Room Treatment - Bassy or Boxy?

I had an 'ah-ha' moment yesterday as I was repositioning panels. Yesterday I thought the sound was bassy. I found a broad peak from about 55 to 65 hertz, and re-read a lot of the room treatment articles. Ethan WIner came through, again, with this article about density tests. First, I noticed that the panels had the foil side pointed in to the room. I don't know why this helps bass absorption, but according to the specifications it helps greatly. So I turned several of the traps over foil side in. His article also provides an explanation for the broad bass peak I observed:

...the absorption widens the bandwidth (lower Q) of those resonances. In my experience, this feature of bass trapping is at least as important as flattening the low frequency response and reducing modal ringing time. With a wider bandwidth the peaks and nulls are audibly less intrusive - instead of individual bass notes sticking out like a sore thumb, a much broader range is emphasized.

This explains the broadness of the bass peak. I think the room has a natural resonance at about 60 hertz, but the bass traps are working and the peak is now broader. But that wasn't the 'ah-ha' moment.

As I was lifting a large box that held a carpet, I realized that the recording made everything sound like it was in a big cardboard box. 'Boxiness' is a well known way to describe an unnatural peak in response somewhere between 400 and 800 hertz. This is a far cry from boominess - but I don't have the ears of a mastering or mixing engineer. I'm sure this would have been immediately obvious to a recording professional.I recorded some voice and guitar and scanned the playback with a Single Band Parametric plugin, another good one from Steve Harris. A parametric filter can artificially raise or lower slices of the audio spectrum. It controls the width of the slice as well as the frequency and direction. I scanned from about 300 up to 800 hertz, using a very high Q; in Steve's control the Q is set by adjusting the bandwidth - wider bandwidth equals lower Q. I used a tenth of an octave width, boosted 20 or more decibels and listened to the effect. There were several areas of resonance. I'm not sure which to focus on, but for room treatment I don't think it matters. I guess that the brick, glass, and wood paneling is causing various reflections above 300 or 400 hertz, and I need to control them. Since I can't cover the windows, which account for 15 to 20% of the wall space, let's focus on the brick and wood. Which is more reflective?

I think the reflectiveness would vary inversely with absorption. StudioTips has a list of absorption coefficients, and unpainted brick is .03 at both 250 hertz and 500 hertz. A low number means the material does not absorb as well as one with a high number. Acoustic mineral wool, for example, will have a coefficient near 1. Wood paneling is 0.19 at 250 hertz and 0.14 at 500 hertz. So it looks like the brick work is more guilty of reflection.

Back to work with the acoustic panels!