View Points: Objective, Subjective & Semantic
Varying takes on microphone evaluation and jargon

By Jack Alexander

Prof. Alexander getting right into a Shure SM57.

There is a famous 1950’s singer currently enjoying a renaissance with young listeners and the reverence of some contemporary artists. When I see him on television, memories flood back of the four-hour sound checks (two KM84 microphones for him, an acoustic bass DI, two C tapes hidden in the piano, and a very hidden mic in the kick) and all of his grief directed at the audio crew and anyone else unfortunate enough to be within earshot: “I’m walking on eggs. This thing sounds like cotton wool. You got a limiter on that? I want a system that works with my mic!” You get the idea.

But he sure knew how to play the presence peak (or proximity effect) of a vocal mic, which leads us into the following discussion. If you’ve been working in professional audio for more than a few months, you may have noticed that as your mouth gets closer to a vocal mic, right up to the point where you are touching it, the response thickens up - it gets heaver in the lower midrange. The history and technical analysis of the proximity effect is more a Professor Jones’s thing, but I am more concerned with how it affects our work in the field.

The evaluation subjects lined up and ready to go.

When informed of my intent to do a little research on the presence peaks of some vocal mics, Doug positively shook with anticipation, figuring that he finally had me on his (measurement) turf. Sorry to disappoint, but we didn’t measure anything during this little exercise that follows, though we fell into a few bits of useful knowledge regarding the various frequency response/distance to orifice gradients of certain vocal mics in a regenerative (mic open into loud monitor, right?) environment.

Technical assistance was provided by the Live Sound Practicum students at the college, many of whom are already working professionals and can be a tough audience. Live Sound Editor Keith Clark also attended, and I can just imagine his take on the proceedings, even-tempered soul that he is.

Sennheiser E 865 results.

There is too much nonsense written in support of performance claims of all audio equipment that often misses the real reasons anyone uses the stuff. After all, if response ruled, then we’d put U87’s (flat response) on everything, and things would be perfect.

But there are unmeasureable criteria (or at least up to now, not subject to measurement) that is applied by working professionals to equipment choices. We may not verbally delineate presence peak variations from mic to mic, but in any high sound pressure level (SPL) monitor work, we are spending a fair amount of time organizing the relationship between the mic, the monitor, and the singer’s approach to mic use.

EVALUATION CRITERIA

I thought we should evaluate at least three distances from mouth to mic - lip contact (show), two inches (sound check), and four inches (aging pop singer messing with engineer). Anyone doing monitor work faces all of these distances and more. We wanted to see what happened to the audible (not measured) response of the mic at these distances.

The presence peak evaluation of each mic in this experiment was logged into XTA Audiocore software. Here’s the result on the Shure SM58.

Signal chain was a Midas Venice mixer, Klark Teknik equalizer, XTA DP224 loudspeaker management system and a Lab Gruppen 6400fp amplifier into two Martin LE12J wedges. (That Lab Gruppen is a very powerful amp, friends. We wanted to remove any amp nonsense from the equation, so we could really hear what was happening.) Though the Klark Teknik was in-line, we only used it at the beginning to get a feel of the situation - for the actual evaluation, EQ was applied only with the XTA.

We found two of our monitors (trying at first to use the ones we’d just painted, so Keith wouldn’t think we were slobs, but they were sub-par sonically relative to the ones with scratches and dented grilles) that more or less matched (I can hear the drumbeat from Doug now), and I threw down a crossover point. Next I yanked up a conventional Shure (not Beta) SM57 with a screen (use the one from Shure, only!) and got the system up to something like a production level. There was some noise from the corner where some joker was measuring the SPL introduced into the space by this mess. I tried to remember the difference between “A” weighted and “C” weighted and gave up and went back to screaming.

Data of Shure SM57 with windscreen.

There are some delay adjustments that really help this particular wedge, and we tried things untweaked and delayed, and the difference was not subtle; thus, those modified values were in place for the whole event. Now that we had a totally unscientific situation, it was time to figure out what was going on with the mics.

We first found that one could forget about the three distances, at least in our relatively small room. The mics dictated the distances, actually. I would start right on the mic, and then back out to where it sounded relatively flat - anywhere from one to three inches. Then we would call the EQ adjustments using the XTA parametric addressed from a laptop (good luck doing this off the front panel) until everyone agreed that the audible response - both distant and touching - was close. The adjustments were saved in XTA’s Audiocore software, and then we’d move on to the next mic.

The take on the Audix OM5.

This was all done with some speed (just like getting a show up, right?), and I had not cleaned up the Audiocore files, so the center frequencies appeared out of sequence to a reader with an organized mind. But actually they are in sequence - the sequence where what was worst got fixed first. The goal was to check out the presence peak distance gradients, not optimize the mic at all frequencies.

To do this (yeah, I checked), nothing was touched above 680 Hz with any of the mics, though some of the “Q’s” used in the XTA were wide, and we reached higher in some cases due to the wide “Q” applied at 667 Hz (Shure SM57/screen) or 680 Hz (Audix OM5). Remember, I did not plan 680 Hz as the limit of the adjustment activity - this happened in a dynamic environment where a mutually agreed subjective goal drove the situation.

EVERYONE THOUGHT THEY WERE HAPPY...

So what’s the point, eh? Let me begin with a little war story that should clarify the need for all this. A well-known Detroit jazz/R&B act played a Chicago venue quite often, and were very much a known quantity to the crew. Eight open vocals, fairly icky attitude toward us, but manageable - good players too. We worked with them frequently.

Jack admits to liking the Beyer 88, but found this data to be entertaining.

One day they turned up with their new vocal mics, and (finally), their own monitor engineer who was a nice kid that didn’t have a clue about live audio. He had correctly identified a mic totally appropriate to the house sound of the act without thinking about monitor consequences. I had a generic tweak on the wedges, and was barely able to contain my laughter when this kid yanked out eight Beyer M500s - a wonderful sounding ribbon mic with absolutely no presence/proximity peak.

Everyone in and with the band was happy that they had their own sound guy and cool mics, and the kid proceeded to yank the vocals up in the monitor rig. Mr. Bandleader starts to squirm at his Rhodes keyboard, and asks for more. The kid bumps it, and even from the desk I can hear that 400 to 800 Hz “hollowness” associated with serious monitor bummer.

Everyone gets ticked off, they try to blame the system and the venue, and finally I gently indicate that there is something that they are used to that isn’t there, and bump all the one-third octave graphic EQs between 100 and 400 Hz (awful monitor practice, ugh). This introduces something like a Shure SM58 peak, which allows the kid to back the thing down into stability, with a peak/chest thump in place that is indeed what the act requires.

Shure Beta 58A result. Interesting to compare with those of the SM58.

They require it because, A: they are used to it; B: their hearing losses are the reciprocal of it; and C: most of the monitor rigs on the planet Earth account for it or else. It doesn’t matter whether it is right or not - yes friends, the Beyer M500 sounds better than a SM58 live, until you try to deal with artist reality, at which point you end up having to introduce the SM58’s proximity anomalies into the sound of the Beyer if you’re thinking of retaining your employment.

And when live operators discuss vocal mics, whether they realize it or not, a lot of the discussion relates to the variances in the proximity effect found from mic to mic. A conventional SM57 with a screen has a lower, narrower, and stronger peak than a conventional SM58, where the peak is higher, wider, and shallower. And the Beta 58 peak is even higher and shallower. The Beyer 88 peak (the only mic I own is an 88, for the record) is a hoot - in our evaluation, we had to cut the thing minus 8.6 dB with a Q of 1.9 at 118 Hz to get it to flatten, with another cut of minus 4.2 dB at 60.1 Hz with a Q of 2.1 also necessary. (Remember those cuts are based on the sound of the mic with lip contact).

In our evaluation, the Sennheiser 865 had the least audible variance over the distance gradient of any of the mics, yet I know that its stablemate, the 835, had a big old honker of a peak reminiscent of an 88 crossed with an old SM57/screen combination. The Audix OM5, noted for high gain before feedback in “leather jacket” situations, thickened at generally higher and different frequencies with proximity than the others. It almost looks like someone modeled the response of the competing mics and made sure that this item sounded different.

TOO MANY VARIABLES

It is difficult to generalize about which peak characteristic supports which production situation because there are too many variables. Once, I worked with a very rich maniac with a very big venue and loads of cool loudspeakers and BSS Omnidrives who insisted on using a Beyer M500 wireless. He consistently operated two inches off the Beyer (you don’t want to know how much time I spent on stage with this joker) and demanded a uniform rise from 80 to 400 Hz at that distance, with a mic that was flatter than death at all distances.

There was 20 kW of monitors, including floors, flying sides, sides, butt mixes (“shakers”), and reverb and delay in everything. We’d get it right, and then he’d drop the mic or it would get a little wet, as they do, and I’d have to fly back and do the whole mess over.

If you don’t have that kind of cash, though, it’s a lot easier to fit the mic to the situation, rather than the other way. Run some of these little tests yourselves, and find out what works for your equipment, clientele, and venue. And before you try to do this by looking at cut (data) sheets, remember that the outcome is in some degree determined by your loudspeakers, amps, and crossovers, which are not factored in when manufacturers are doing data measurement.

Live production is so impure and non-technical, just like those damned peaks.

Editor’s Note: Prof. Alexander was compelled to add the following after viewing Prof. Jones’ comments.

As we really don’t like things out live with a rising high end (feedback - you ain’t got that in the studio, do you?) we have a tendency to foul up the terminology regarding proximity effect and presence peak. Of course it isn’t the same thing, but we often use the words interchangeably because:

A) We’re slobs. If we were anal we’d be recording engineers and starve, or be consultants and suffer from massive boredom from dealing with mojos in suits.

B) The word “presence” to experienced (O.K., old) live engineers, especially those like moi who were trained by the spawn of the British Commonwealth and most especially those who do monitors, relates specifically to energy in a male vocal from 80 Hz or so through around 315Hz. Mics that are excessive or rising in the region from 2 kHz to 8 kHz are viewed with disdain live. The systems are heavy there as a rule and we don’t need any more of that stuff from the mics, therefore the word “presence” carries no meaning to live engineers when they consider those particular frequencies. More “presence” in the room to a live engineer in this context means that we can hear the lower parts of the male vocal distinct from B/DI and snare/low.

C. If you’re interested in the correct “live” terminology for what the wait staff (couldn’t resist, sorry) in the studio call “presence peak”, it is called “cut”.

So if I correctly combine “live” jargon with the language of science, then we end up with “cut peak”. I think I’ll stick to the old way.

Jack Alexander instructs on topics allied to performance audio at Columbia College in Chicago. Reach him at jalexander@livesoundint.com

March 2003 Live Sound International