Audio samples are below.
Summery: If you record to a DAW you may want to take a second look at your dynamic mics. About a year ago I started to have experiences counter to what I had read. Again and again I preferred dynamics mics over condensers on what were typically condenser tasks. Here is my experience, some samples, and my theory on why this is.
Strange Happenings: About a year ago I tried an SM57 along with some condenser mics on an acoustic guitar. To my surprise I liked the 57 best. A lot of people hate the 57 and wouldn’t consider using it on an acoustic guitar. It rolls off the high end and the low end. It has a big peak between 3-7k. And it isn’t very sensitive. So why would I like it better on acoustic guitar? An acoustic guitar’s nuances, I was told, were captured better with a sensitive condenser mic. Then I had the same experience with capturing cymbals and a rap performance (The rap performance samples are further down the page). Now I use my dynamic mics more then my condensers. I started wondering why?
Were my ears getting worse with practice? Doubtful. Did it have something to do with my converters and high end? Was the precision of digital coupled with sensitive mics too accurate and sterile sounding?
Before I put forward my theories on this I want to be clear about what I’m saying. I’m not saying my dynamic mics were more sensitive to transients. They weren’t to my ears. I’m not saying they had a wider frequency response. I’m saying they mixed better out the gate.
Another caveat: Not all condensers behave like typical condensers. This is true of dynamics as well. If you look around you can find condensers that sound like dynamics and vice versa. My argument is an oversimplification for sure.
It’s digital: My first theory is still the one that convinces me the most. With digital being so precise and accurate you may not need very precise mics. When I first started looking into mics about 10 years ago most of the ideas came from books. These books were written by talented engineers who came up using tape. I’m told, and believe, tape does a few things. It rolls off the high and low end. It compresses. If pushed it adds pleasant harmonics. Compared to digital it’s not so accurate but that’s not a bad thing.
I personally do not want to hear all the high end trash and low end mud in every source. Most systems can’t reproduce it and the human ear is much more in tune with mid-range anyway. You know… Nail the mids, nail the mix.
If you start reading modern books on DAW mixing you might notice the practice of high pass filtering everything at around 50hz. Further proof that the “perfect” nature of digital needs some taming.
We’ve all read the microphone rules. They aren’t rules, more like guides: Use large diaphragm condensers for vocals. Use a SM57 for guitar cabs. Use matched condensers for drum overheads. But are these cultural inertia? Something handed down from a time when recording technology was very different. A time when no one went digital and there were no cheap Chinese condensers.
On a relevant tangent my first book on mixing was bought 10 years ago. It goes through the nuts and bolts of mixing from a pros perspective. Width. Depth. Frequency spectrum. What it does not deal with is the evolution of recording. For example most of the people reading that book probably aren’t pros. They don’t have great rooms, great mics, or great monitors. A much better book for people like me is Mike Senior’s. He deals with the evolution of recording technology. His book takes a look at the problems and solutions for people working with DAWs in less then multimillion dollar facilities.
Is there a bigger place for dynamic mics with today’s digital recording? Has cultural inertia biased us to discount these less sensitive mics? Is is just coincidence that dynamic mics do some of the same things tape did? Compress frequency range. Compress dynamics.
Here are some samples of a rapper I recorded with 2 condensers and 2 dynamics.
All of these sound alright to me solo. But when I try to get them to lay in the mix the SM7 was my favorite. It required much less work.
If you listen close you will hear the condensers make pops, clicks, and sibilance more pronounced. This is what I’d expect from a sensitive mic. But I didn’t want or need it on these vocals. In the mix this sensitivity made the vocals seem too upfront and disconnected from the music. After adding some compression the issues got worse. The dynamic mics on the other hand smoothed a lot of this out naturally. That means I can do less processing. My preferred method is to process sound as little as possible. I’m no purist. I just think you get better results that way. Put the right mic in the right place and things will mix naturally. Use the wrong mic and placement and you’ll end up doing a lot more work and it still won’t sound as good.
Another Theory – Converters and High Frequency Content: Just by coincidence I was reading one of my favorite magazines, Sound On Sound, and came across an interesting Q and A about sampling rates. The question was an old one I’ve seen asked a million times. Basically the reader wanted to know if a certain sample rate was best. I’m pretty set on using 48k but the response looked interesting so I decided to read on.
Lets see if I can sum up one of the points without a separate blog post. Um… First I’m going to back up a bit.
There is this thing called the Nyquist Theorem. To accurately sample a signal at frequency X your sample rates must be at least 2X. For example you’d need at least a 40k sample rate to accurately sample a sine wave of 20k. The threshold of human hearing is about 20k. It’s probably no coincidence that CDs have a sample rate of 44.1k allowing them to accurately sample sound at frequencies up to 22.05k.
You might wonder what happens to the frequencies too high for the sample rate. After all real world sound can easily exceed the human hearing range. If you ignore the Nyquist theorem and use a sample rate less then 2 times the signal frequency you get this thing called aliasing. It’s a bit hard to describe with words but easy to understand with a diagram. Here’s one. So aliasing will introduce lower frequencies never present in the original. That can’t sound good. And that’s why we have… filters.
Filters remove the high end content that the sample rate can’t handle according to the Nyquist theorem. There is no aliasing and everyone is happy. But all filters aren’t created equal. Some are badly designed. Bob Katz makes the argument that the sonic difference in sample rates is actually a filter issue not a sample rate one.
So what if a badly designed filter lets through some high frequency above what Nyquist allows? You get aliasing. So the Sound on Sound idea is certain sources with lots of high end content like cymbals can cause aliasing if the filter is poorly designed. With me?
This is where I started to think about condenser mics. My condensers certainly deliver more high frequency content then my dynamics mics do. But I seriously doubt my RME Fireface has poorly designed filters. Even so this may be helpful information for the thousands of people using super cheap AD converters.
Finally: We’ve all heard it said before. There really are no rules. So why not try that dynamic mic on atypical material? You may be surprised. I was.