housed in the equivalent of a blimp with proper windjammer). To get better wind protection for the AB40cm rig would require a massive and heavy (by my standards) blimp.

1. AB20cm - narrow, too narrow...
2. AB30cm - getting wider, pretty nice...
3. AB40cm - wider still but nice balance of width and centre-image?
4. AB60cm - very wide but hole in the middle?
5. Quasi-binaural (20cm spacing with foam cylinder baffles) - improved placement of sounds in the stereo field?

Would be very interested to hear what other folks think. Please comment if you care to.

Omni Spacing Test

Stereo soundscape recordings. Sometime nature, sometimes people, sometimes machinery. Sometimes quiet things like a single branch creaking in the wind on Dartmoor, sometimes odd moments in the city. Mostly local, but sometimes travelling. 

Omni Spacing Test

Douglas Langley