Friday, November 15, 2013

Preamp quality - a simple but effective analogy

Preamps are funny things.  They're generally considered to be as invisible as possible to the recording process.  The obvious exception is that some of them have a "sound" to them that is desirable.  You don't buy a Les Paul because it is a transparent sounding guitar.  You buy it because it sounds like a Les Paul.

That exception aside, we generally want our mic pres to be as clear and as clean as possible.  So, you go out and get your entry level preamps when you're just starting out, and say to yourself, "Gee, they sound fine to me.  I don't hear any noise.  It sounds pretty clean.  Why would I pay five times the price?"  And then you splash out and spend five times the price and you hear it, and you think to yourself, "Gee, I guess it sounds better.  But at five times the price, the difference is barely perceptible."

I mean, really... listen to the Behringer preamps  and they do sound pretty darn good.  A/B them against even a fairly semi-pro unit like my Steinberg MR816 ($1000 for 8 preamps - not exactly in the $1000 per channel just yet...), and you'll hear that "not as different as I might have thought" that I'm talking about.

What is important to understand is that the difference is cumulative.  Record 16 channels with entry-level pres and you'll fight like heck with it to get it to where you want it to be - if you ever actually get it there.  Record 16 channels with much better preamps, and the project will almost seem to mix itself.

Here's why.  (click on the image to make it larger.)

The image on the left represents the better preamps, and the image on the right represents the cheaper ones.  Looking at the single red dot at the top does not reveal that much of a difference.  However, in the context of the overall picture, made up of a number of dots, the difference becomes much more noticeable.

THAT is the difference between entry level preamps and something you'll pay more for.

Saturday, May 4, 2013

Sound Treatment and Sound Proofing

Sound Treatment:

The first thing to note is that sound treatment and sound proofing are not the same thing.  Sound proofing is about making things sound quiet.  Sound treatment is about making things sound good.  These things really are mutually exclusive.

Consider the recordings you have made on your phone, or with a video camera of everyone singing Happy Birthday to grandma on her 90th birthday.  You know that hollow, boxy sound?  People blame it on cheap microphones.  That may be partly true, but the bigger culprit is a lack of sound treatment in the room.  What happens is that sound waves are hitting hard surfaces and reflecting off of them, and ultimately, back into the microphone.  In a very large room, you’ll hear a natural reverberation or maybe even an echo.  In a smaller room, those sounds will be so close together that they will just smear and make a mess.  So, there are two factors at work here - the size and shape of the room, and the reflective surfaces.

Imagine a 10x10x10 room with a sound source exactly in the middle.  The sound travels from the source and bounces off the walls and around the room.  Because all of the surfaces are pretty much equidistant, the reflections make a huge mess in the center.  That is, admittedly, over-simplified, but it gives you a sense of where some of the problems do lie.  One remedy to this problem is to construct the room so that the length, width and height are not even multiples of each other.

Reflections can also be controlled by angling the walls.  In a square room, the walls are parallel, and the corners are square.  Sound bounces off them and off each other quite a lot, with bass frequencies taking up residence in the corners.  Many recording and mixing spaces are constructed to be pentagonal and with a sloped ceiling - no parallel surfaces to reflect sounds back into the middle of the room, and no square corners for them to build up in.

But the size and shape of the room are usually already predetermined.  Your final solution is to add material to the walls (and the ceiling and floor!) to absorb some of those reflections.  Be careful, though, as you don’t want to over-do it.  A typical material is semi-rigid fiberglass, such as Owens-Corning 703 is the go-to product.  This can be difficult to find in Canada but was able to locate a very similar material in Stoney Creek at Glass-Cell Iso Fab.  They have other locations in Ontario.  As a last resort, you could source out some Roxul “Safe and Sound,” but it is NOT the same thing.

Here is a sequence of photos that shows how I built my own absorptive panels, including a “cloud” for my home studio.  

The link also shows how I built what are called “bass traps” in the corners.  Bass traps help to prevent sounds from building up in the corners of rooms.  Low frequencies are particularly subject to this, which really does mess with how sound is perceived and distributed in the room.  Essentially, a bass trap is a triangular wedge that absorbs frequencies, while minimizing the “squareness” of the corner.  

The difference this makes is truly remarkable.  Before treating my room, I did a frequency response measurement of my space, and it was horrifying.  I had some frequencies that were represented literally more than twice as loud or twice as quiet as others that were only a few notes away.  No wonder I couldn’t mix anything to save my life!  Now, with some changes made to how and where I was placing my monitors, along with my acoustic panels, the room is flat from about 60hz to almost 20khz, within +/- 3db.  It’s not perfect, but I can mix in there now with confidence that what I am hearing in my room when I mix will successfully translate to other systems in other rooms.

Sound Proofing:

There is no such thing as cheap soundproofing. It's complicated, and it's expensive. Anything that does not match that description is going to be dodgy at best.

Essentially, soundproofing comes down to a simple formula:

mass + insulation (even air space is fine) + mass = soundproof. The effectiveness of this equation is dependent upon how much mass you have, the quality of insulation, and how air-tight the air space is.

Generally, what is done to soundproof a space is to build a room within the room such that the floor, walls and ceilings are not connected to the outer floor, walls or ceiling with anything that will transfer vibration.

The "outer room" will have studs placed not against the wall, but in from the wall a little bit. The wall frames and floor of the "inner room" will be elevated and placed on dense absorptive foam rubber or similar material. The "inner room" will also be connected to the ceiling of the outer room with a similar kind of material. The "inner room" will be constructed with drywall or something - maybe double-thickness - placed on *both* sides of the studs. This gives you an airspace between the two layers of drywall, and an air space between that and the outer room.

The studs for the walls will also typically be offset from one another so that the outer wall and the inner wall do not touch - almost like one frame for the outer part of the wall and another frame for the inner part of the wall.

Think of this... you know that old trick with the tin cans and the string where the sound travels along the string? Well, you can sound-proof the crap out of your walls and ceiling, but the sound will travel along the floor across the joists and follow the joists to the outside. Similarly, if your inner room and outer room are connected... say... the screws for the drywall going into the studs, and then those same studs touching the outer layer of drywall, you have the potential for sound transference from the inside to the outside.

Think of this too.... sound travels through the air in waves. That means that if air can get out, then logic says that sound can get out. No prob.... just block off all the air holes with enough mass and insulation and enough mass again to absorb all the sound waves. Er... wait a sec..... but if air can't get out.... how can it get in? Ooops.

Oh, the egg cartons.... the reason they don't work.... practically no mass! And they typically get installed such that they touch the walls, so sound travels through them to the wall and to the outside.

Further Reading:

I STRONGLY advise three resources:

This book is written in very friendly language, and is really well explained. I can't recommend it enough. I have it. The author is Rod Gervais, and the book is, by most people's standards, the bible on building a home (or garage) studio. It also recommends techniques, materials, a bit of acoustical physics, etc.

This forum is hands-down the best resource on the internet for studio construction. John Sayers builds studios into the 100's of thousands of dollars and you can see his work on the site. He frequents the forum. Rod Gervais, the author of the book above also frequents the forum. So do many other bedroom experts, as well as many who are actually acoustical engineers.

Ethan Winer is the developer of a company called Real Traps, an industry leader that specializes in acoustic room treatments. This is his acoustics forum, and he posts there frequently.

A primer on microphones

When discussing microphones, I like to use two analogies.  First, I will suggest that mics are a lot like saws. There are many different kinds for many different purposes.  Different microphone designs and technologies will make one microphone much better suited for one application than another, but for a different application, a different design or technology will be required.  Like saws, where you *could* use a scroll saw to cut a board, you *could* use a small diaphragm omni condenser on a lead vocal... but in neither case is it typically going to be the best choice.

Generally speaking, the technologies involved in microphone design involve either the size of the diaphragm, the way the diaphragm converts sound energy to electrical energy, and the capsule design, which affects how the microphone responds to sounds coming in different directions.

Diaphragm size:

Microphones have a very thin, flat moving object that is made to vibrate back and forth as it converts sound energy to electrical energy.  This is called the diaphragm, and they generally come in two sizes - small and large.  Small diaphragms tend to be more accurate, but large diaphragms tend to sound more robust, or even “larger than life” sounding, in much the same way that a larger speaker tends to sound more robust.

Dynamic vs Condenser:

Most microphones are either dynamic or condenser (sometimes called electret).  The technology used does make a difference, but not as much as you might think. Many people are too hasty to make the over-generalilzed conclusion that a condenser microphone is “better” than a dynamic one.  The SM7 is a dynamic mic with a large diaphragm. It was used to record the lead vocals on Michael Jackson’s Thriller album.  It's a wonderful sounding vocal mic.  However, a condenser mic is a more "standard" choice for vocals. They tend to more airy and open sounding, and are more sensitive to transients and dynamics.  Dynamic mics introduce a little bit of natural compression to what you're recording, (a function of the physics that determine how they operate) which may or may not be desirable.  Dynamic mics are less likely to feed back than condensers, which make them generally better-suited for live sound.

Both dynamic mics and condenser mics come in both small and large diaphragm varieties.

There are other microphone technologies, such as ribbon mics and PZM (pressure zone microphones, also called “boundary” microphones), but by the time you’re likely to buy one of those, your knowledge will likely surpass the scope of this article.

So...  we have small-diaphragm dynamic (or more like medium diaphragm) mics such as the ubiquitous SM57/58.  These are generally rugged microphones best suited for miking instruments and live vocals.  We have large-diaphragm dynamic mics, which are good for lower, resonant sources, like the “big radio announcer” voice, or a kick drum.  There are small-diaphragm condensers, which are great for instrument spot mics, and capturing accurate sounds in a room, like for miking an orchestra, or for drum overheads.  Finally, there are large-diaphragm condensers, which are primarily suited for lead vocals, but can be applied to many other things that you might want to sound detailed, yet larger than life - an acoustic guitar, or even drum overheads.  

Polar Patterns (directionality)

So, within all of these combinations above, any of them may be tailored to pick up sound better in some directions better than others.  The two most extreme directionalities are hypercardiod and omnidirectional.

A hypercardiod mic will pick up sounds mostly directly in front of it.  This is useful when you want to isolate one instrument from the sounds around it as much as possible, like miking an acoustic guitar while rejecting a singer in the same room.  Another significant advantage with a hypercardiod mic is that, because they are so directionally specific, they are less prone to feedback in live situations.

At the other end of the spectrum is the omnidirectional mic.  Omnis pick up sound more or less equally in all directions.   These can make for very useful room mics, where you want to pick up many instruments at one time.  Omnis tend to not make very good stereo overheads, though, as both mics generally wind up picking up the same thing, meaning there is little separation between right and left channels.  

Cardiod mics are the most common as they are reasonably directional, but not too picky.  Consider a live application where you have a singer/guitarist who plays while standing in front of a microphone.  A hypercardiod mic would have the singer fading in and out as he/she made small movements away from the focal point of the mic.  A cardiod mic would be more forgiving.  

Finally, we have figure-8 polar patterns.  These are found in microphones where you visually have a “front” and “back” of the mic. Visually consider what the number eight looks like - a top and bottom bulb, connected by an extremely narrow “waist.”  The top bulb would represent what the front of the mic picks up, and the bottom bulb would represent what the back of the mic picks up.  The mic does, indeed, pick up sound from both sides.  This could be used for recording a pair of backup singers, for instance.  What just as useful, though, if not more so, is the “waist” of the figure-eight.  That “waist” represents the side of the microphone.  A figure-eight microphone rejects all sounds coming from beside it.  This area of rejection is called the “null.”  A typical application of figure-8 microphones is recording a person who sings and plays guitar.  One figure-8 mic will be used on the voice, but placed sideways so that the null of the microphone (the “waist” of the 8) points to the guitar, thus rejecting as much of the guitar as possible. Another figure-8 microphone will be used on the guitar, also placed sideways, so that the voice of the singer is in the null of the second microphone, thus rejecting the voice.  This means a minimal amount of bleed between the two microphones, allowing for a good deal of control of both the voice and the guitar when mixing.

Choosing a microphone:

My second analogy is that microphones are a lot like wines.  You know how some people or restaurants will have a collection of wines, because the trick is matching the right wines with the right meals?  Mic selection works the same way.  Some singers - not many, mind you - will sound fantastic with an SM58 thrown in front of them. I've only experienced that once, though.  You could liken that to the bottle of Baby Duck that was just perfect with the potato salad at the family picnic last summer.  Most singers - not all, mind you - will sound fantastic (or as fantastic as they're going to sound) when you throw a U87 in front of them. That would be the equivalent of saying that you can serve Dom Perignon with almost anything. That's one of the reasons that, like that bottle of Dom, the U87 is so expensive. (not even close to a Sony c800G, but the last time I spent that kind of money on anything, I drove it home...)  To go back to the Thriller example I used earlier, the producer surely had access to a U87.  There is no doubt about that.  But in that particular case, it was not the bottle of Dom that got served.  The producer found that the SM7 wine matched the Michael Jackson entree better.  

Finally, I want to kill any notion you might have of using frequency specs to determine the quality or suitability of a mic  I thought that flat and transparent would be the holy grail of microphone ideals at one time too.  If you look at a frequency response curve for some of the most expensive mics, and they are NOT flat. That "not flat"-ness is the character they impart, and it is that character that people are willing to spend big bucks on. You don't pay big money for a Les Paul because it has a good balance of tone across the frequency range.  You buy it because it sounds like a f**king Les Paul! Same thing with mics. (and preamps!) By comparison, the most accurate mics in my collection are a pair of Behringer ECM8000's. Their frequency response curve is more accurate than any other mic on the market until you start hitting the $3000 range. They are small-diaphragm condensers (no surprise). They were about $40/ea. I rarely use them.  They are so accurate that they are entirely uninteresting, and entirely unflattering.  I am reluctant to part with them though. They may well, because of their accuracy, be just the right wine for the right meal at some point.

One last thing....This may be the greatest resource on microphones on the planet:  

It is a condensed version of a 97-page thread from another site.  Harvey Gerst, the author, has 50+ years of experience as a gold record songwriter, recording engineer, producer, and product design specialist. He's worked with Bob Dylan, Albert King, Jimi Hendrix, The Doors, and Jefferson Airplane. He designed prducts for JBL, Trident, Morley, etc. In short, there are few people who know more about this stuff than him, and he has chosen to share it.