The setup - three ORTF stereo pairs.
Intent
We’re working on expanding our immersive audio recording capabilities, particularly in the realm of immersive music. One of the ways we’re looking to expand is into spaced microphone arrays. Everything Blackguard Sound has commercially released up to this point has been recorded with coincident arrays (1st or 2nd Order Ambisonics). There are many benefits to Ambisonics, but there are some tradeoffs, too. We’re primarily concerned with the size of the sweet spot, which can be quite small. Outside of that sweet spot phasing becomes apparent and it can feel a little dizzying. We’re working on postproduction methods to decrease this effect, and the results are encouraging. But one of the benefits of spaced arrays is a more stable, enlarged sweet spot. That said, we will most likely continue to use Ambisonics mics for field recordings due to the obvious logistical challenges of bringing spaced arrays into the field.
We’ve experimented in the past with quad spaced omni (with meh results) and a proprietary technique devised in conjunction with Darren Schneider at Full Sail University. This novel technique places the performer(s) in the center of 11 microphones at the Dolby recommended Atmos speaker placements, pointed inwards. Then in post, each mic is converted into a Dolby Atmos object and panned to its placement in the room. The results of this experiment were more encouraging. The immersive image is very high fidelity and interesting, but perhaps hyper-real. It almost places the listener inside the instrument or inside the performer’s perspective. We thought it was pretty cool, but the feedback from classical music aficionados was almost exclusively ‘I am not used to that perspective, and I don’t think I can get used to it’. So, our research into spaced arrays continues.
The Full Sail immersive mic technique in action
One spaced array in particular has sparked our curiosity. The PCMA-3D (Perspective Control Mic Array) by Dr. Hyunkook Lee. It’s influenced by existing classical music recording techniques but enhanced to be compatible with immersive audio. This technique requires eleven mics (a mix of cardioids and super cardioids) arranged in a one-meter square with seven mics on the horizontal plane and four aimed upwards (coincident with the four corners of the square). Phase one of our spaced array journey will be testing. Phase two will be purchase and adoption. Buying eleven high-end microphones is a substantial investment, so we will be renting mics for our initial experiments. But even renting is an expensive proposition. - it will cost about $1,700 to rent eleven high quality mics for a week! So, with that in mind we devised this microphone shootout between the three front-runners in the hopes of narrowing down our options.
The Three Options
Schoeps MK 4 / MK 41
These are the mics to beat. We have a stereo pair of MK 41s, and they are our favorite mics for sound effects recording. They have an uncanny ability to capture the space of a room. On headphones, it’s a borderline immersive experience. I have been genuinely startled while editing under headphones when an off-axis sound seemed to come from behind me. Since the goal here is maximum immersion, it’s going to be tough to beat this mic. There are two downsides to these mics, however. The first is expense. These retail for $2,000 each. Multiply that by eleven and we’re looking at the cost of a reasonable new car. The second is their susceptibility to humidity. As much as we love these mics, it’s just not possible to use them outdoors here in Florida (or any other sub-tropical or tropical climate). As stated above, we’ll probably continue using Ambisonics mics in the field, but if we’re going to spend a crazy amount of money on mics, we’d prefer to have the option of using them outside. We used the MK 41 (super cardioid pattern) for this test.
Sennheiser MKH 8040 / MKH 8050
We have tons of experience with Sennheiser mics. The AMBEO is our go-to mic in the field, and we’ve used MKH 50s and MKH 60s recording dialog on countless film sets. The 8040 and 8050 are updated versions of that older generation. The 80 series is meant to be interchangeable with the older mics, but we’ve heard anecdotally that this isn’t entirely the case. However, for the purposes of this test we’re using the older MKH 50 since that’s what we have access to. This approach may prove the be flawed, but I suspect these mics are close enough to help with our decision. The upside of the Sennheisers is that they are slightly more affordable at $1,500 each. Also, they are RF condensers, so their high humidity performance is immaculate. Even if we aren’t utilizing the full array, we can use these mics in the field for a variety of other applications. We used the MKH 50 (super cardioid) for this test.
Neumann KM 184 / KM 185
These are the wild cards. The KM 184 is well known in the music industry but not widely used in sound effects. I have used the Neuman KMR-8 shotgun mic for dialog. Like the Schoeps, it does not deal well with humidity. One time on a short film set, there was a company move from inside to outside. I think the KMR-8 lasted about 20 seconds before it started sputtering from humidity! But the Neumanns are well regarded for studio work, and their price point makes them much more attractive. At $850 each, it’s a much more reasonable proposition to get eleven of these. My main concern for studio work is a presence boost around 4K, where the other mics are reasonably flat. There’s only one way to find out if that boost is going to be an issue! We used the KM 184 (cardioid) for this test. Yes, the other two mics in the test are super cardioid, but we live in an imperfect world.
That’s a bunch of German mics.
Methodology
Placement
I decided to use ORTF placement for the three stereo pairs. This is my personal favorite stereo technique. My experience with the MK 41s is exclusively using them in ORTF configuration and this is how I’ve achieved borderline spooky imaging with them. ORTF requires the mic capsules to be 17cm apart and at a 110-degree angle. I tried to get them as close together as the mounting hardware would permit. I was able to place the three ORTF pairs in a vertical stack approximately two inches apart. I hope that’s close enough to not introduce any major variables. The timing discrepancy between the earliest pair and the latest pair was less than 1ms, so they were very close. To my mind, the more important variable was being able to record the exact same material on each pair simultaneously for the cleanest possible comparison.
Detail of the setup
Amplification
This is the variable I was most concerned with. For this test to work, the gain has to be identical for all six mics. Ideally, the preamps would be digitally controlled. My RME UFX+ only has four digitally controlled mic pres. Our Sound Devices MixPre10 has 8 linkable preamps, but I didn’t want to have to dump files afterward or set it up to work as an interface. My remaining option was to use a Sonosax SX-ES64, which is a high-end production mixer. It has a unique 3-way gain selector (instead of a mic/line switch). I set the switch to the mid setting and kept the gain knobs all the way down. From there, I took six pre-fader outs and plugged them into a Ferrofish A/D converter. It’s possible to link the gains on the Ferrofish, so I was able to ensure that all six mics got the exact same amount of gain.
Sonosax Production Mixer
Linked gains on A/D Converter
Test Material
I decided to use four very different sound sources to get the widest possible variety of volumes and timbres. The four sound sources were: Alto Saxophone, Classical Guitar, Voice, and Percussion. I then recorded each instrument at two different perspectives that I’m calling ‘near’ and ‘mid’. Near was about five feet from the mic array and mid was about 15 feet away. Blackguard’s studio space is not really intended for tracking acoustic instruments. But it’s a treated room, and it’s relatively quiet so it’s not the worst. While I wouldn’t want to use these recordings on an album, I think they’re perfectly fine for the sake of comparison.
Sound sources
Analysis
I recorded both performances of each instrument into Pro Tools. Once I was done recording, I decided to measure the performance in two different ways. The first way is objective analysis, measuring levels (pseudo) scientifically. The second is subjective analysis using my ears to pick a favorite.
Pro Tools timeline
Objective Measurement
I measured each recording using iZotope Insight 2. I was looking at three things: Integrated LUFS, True Peak, and Noise Floor. The integrated LUFS gives me an idea of the mic’s output. The True Peak (when compared to the integrated LUFS) gives me a rough sense of dynamic range. Those two measurements are straightforward. The noise floor measurement is probably flawed, but it is an ‘apples to apples’ real-world comparison. In each recording, I made sure to remain perfectly quiet for about five seconds. I then separated this ‘silence’ into a new clip and measured the LUFS again. However, the LUFS meter doesn’t do great with very quiet recordings, so I raised the clip gain of these ‘silent’ clips by 12dB to get a reliable measurement. The studio isn’t exactly an anechoic chamber but given the same source material and same gain the only variable SHOULD be the mic’s inherent self-noise.
Insight 2 measurement
Detail of noise floor measurement
Objective Results
I bet a real acoustician would eviscerate this analysis, but for my needs I feel I got some good information. For example, the MKH 50 was WAY hotter than the other two. The MK 41 consistently had the lowest output. The KM 184 had the lowest noise floor and the most dynamic range. These things were true across nearly all instruments and perspectives. Surely, there’s value in this intel.
Some objective measurements
Subjective Measurement
I decided to evaluate each mic pair on two subjective aspects: tone and space. Tone being the frequency response of the mic and the resulting attractiveness on each source instrument. Space being how convinced I was by the spatial imaging. This was the biggest benefit of recording in our studio. I know the space well and it was very easy to hear the difference in each pair’s spatial performance. I assigned a rating to each mic for every instrument and perspective: one for my favorite, three for my least favorite. The mic with the lowest score wins the subjective analysis.
Subjective grading
Subjective Results
The MKH 50 was the easy winner in the tone category. The highs are clear and articulated without being harsh. As an added bonus, the MKH50 (and the 8050) extend well into ultrasonic frequencies – not relevant to this test but as a sound designer I’m a fan. The MKH 50 also had the best low-end response of any of the mics. This was especially apparent on the percussion recording and I feel the other two mics really lost something in this regard. The MK 41 and the KM 184 were much closer to each other, tonally. I preferred the MK 41, but only slightly (there was only a two-point difference between them). Given the price difference between these two mics, I was pleasantly surprised by this.
The MK 41 was the clear winner in the space category. It has a nearly magical way of making the space feel authentic. I suspected the MK 41 was special in this regard and the results of this test confirmed my suspicion. The KM 184 was the next best, but I still preferred the MK 41 by a nearly 2:1 margin. The MKH 50 lagged far behind the other two. When compared directly to the Schoeps mics, the Sennheisers practically collapsed to a phantom center that feels like a single point as opposed to really feeling the width and depth of the room.
So, there’s a bit of a problem here. The MKH 50 was my favorite from a tone perspective, but my least favorite from a space perspective! What were the final combined scores? Maybe that will help…The KM 184 finished last with 35 points. The MKH 50 finished second with 32 points. The MK 41 finished first with 27 points. OK – so we have a winner but it’s not a blowout.
Final results - overall, tone, and space - low score wins
Synthesis and Key Takeaways
The MKH 50 was the outlier – it consistently felt like a totally different beast on every source. If this decision was based purely on tonal quality the MKH 50 would have run away with the victory. However, the whole point of this exercise is to evaluate these mics for use in an immersive array and the MKH 50 lagged far behind the others in this regard. Despite all the other things the Sennheiser mics have going for them, I think they need to be excluded from consideration. Perhaps this is unfair, and we need to acquire a pair of the newer 8050s for truly fair testing…
The Schoeps MK 41 and the Neumann KM 184 felt much closer to each other. The KM 184 was just a little harsher (that presence boost rearing its head) and a little narrower. But I need to stress ‘a little’. I had to really dig in and A/B a lot to come to this conclusion. For me, this was a surprise for a couple of reasons. First, the difference in polar patterns. I was concerned about using the cardioid KM 184 for this experiment since the other two mics are super cardioid. But it turned out to be less of a factor than I anticipated. I’d really like to get ahold of a pair of Neumann KM 185s, which is the super cardioid version, and compare it against the MK 41. I wonder if that mic would close the gap in the space category…
The technical specs were consistently quite good for the KM 184. Lowest noise floor. Highest dynamic range. Middle of the road output. Least expensive by a lot. Truly impressive.
This leaves us with a tough decision. The MK 41 is my clear favorite. But is it $1,100 better on a per-microphone basis? Definitely not. Let’s do some quick math. A $1,100 price difference per microphone times eleven total microphones in the array comes to a $12,000 difference! That’s going to be hard to stomach. There’s an argument to be made that quality and results are all that matter. I tend to agree with this sentiment, and I’ve never been one to shy away from spending up for professional tools. However, the difference between these two mics is not that wide. In the short term, I think we need to rent a pair of the Neuman KM 185s and do another shootout 1v1 with the MK 41s. If the gap is closed even further this decision will get a lot easier. For now, my intuition says the KM 184/ KM 185 combo is the right choice but stay tuned to see what we ultimately decide.