I found that installing these panels made a lot more difference in the quality of my recordings than upgrading a preamp or a/d converter, or even buying a new microphone. By improving the sound in the room, the acoustic treatment made the whole recording process much easier and more enjoyable. So when people ask me how to improve their recordings, one of the first things I suggest is room treatment.

Minimum Impact, Maximum Result

Unfortunately many people don’t have a dedicated recording space. They can’t stuff a dozen panels into corners or lean them against walls, and they often ask how few panels they can use and gain any benefit. My standard answer is two, just two panels can reduce reflection in one small area, an area just right for recording acoustic guitar or vocal or many other acoustic sources.

There are a number of commercial products born out of this idea of a zone of controlled reflections. Some of these are rather small and thin, so I’m doubtful that they’re strongly effective. Some are fitted with frames and/or attachment hardware. This adds flexibility and a pro look, but also adds weight and cost. My preference is a pair of simple unframed panels, 2 feet by 4 feet by 4 inches of OC703 compressed fiberglass. I made mine with burlap and hot glue, but premade panels are available as well. Here’s the blog entry on building these panels: Building a broadband absorber (on the cheap)

I deploy the panels in a V surrounding the microphone(s), often by simply leaning them against the mic stand if it’s sturdy enough. If needed I lock the panels in place using a bent coathanger poked into the fiberglass to add a little stability. I’ve experimented with different positions for the panels and this arrangement has seemed the most effective to me. The sound leaving the guitar, or at least a lot of it, passes through the panels as it leaves, so it hits the microphone full strength, then gets attenuated by the panel. Then after the sound hits the wall it returns through the panels for further reduction before hitting the back and sides of the mics.

Let’s Listen

In order to evaluate this two panel approach I set up in our guest bedroom, a 12 x 14 x 8 foot space with no room treatment and only a futon for furniture. This is a fairly reflective room and seemed like a good place to test the two panel solution. I used the Zoom Q3HD as the “mic” and placed it about 24″ from the guitar, with the stick-on wide angle lens attached Zoom Q3HD with a Stick-on Wide Angle Lens then I setup the Xacti HD2000 for a wider shot.

Here’s the resulting video:

Conclusion

I hope you’ll listen and make your own judgment, but my opinion is that the two panels do a remarkable job of cleaning up the early reflections and reducing the “small room” sound. The direct to reflected ratio is improved with a couple of benefits. Since the reflected sound includes a lot of high mids, the sound with the panels is “warmer” with those reflections reduced. The other effect is to make the recording sound like it was made with a closer mic position or in a much larger room.

I’m pleasantly surprised by the effectiveness of this two panel approach. They’re small enough to be stored under a bed or behind a door, they’re light enough that they can be moved into position quickly and easily, and even purchased commercially the cost is under $200. It’s possible that one or two more panels might improve things further, but in my preliminary comparisons it seemed that the big improvement came from two panels, and one or two more panels did not provide much additional benefit. Your situation could certainly be different, of course.

Now for the icing on the cake – the development team has linked in the FFmpeg video libraries and given REAPER the ability to do simple video editing. I’m totally happy with my video editing system since I moved to Edius Neo 2.5, but when I played around with video in REAPER I realized that this is a tool many musicians want and need.

Many of us like to post simple performance videos on YouTube – not big production numbers but simple clips showing off our latest tune or instrument or technique. And we want to do this on a budget. We use inexpensive pocket video cameras, and they have at best mediocre sound quality. Many of us also have a fairly high quality recording system of some kind. Now with REAPER we can create our videos just like they do it in the movies – recording the sound on a high quality audio system positioned for the best audio quality while capturing the video at the distance needed for framing and perspective.

Using video in REAPER is a snap. First, go to the CockosWiki Video Support page for instructions on adding the FFmpeg libraries to the REAPER program directory.

Once the FFmpeg libraries are in place, just open the video file like any audio file, or drag and drop from your file browser into the REAPER track window. I’ve tried clips from three camcorders so far, the Kodak Zi8, the Flip Ultra HD, and the Sanyo Xacti HD2000. REAPER handled all of them with ease. Only the .mts files from my Panasonic Lumix TS2 failed to open, and I was able to convert them to an AVI that REAPER liked.

If I’m shooting video around my studio, I’m all ready to record audio into REAPER, so there’s another step out of the way. If I’m shooting on location it’s a snap to bring the recordings back to the PC and drop them into REAPER. The whole process is so easy that instead of writing detailed instructions I decided to simply shoot a video of shooting a video and demonstrate the syncing process.

So that’s how easy it is. Drag and drop, drag and drop, trim, drag and drop, trim, and render. When I think of the hours I’ve spent calculating frames in Avidemux, or waiting for Premiere Elements to redraw, or being told by Windows Movie Maker that it can’t deal with my file …

Here’s the video that I created in REAPER in the process of the tutorial. The Flip would have been happier with more light, for sure, but the quality you see in this video is right there with the original from the Flip. And the audio is in a whole different, and better, league. Naturally, since I was working within REAPER I could have easily added effects, applied EQ, worked over the audio to improve the result.

My understanding of microphones and audio says this myth is not true. I’ve been told by knowledgeable experts that mic sensitivity is linear below physical clipping regardless of the transducer technology. I actually asked this question some time ago over on rec.pro.audio – a usenet discussion forum where professional audio engineers hang out. Some of their responses were rude, but they basically established that preamp gain is all that is required to match sensitivity between a dynamic and a condenser mic. Here’s a link to that thread.

Now I believe those guys. I thought it might be interesting to devise a demonstration of sorts, by plugging in a dynamic and condenser mic, playing a reference tone through a speaker in front of the mics, then adjusting preamp gains so the levels match. Then I could generate a quieter and quieter test signal by walking away from the mics making noise. Recording this diminishing sound with both mics would tell me if one could “hear things” the other could not.

Oh No!! Is the Myth True?

To my amazement, when I did conduct a demonstration for myself, I could hear more distant, quieter sounds on the condenser mic. It seemed as if the myth was true.

I knew the error was not the physics or engineering of mics, but rather something I had overlooked in setting up my test. After a day or two of research and pondering the light came on. I realized that the different pickup patterns of the two mics made my calibration procedure wrong. I was calibrating for direct, on-axis sound, but I was measuring diffuse off-axis sound. I needed to do the calibration using only the diffuse sound field, which meant moving the speaker some distance from the mics during the calibration.

A Better Test Design

At this point I also figured out that I didn’t have to move the noise source to reduce the level of the test signal, I could create test tones that got lower and lower in level and play them back from the same spot as the calibration (well, duhhhh).

Here’s the result. For the best fidelity, here’s 45db-66db-xt.wav. Or if your connection is a bit slow, the compressed version is 45db-66db-xt.mp3.

You’ll need to download one of the files and pull it into a player that can select only the right or left channel. The mic in the left channel is the condenser, a Shure KSM141. The mic in the right channel is a dynamic, the Shure SM57. This is a little excerpt from the demonstration recording. I selected the area from -45 to -66 dBFS, which is where the test tone slipped into inaudibility. I raised the level of these files substantially and there is plenty of broadband noise, so be careful not to play them too loudly. The condenser mic is in the left channel, the dynamic in the right. Listen to first one side, then the other, and see if you can hear tones at lower levels from one mic or the other.

Conduct Your Own Demonstration

If you’d like to conduct this demonstration with your own mics and room, all you need is a calibration tone and a test tone series. You can generate the calibration tone in most audio editors, and you can download my test tone series.

If you can’t figure out how to create a tone in your favorite audio workstation software, download Audacity and install it (I recommend the Beta 1.3.xx or later version). Start Audacity and choose

Generate | Tone

then fill out the Tone Generator form:

Waveform: Sine
Frequency (Hz): 1000 (a 1 Khz test tone is the normal industry standard for basic testing)
Amplitude: .6 (a very loud long 1 Khz tone can damage your speakers and possibly your ears)
Duration: 600 seconds (10 minutes should be enough)

Click OK and you’ll see a strange solid waveform. That’s your calibration tone. Just export it from Audacity:

File | Export
Save as type: (either MP3 or WAV Microsoft signed 16 bit PCM)
(Choose a directory and file name)

You can download my test tone file here. The test tone file contains volume level announcements and 1000 Hz tones starting at -9 dBFS and going down to -90 dbFS in 3 dB increments.

Here’s a screen shot of the test tone file:

Mic sensitivity test tone in Adobe Audition 3

and a sample of the tones, starting at -9 dBFS and going to some of the lower level tones:
test_tone_xt.mp3

To conduct your own demonstration, connect two mics to your recording system. Place the mics at least 6 feet from the speaker, then play the calibration tone. Adjust the preamp gain so the two mics show the same input levels. Do this very very carefully, this is the most critical step in the process.

Next, simply play the test tone file while recording the two microphones. You’ll want to wait until a quiet part of the day, and be prepared to sit very quietly while the test file plays. When you’ve completed recording the test tone sequence, listen to one of the tracks you just recorded. When you can no longer hear the test tone, switch to the other track (other mic) and listen again. If your experience is like mine, the test tones will fall into inaudibility at the same level for both mics.

Better Recording By Knowing Our Tools

Mics are fascinating devices, but they’re engineered objects in the physical world. We can make better use of them if we have a better understanding of the way they really work instead of relying on incorrect assumptions and erroneous analogies. In the past audio testing required lots of expensive dedicated equipment, but now with our computer audio systems we can easily perform simple but fairly sophisticated evaluations of our audio gear, and learn to make better recordings in the process.

Sometimes when I’ve listened to mic samples I thought I heard these dramatic differences, but after a bit I realized that I was listening to different performances, not different mics. Sure the mics had been changed, but the player was hitting the strings differently and playing different riffs at a different volume – so how could I tell what part of the difference was the mic, and what part the player?

Since then I’ve tried to do some mic tests of my own, and I’ve tried to educate myself on audio testing. At this point I’m beginning to think that the differences in microphones are a lot more subtle than I had been led to believe, which makes a careful test even more important. As I’ve mentioned before, very small differences in volume are registered by our ear/brain combination as differences in quality rather than loudness. I’d like to demonstrate the steps I now take to try to make my mic comparisons, and preamp and a/d comparisons, meaningful.

Mic Setup

The mics need to be close together, but not interfere with each other. I try to determine the location of the diaphragm, the part of the microphone that gets hit by the sound waves, on each mic, and align them along that position. This is pretty easy with large diaphragm mics, often a bit tricky with small diaphragm units. All the mics should be on axis to the source or all should be equally off axis. Check pattern switches, rolloff settings, pads, any feature of the mic than changes its response. I’ve missed this step a few times.

I also arrange the mics so they don’t interfere with each other. For instance, it’s not a good idea to tie a bunch of small diaphragm mics into a bundle, because the ports behind the capsule contribute toi the frequency response and pattern of the mic. If those ports are blocked the mic is not performing normally.

There are three mics being compared in this example. The CAD M179 is a low cost continuously variable mic in an unusual looking housing. One salesperson began a drumbeat for these mics and they are now widely recommended in internet recording forums. The Rode NT2a is one I’ve used before. I think it does a fine job of capturing audio with low noise and no noticeable distortion, but many forum posts are negative toward Rode in general and the NT2a in particular. The Audio-Technica AT3035 has recently been discontinued, replaced by the AT2035. In the past I’ve read a wide range of comments about this mic, with some calling it a secret gem and others decrying its overwhelmingly bright character.

Here’s a picture of a mic array arranged for testing. A cheater clamp that fastens to the mic stand main tube can help fit an extra mic into the array. Adapters to attach mics at an angle can be very handy as well.

Three large diaphragm mics in an array for comparison

I often hear this arrangement, with all the mics equidistant from the source, criticized because different mics work best in different positions. I strongly suggest that if you wish to evaluate different positions you try all the mics in all the positions. It’s a little more trouble, but it’s a great learning opportunity, and omitting this step means no valid comparison can be made. We’re right back to asking what part of the difference is the mic, what part is the location with no way to answer the question.

Of course, my aim here is different from trying mics and positions to get a good recorded sound. I’m after data, not art. When the aim is art, the ears rule!

Equipment wise, you can do a useful comparison with as few as two recording inputs. In fact, comparisons that are done two by two are probably the most revealing and useful. Still its fun to line up three or four mics for a single session – in which case it’s helpful to have four identical channels, especially if the comparison will be made public. In my experience, more posters will criticize a change in preamp than will mention a change in performance, amazingly enough.

Here’s a video that illustrates the setup process and takes you through gain calibration, discussed below:

Microphone Comparison - A Tutorial - part 1 of 2 from Fran Guidry on Vimeo.

Gain Calibration

Once we’ve arranged the mics to capture a single performance we need to minimize the volume difference in the our samples. I like to start with a test tone played into all the mics from a speaker only a foot or so away. A 1000 Hz tone is not very challenging, so even a low cost computer speaker can be used as the source. A test tone is easy to come by and it doesn’t need to be calibrated for volume since we’re interested in relative rather than absolute levels.

I’m using Reaper to conduct this test, and a little inquiry on the Reaper user forum and learned about MDA Test Tone, a plugin that provides the beep.

In Reaper I created a project with MDA Test Tone on one track, and three tracks for recording the three mics. I also adjusted the range of the Reaper console meters. With these connections in place I hit record and adjusted the preamp gain for each mic so they all were very close to -18 DbFS.

Next I discarded these recordings and captured three more tracks to store a consistent level for each mic. These 1000 Hz tones are used to fine tune the track levels later.

Recording

In the next video we actually do a little recording. After all the setup and calibration work, the recording process is anticlimactic. Move the calibration speaker, tune up the guitar, hit Record in Reaper, play a few bars, and we have our clips in the can. Be sure to record the clips on the same tracks as the reference tones we created in the last step. That way when we adjust the reference tones we adjust the clips at the same time.

For straight mic comparisons, I like to position the guitar about 32″ from the mics. This avoids proximity effect and hopefully presents a fully developed guitar sound to the mics, instead of one mic picking up the neck of the guitar while another picks up the bridge.

Here’s the video demonstration of recording simultaneous tracks and fine tuning the gain:

Microphone Comparison - A Tutorial - part 2 of 2 from Fran Guidry on Vimeo.

Volume Fine Tuning and Rendering

As the video illustrates, working with that 1000 Hz test tone can be pretty annoying, but we need to delve into it one more time. Actually, this time we can turn the volume down, because we’re adjusting levels “in the box” – that is, internally in the computer.

The helpful folks at the Reaper Forum pointed me to the Sonalksis FreeG plugin. This tool adds high resolution metering and gain adjustment we can use to tweak the levels of our clips that last little bit.

Simply place the Reaper track cursor so the recorded test tone will be played, reset the FreeG meter, and play a bit of the clip (the space bar starts and starts playback). Note the RMS level, and repeat for each track. Then, to be fair, apply a gain adjustment to each track so they all show the same RMS level. In the video I’m fiddling with the Gain knob using the mouse, but after I recorded the session I realized that you can simply type the desired gain change into the value box, making it very easy to get the level just right.

Finally, click and drag the cursor to make a time selection of the recorded clips, then render each track separately as illustrated in the video. It’s a good idea to render to 44.1/16 format if you plan to make the clips public, because everyone can play this CD standard format.

Blind Comparison

I have had the experience many times of listening to mic comparison clips and clearly hearing the difference between them when I knew their identities, then finding that I could not hear a difference at all when I hid the identification in some way. Even getting momentarily confused about the source of the clips has been enough to change what I “hear”. Our brains are at least as important as our ears in defining what we hear, and our brains like new stuff, shiny stuff, expensive stuff. So when we know a clip was made with our shiny new expensive mic, we’re going to “hear” how wonderful it sounds.

For a test to be meaningful, we need to hide the identity of the clips somehow. This is pretty hard to do when working by ourselves. The foobar2000 audio player offers one solution, with the ABX testing utility built-in, as described in this blog post. This is a powerful tool, because it not only offers a way to test clips double blind, it helps us determine if we can hear any difference at all before we try to determine a preference.

Mic comparisons are useful for our own recording knowledge, but it’s even better to share. There are lots of folks hanging out at recording forums on the internet who are looking for information about mics and other recording gear. I like to contribute when I can by posting comparison clips, but I think it’s important to make the original post without identifying the devices used. It’s more informative, and more fun too.

So here are the three clips I recorded in the video, with no EQ, no compression, no reverb, no processing of any kind except to match volume levels and trim ends. To recap, the three mics being compared are the CAD M179, Rode NT2a, and Audio-Technica AT3035. Naturally the clips are not in the order listed.

download 20090625-J.wav
download 20090625-K.wav
download 20090625-L.wav

I’ll post the key to the clip identities in a future blog entry. Or if you post your opinion and preference in a comment here or on one of the forums I visit I’ll email or PM the information to you.