Many people say a mic spec sheet is useless, it can't tell you how a mic will sound, whether good or bad, on any particular source or any source at all. While that is true, that's not what the spec sheet is for. It's more to save us time in not bothering to try out mics that are completely inappropriate for our intended source and the rest of our gear.

As I imagine you might know, mics come in two basic flavors (with some unusual others I'll ignore for this article): dynamic and condenser. Ribbons are dynamic. Some strange Europeans call condenser microphones "capacitor microphones". (then again they call apartments "flats" and car trunks "boots".) Condenser is an old word for capacitor, so take your pick. And there! That's the first spec: microphone type; sometimes called operating principle.

Within that same spec, you might also see a reference to polar pattern, perhaps in an oblique way: "pressure" or "pressure gradient".  In simple words, a pure pressure mic is omnidirectional, and a pure pressure gradient mic is bidirectional, or figure-8.

Also, you might learn some more information about the type of mic. For example, condenser mics are further subdivided into two categories, electret and externally biased (or polarized). The latter are sometimes called "true" condenser mics, but an electret mic capsule is just as much a capacitor, so I think that's prejudiced!  OK, I'm "biased" too... (sorry for the pun!) And, finally sometimes the diameter of the capsule diaphragm is stated. "Small" and "large" are the usual descriptors, but sometimes the'll be kind enough to specify the actual diameter in millimeters or inches.

So what have we learned about this mic so far? Whether it's a dynamic or condenser style mic, it's pressure gradient or polar pattern and it's capsule size. What does that all mean to us?

well, dynamic mics and condenser mics have very different applications. Broadly generalizing (with many exceptions), condenser mics tend to have better transient and high-frequency response. They also tend to be much more sensitive (We'll talk more about sensitivity when it appears further down the spec sheet.).

Diaphragm size can affect those specs as well, though there is some disagreement as to how much.

We also know one important item: condenser mics need power. This power is usually supplied via "phantom power" supplied by the preamp, although an external power supply or an internal battery are two other possibilities. Dynamic mics do not need phantom power, except for some recent ribbon designs that have an internal preamp. Furthermore, the voltage required often differs by type; externally biased condensers often want a full 48V, but electrets are frequently happy with anything from 12V to 48V. How much power does our mic want, though?  Usually for condenser mics you'll see a spec for current, in mA, instead of volts. The maximum possible under the phantom power standard is 10mA, but most mics use far less. 2mA to 5mA is typical. If you see much more than that on a mic spec sheet, be a little careful. Not all preamps can provide the full 10mA defined in the phantom power standard, and as we shall see in a couple of weeks, preamp manufacturers are very bad about admitting that.

Next up is polar pattern. I have already mentioned the two extremes; all other polar patterns are in between: subcardioid, cardioid, hypercardioid, and even more in between patterns. Different polar patterns are generally selected for their pickup and/or rejection of sources at various angles, or more artistically, for the characteristic frequency response they offer. Usually along with the pattern, you'll get a polar response graph, at least at 1kHz, but often at a range of different frequencies.

You can see a lot from this latter type; if a mic has a consistent off-axis response at a range of frequencies, it might be a better choice for use in a stereo pair than a mic that has all kinds of crazy things going on at diffeent frequencies. That one you'd probably want to save for on-axis use. Just keep in mind that, because of the nature of sound itself, all polar patterns are somewhat omni at low frequencies, and somewhat directional at high frequencies.

Naturally, next on our spec sheet is the frequency response. Don't flip out over the hard numbers you see here. You don't necessarily want every mic you use to have 20Hz-20kHz response.  That, by itself, doesn't make a mic "better". Microphones aren't studio monitors! What does that numeric range mean, anyway? Most manufacturers will list the "corner" frequencies of the mic's response at the extremes; that is, the 3dB or maybe 6dB down points at the high and low end. So a mic like the Shure SM81, for example, is listed as 20Hz-20kHz, whereas the SM58 is 50Hz-15kHz. Does that mean the SM58 has no response at all below 50Hz?  Such that if you use it on bass cab, the low E will magically disappear?  No, of course not.  It's a smooth downward falling curve, so maybe -8dB at 40Hz. (Also, there is proximity effect, which causes directional mics like the SM58 (and SM81) to have increased bass response when the source is within a fcouple of inches. Better to look at the actual frequency response curve graph.

So next we look at the frequency response curve. Some will say these are notoriously inaccurate. From a measurement microphone perspective, that is probably true. If you take any one example of a microphone, you might get a fair amount of deviation from the published curve, which is typically an average of 10 or more units. Put simply, no two microphones, even of the same make and model, will have identical curves to each other, let alone ientical to the one printed on the spec sheet. That graph is not intended to represent precision; it's main puropse is to represent overall character.

So instead of reading every nook and cranny of the wiggle on that graph, instead look for broad patterns. How fast does the high end rolloff? Are there peaks in the presence range for a tailored response, or is this a relatively flat-response mic? Or is the thing just really dark? Unfortunately, the manufacturer won't tell you if those presence peaks sound like audio crack cocaine or nails on a chalkboard. But it might keep you from buying a mic for a totally inappropriate purpose.

Next time: math and mics.

The Business of Music, Part 1: Show Me The Money!

CAPITAL INVESTMENT AND PAYBACK PERIOD
Even the most modest of home studios requires significant capital investment. Thousands of dollars Will be invested in hardware, software and acoustics by the time you have even the minimum requirements to qualify as a competitive independent home studio in a market infamous for having clients with very little spending money. Until the studio's net revenues exceed the costs in capital investment, one cannot even begin to think about making money.

Therefore, many people look at a capital investment in terms of payback period. It's easy to calculate: capital investment divided by net profits equals payback period. Let's say you can spend $10K in gear and treatment, and bill 1,000 hours a year (in your spare time) at $25 an hour. That's $25k a year in gross revenue, or just a bit over $2000 per month in gross revenue. That's a payback period of 5 months. Hooray! It's all free money after that, man!

Or is it?

I mean, if it's that good, why does anybody ever open a Mickey D's?

Payback period is a somewhat useful concept, but it's actually excessively optimistic and pessimistic at the same time. You'd think those extremes would cancel out and yield a useful figure. Sometimes they do. But knowing when they do often requires more experience than the average startup business owner has to muster. The actual numbers get much more complicated than that.

What if you hired an engineer to run your sessions and you just pocketed the profits?  Now what do you get net, maybe $5-10 per hour? That's a one to two year payback period instead of five months.

But we're not done yet.

RESIDUAL VALUE
Where payback period calculations fall on the pessimistic side is that they ignore residual value. Let's say you operate for two years, and hire an engineer at $20 per hour. That's the break even point; that makes your net profits zero, right? Wrong. You still have $10K of gear. It might not be worth $10K anymore, let's say depreciation has brought the value down to $6K. So you sell that off and your net profit is $6K.

So while you have only broken even on paper, the assets you purchased still have residual value. So, now your $6k ahead of the game because that is the value of the assets in which you originally invested your capital.

Almost.

TIME IS MONEY
Where simple payback period can be a bit too much on the optimistic side, the giant misleading concept of payback period, is the lack of this fundamental principle of finance: money has time value. $6K in two years isn't worth as much as $6K today. You'll probably think of inflation as the culprit here, and that's a big part of it, but inflation is only part of the scene.

Every investment has an opportunity cost, which is your cost of capital, or your marginal interest rate for borrowing (or investing). Marginal means the rate you'd pay off first, or the rate you'd get if you borrowed more money. The more time it takes to pay that cost of capital, the higher that cost rises. So even if you have $10K in cash laying around, there's still a capital cost associated with spending it on gear rather than paying off the balance of your loan or mortgage.

That incredible vintage hardware you found on eBay for only $1000 would add another two weeks to the original 5 month payback period.. That new microphone? Another week. The hundreds or thousands of blank CD-Rs, DAT tapes, etc. Add a week for every $500 spent. And that's just for the principal capital cost. The extra costs in finance charges, depreciation, etc. that are incurred because of the extra time required for the payback extends the payback period just that much further.

But that's still not right.

I AM SOMEBODY
If you are evaluating your business plan (you have one of those, right?), you need to consider both residual value of assets and time value of money to get it right.

You need to also value your own labor. Here's why: One thing you are trying to do in a financial model is evaluate the profitability of a capital investment. The rate at which you can get away with billing your clients is largely determined by two factors: your level of engineering skills, and your facility/gear. (OK, really it's who you know, but this is a finance lesson, not marketing!). You don't improve your engineering skills by buying gear, we all know that. So if you want to know if buying a particular piece of gear is worthwhile, you must consider how it will affect your hours booked and hourly rate, which drop to the bottom line. But in terms of return on capital, you need to remove the value of your labor to see whether or not it's worth charging it on your Musician's Friend credit card at 21% interest.

I didn't include interest expense in my spreadsheet model (see below), but feel free to change the marginal interest rate to 21% to give you a pretty good idea of what such expenses would be like. If the present value of net cash flow drops below zero, that basically means you have lost money on your capital investment, even if you've made some scratch from your salary.

CRUNCHING THE NUMBERS
I have with this article included for free download a simple Excel spreadsheet (click the image to download.) that makes some of these basic calculations. Just click on the spreadsheet image to download.

I've added a few more assumptions in there; hours booked and rate grows as you reputation spreads. By Year 5, you're ready to go full time. Depreciation is 10% per year straight line, which is certainly not a tax measure, that's supposed to be actual reduction in residual value. So if you cash out after Year 5, you'd have the residual value too. But I have stubbornly stuck to including your salary as an expense.

Feel free to play with the numbers and see how they affect the actual financial calculations, and the importance of each of the variables in the overall scheme of running a studio for profit.

IT'S NOT ALWAYS THE MONEY
One final point: if your operations never improve beyond Year 1, you are better off sticking with your day job and picking up overtime. I know, I hated working a day job too. There are non-financial considerations to this business that make it worthwhile anyway. But don't lie to yourself about the finances because of it . . .

Self-Mastering Objectivity

---

By Jon O'Neil on July 9 2008 01:04 PM | Permalink | Author Info

---

You've maybe read my rants and raves about hiring services vs. DIY.  Let's say you've carefully considered that, and it makes sense to DIY.  But in order not to waste too much time and test CDs to drag out to your car, try to inject some objectivity into the process.

OK, so how do you "master" your own stuff?  You know the disadvantages . . . the same monitoring chain you used to mix, the same pair of ears.  What are your advantages?  Well, first and foremost, your time.  You have lots of it.

Let some time elapse so you can listen to the mixes fresh.  Then, before you do anything else, burn a CD and keep it in your car (iPod, whatever) for a week.  Take some notes.  When you get back in your studio, try to figure out why you didn't notice that stuff while mixing.

Some of those changes you will make--perhaps most--will be remix decisions, and that's OK and maybe even good.  But at some point, you gotta decide the thing is done.  I also believe that there are certain things that are actually best done on the master bus, just to add some glue.  But those aren't most things or many things, so don't be lazy about mixing.

Objective things:  If you use nearfields to mix, don't use them to master.  What, you say?  You only have one pair of monitors?  Fine, just don't use them as nearfields.  Spread 'em out to six feet and put yourself six feet away from each (equilateral triangle).  Wait, you say, your room is totally untreated, that means lots of problems with echoes and standing waves!  True, but that's also how most of your listeners are set up, if they are not using headphones.  The goal here is to hear your mix in a different situation than your mix environment.  When you lack an excellent listening space and high-quality full-range monitors, your alternative is to listen to your mix on a lot of different systems.  So reposition your monitors to create a new system without actually spending any money!

While I'm on the topic of headphones, listen on headphones.  You can't mix on headphones, you can't really master on headphones, but I do every fade using headphones, because you can hear way way down to the bottom of your dynamic range without having room noise mask uglies that may lurk there.  Does your track have excess hiss as the last note dies?  Is that decaying guitar full of 60Hz hum?  Is there video whine that just won't go away?  You might not be able to hear that on your monitors, but if you can in cans, your listeners will too.  Use some noise reduction or get those fades fast and tight.  An "S" curve fade is great for a longish fade that squelches low-level noise quickly at the same time.

 

OK, you're in the thick of making those master bus changes.  I am not going to get into what you should do, because I don't know, I can't hear your mix.  And that part is very subjective, and described many other places.  But I will give you one hint:  you need to hear clearly what each plug you add is doing, and frequently.  Ears have poor memory.  Putting plugs in bypass is a good thing, but it works best if the volume level doesn't change when you bypass the plug.  If the plug has an output level control, USE IT!  Match the level in and out of bypass.

But eventually you're going to get to the final stage, which is probably going to be a limiter.  How do you know if your limiter setting is making your track punchy loud or just sounding like distorted bunghole (which would be a bad thing, the bung wouldn't form a tight seal and the wine would be spoiled!)?

Answer:  if your limiter doesn't have an output level control, set a gain change plug after the limiter to reduce the level to match the pre-limiter level.  Now you should have a master bus chain where any individual plug or the whole effects chain can be put into bypass without changing output volume more than 1dB or so.  And then when you are finally really all the way done, remember to set output gain back to full scale!

And that's it.  And oh yeah, buy an SPL meter and calibrate your monitors to proper listening level.  But that's a tale for another time!