early reflection of sidewalls: absorption or not?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Member
Joined 2009
Paid Member
Hi Plasnu,
Bob Katz's room is not rectangle and it does have an RFZ at listening position.
I've seen pictures of other mastering room following this principles as well ( some Hidley's design iirc, one in Nashville with RM7 monitors but i can't remember the studio name atm).
That said you are right those are exceptions most mastering studios i've been are rectangle but most of these rooms were huge (for the one i've seen) and speaker placement and room dimensions made for low lateral and vertical ER.
 
Last edited:
I wrote an article comparing two speakers with very different directivity indexes. You can really hear the difference the room sound adds from the wider directivity speaker, as I made binaural recordings of each speaker: KEF LS50 (David) Versus JBL 4722 Cinema (Goliath) Speaker Comparison with Binaural Recordings - Reviews - Audiophile Style

Since both sets of speakers in the room are setup along the long wall, and fire across the short dimension, the are no side wall reflections, as they are several feet away from the speakers. However, on the binaural recordings, one can clearly hear more room with the wider directivity speakers (i.e. KEF LS50) versus the much narrower directivity (JBL 4722) speakers.

While not directly related to sidewall reflections, it can determine ones preference for more direct versus more reflected sound. Which I believe is what is boils down to. A preference neither right or wrong...

In the case of measurements, and specifically an ETC measurement, the rule of thumb is that the first reflection and subsequent reflections should be -15 to -20 dB down from the main peak, within the first 20 to 40 ms of early reflection analysis.

May I humbly suggest in your specific case, the only way to find out is to listen to your setup with no treatments and then with enough acoustic treatment to make a difference, and listen again. Then you can determine your preference for more room sound or more direct sound, wrt sidewall reflections.

I spent +10,000 hours in control room chairs recording and mixing music. I prefer the in-the-corner or in-the-wall mounting of speakers over monitors sitting on the mixer bridge. Gives the cleanest sound minimising early reflections and to my ears produce a cleaner, more coherent sound field or auditory scene. The point I am trying to make is that it is your personal preference that matters most, as there is no right or wrong...
Mitch,

Since this is now an older thread (~two months), I wanted to follow up on your comments. Your setup is in some ways similar to mine: your JBLs have very good horizontal directivity down to about 630 Hz--while the Klipsch Jubilees in the corners of my room have nominally ±90 hor./± 60 vert. degree directivity (broadening below 200 Hz) to below the room's Schroeder frequency of ~100 Hz. My room also has acoustic treatments in the nearfield of the Jubilees to control the 0.7-5ms early reflections from the front three loudspeakers (with center K-402-MEH) that lock in the phantom center imaging and clean up the early arrival reflections that significantly add to the ear's "confusion", thus making the resulting soundstage extremely clean, clear, and non-fatiguing.

Your review of the KEF LS50s is also directly shared by a good friend that also owns Jubilees and recently bought a pair of the KEFs to put them in-room in front of and inside of his Jubs (which I thought was a big departure for his tastes and interests).

  1. So the the motivational factor in writing this reply is the huge difference in directivity that these two loudspeaker types produce and the two completely different listening experiences that result. Additionally, the loudspeakers are placed in-room fairly differently with the KEFs avoiding reflections from 0.7-5 ms by elevating stands and placements away from the room boundaries, while the JBLs are supporting nearfield reflections (albeit way down in SPL from the direct arrival SPL) in what looks to be 3-4+ms time delays. The resulting subjective differences in sound, especially with certain recordings having more or less ambiance in the recordings themselves, is huge. The dimensions of the soundstage, the clarity and "micro-imaging", and the differences in timbre will be a large shift from one to the other (even found just using your on-line binaural files in your referenced article).
    _
  2. I would certainly concur with your recommendation that the only way for the OP to determine acoustic treatments to taste (and it is taste that we're talking about, not setting up loudspeaker-room acoustic performance to measurable criteria) is to listen to a bare-room case, then add absorption and diffusion to taste. This is due to the more extensive wide-band "second room" reflections from the very wide and uncontrolled polars from the KEFs, and to a much greater extent, controlled reflections above 630 Hz with the JBLs. I would add that there is as much to be gained in directivity going from 100-630 Hz as there is from 630 Hz to about 10 kHz (above which frequencies basically don't reflect without significant loss in SPL).
    _
  3. I would add that this is basically mimics the dichotomy between studio monitoring facilities vs. home sound reproduction system, as illustrated in the writings of Toole, et al. This is apparent even the tastes of mixing and mastering engineers wanting much fewer in-room reflections in the studio for hearing more clearly in the studio than in their own homes. Did you catch the recommended RT values for the "Moulton Labs" small control-room layouts that the OP said he wanted to mimic:
    3. The reverberation time should be roughly the same at all frequencies between 60 Hz. and 8 kHz., and it should be quite short (ca. 80 to 200 ms.).
    This is about half the RT values than is common in home stereo listening rooms-and in my experience, involves a large amount of absorption surface area and sufficient depth of absorption panels materials to achieve this. I don't believe that I'd like this in my listening room with full-range directivity loudspeakers. In fact, I'd say that this would definitely kill all room characteristics that exist and that lead to the pleasant long-delay room reflections that extend outside the Haas interval of hearing, and that increase the perceptions of envelopment.
Overall, I have to say that starting with full-range controlled directivity loudspeakers in room greatly clarifies and simplifies the entire acoustic treatment process (over direct-radiating loudspeakers having no inherent directivity other than cone diameter)--with suppression of 0.7-4 ms early reflections being the most effective to achieving soundstage and phantom center imaging being the prime requirements (in terms of adding room treatments).

In the case of integrating into the room direct radiating loudspeakers of the "DIY bending-wave" (a.k.a., distributed-mode loudspeakers) that behave more like dipoles than monopoles, I really don't have many words of wisdom, other than, perhaps "I hope that the soundstage that you get from them pleases you...because I find it a bit lacking in terms of resulting sweet spot size and sensitivity to in-room positioning.
 
Mitch,
In the case of integrating into the room direct radiating loudspeakers of the "DIY bending-wave" (a.k.a., distributed-mode loudspeakers) that behave more like dipoles than monopoles, I really don't have many words of wisdom, other than, perhaps "I hope that the soundstage that you get from them pleases you...because I find it a bit lacking in terms of resulting sweet spot size and sensitivity to in-room positioning.

My design does not quite fit your description; It has only a diameter of 12cm and needs a woofer below 200Hz (close to the Schröder frequency). I did not make full polar plots but on-axis and 30º looks almost identical and 60º is only starting dropping above 7kHz. So it has wide dispersion and that is the reason I am discussing early lateral reflections.
I wanted a speaker not with a sweet spot but a sweet area!
 
Last edited:
@Chris, thanks. The JBL 4722's have constant directivity down to 400 Hz according to their specs and I have reviewed both the Kii THREE and Dutch and Dutch 8c, both of which have measured constant directivity down to 100 Hz using cardioid tricks. I still prefer my JBL's, but I also use DSP...

@Monteverdi As Chris pointed out, I missed your RT60 requirement as mentioned by Moulton Labs. Have you ever been in a room with this short reverberation time? Also, have you worked out the Sabines required to get that short of a RT60 in your size room?

Just curious to ask as I have been in a few really "dead" control rooms and personally I did not like the sound as it sucked the life out of the music. My room has an awful room ratio, but a smooth reverb time across the band at 600ms which is the upper limit for my size room, with the lower limit around 400ms 30ft x 16 x 8. My personal preference is for a livelier room. For your size room and getting under 200ms is going to require a lot of absorption... and cost plus you may not like it if you have never been in that dead of a room before... Thoughts?
 
I remember that I measured a RT60 between a little over 500ms (but not as smooth as I wanted to) for my previous room which was almost twice as large than I am building now and I thought it was a good balance. I feel that RT60 for a room is a flavor choice. Primarily I want to listen the music but also to appreciate the space one is in acoustically.
 
IMO (at moment :)), taste operates the differential, and Rt60 is a global parameter.

RT60 = 0.6s @ 500Hz octave + RT60 = 0.3s @ 2KHz octave, appears to me more "dry", compared to RT60 = 0.3s @ 500Hz octave + RT60 = 0.3s @ 2KHz octave.

Anyway, in my room, with RT60> 0.6s, I moved to nearfield or headphones, for complex music, no matter if orchestra or progressive rock.
 
I don't intend to turn this into a discussion of reverberation time (RT), but I think that it's worthwhile to firm up the use of terminology a bit in this section of the forum. The application that I generally use (REW) doesn't actually calculate RT60, but rather a succession of other values: EDT, Topt, T20, T30, TS. In general, I find that most people are probably talking about T30 or perhaps T20 rather than T60 (for reasons that should be obvious in home-sized listening rooms).

One of the requirements listed in the Moulton Labs small control room layout is that "the reverberation time should be roughly the same at all frequencies between 60 Hz. and 8 kHz". I believe that they are probably referring to T20 or T30, (i.e., not really RT60). Bicicletta mentioned a varying reverberation time curve (0.6s at 500 Hz, 0.3 and 2000 Hz), which actually doesn't conform to the RT requirement from Moulton's web page. So the discussion of non-uniform RT values is seemingly outside of the constraints listed by the OP.

Additionally, I find that there is one related measurement that stands out as sharply varying from loudspeakers having good directivity control and those having essentially little directivity other than that of the diameter of the direct radiating diaphragms vs. frequency. That measurement seems to be EDT, or "Early Decay Time".

As an example below, a plot of the various RT curves taken from my listening room is provided whereby the microphone is 1m in front of the center K-402-MEH, which itself has controlled directivity to below 100 Hz:

790547954_ChrisAsListeningRoomRT60-1MetreCenter.jpg.fd34b1a2aef42c92ed8d388eba9ac0f1.jpg


The RT plots are also typical of the measurements of the two Jubilees on each side of the MEH. In particular, I've noticed that the EDT curve always falls away from the other RT curves in a fashion that seems to conforms to the 1/(directivity index), while measurements from other rooms using DR loudspeakers looks more like the other RT curves, i.e., like the EDT curve doesn't have a "1/DI" characteristic. This fast EDT decay time also seems to be reflected in the control room practices using lots of absorption to achieve the 0.2 RT values that were recommended in the Moulton control room setup requirements (thus largely negating the effects of side-wall reflections).

My observation is this: would it not be roughly equivalent to use higher DI loudspeakers with wide-band directivity than to use tons of absorption to achieve a T20 of T30 of 0.2 in a small control room environment, rather than specifying an RFZ (reflection-free zone) that's recommended by D'Antonio, or "Ambient Ambechoic" (see pg. 13)? One of the quotes from the REW help file on this subject: "A fast early decay (low EDT figure) indicates better clarity than at positions where the EDT is higher."

Additionally, the requirement in the Moulton Labs list for such a low RT value (pick one RT curve type that you're talking about) seems to be an integral part of the formula that they are specifying, i.e., it is actually a non-tradable requirement having higher precedence than most of the other requirements listed. If you decide that you're going to use a much higher RT in-room than Mouton recommends, then you've basically negated the entire concept that is being proposed by Moulton.

Chris
 
Last edited:
Chiming in here from my experience. I've looked previously into the research and from my own experience adding damping on the wall behind the speakers has always been beneficial in terms of more detail and less background noise. I have DIY speakers with wide dispersion designed to be close to the wall using it as bass reinforcement and heavily toed in. But WAF is a factor so now I just have some damping panels behind speakers.

I have got less impact on sound by treatment of the sidewalls so for now I keep them as is. Room is a small living room with carpet, fabric sofa, pillows, plants etc and stuff on the walls (paintings, decorations etc). I would like more damping bending speakers but it is always a compromise.
 

Attachments

  • IMG_3390.jpg
    IMG_3390.jpg
    698.7 KB · Views: 143
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.