The rule of thirds

September 6, 2011
 by Paul McGowan

Many of my readers have asked for some setup help in these posts and what works for me is to simply throw one in here and there as the mood strikes.

Perhaps one of the most fundamental setup procedures is speaker and listener placement in any given room. You have to have a starting point when you setup a system and using the rule of thirds, originated and popularized by one of the best in the industry Harry Pearson (HP a true high-end treasure), is by far the best.

Our goal will be to achieve a disappearing actin your living room where you can’t pinpoint the location of the left and right loudspeaker when they’re playing music (unless you’re like the customer I wrote about yesterday).

Using the rule of thirds is simple: place your loudspeakers one third the total distance of the room from the rear wall and your listening position the same one third away from the opposite wall. For example, let’s say you have a room that’s 15 feet long and 12 feet wide. The loudspeakers would be placed 5 feet from the rear wall and your couch would be 5 feet out from the opposite wall. That’s it.

Remarkably simple and surprisingly close to what you will wind up with in the end. Now what’s left if for you to make minor tweaks – how far apart are the speakers, how much toe in, etc.

We’ll cover more specifics in a future post.

Subscribe to Paul's Posts

15 comments on “The rule of thirds”

  1. An excellent focus for dialog. However, my I humbly suggest that we look at differentiating the difference between music and sound? Pleasing music is a wonderful experience but a world without sound is a bleak existence to be sure.

    I cannot live without sound, wind, birds, rain, kids, tone, voices, cars, everything that goes into the manifestation of enhancing my life experiences and of course the choices I make to fine tune my emotions from time to time in my chosen and manipulated listening environment.

    Distance between speakers recedes into more of an interesting phenomenon rather than a goal. Here today, there tomorrow…, somewhere else next week or next year….

    Most important is how I implement the ambience to monitor my environment during moments of calm, stress, productivity and rest and relaxation.

    Sound is similar to having a magic wand at my disposal…..

    1. I have very little experience with REL because I am a servo bass fan. However, people I trust tell me only good things about the REL and I believe them. If I had to choose a subwoofer today, I would probably choose a REL.

  2. Perhaps if the only goal is the “disappearing act”, then the rule of thirds is best. For anything approaching a full range speaker in a moderate size room, it seems important to take room nodes into consideration in order to get the most balanced bass response possible. In an idealized room, minimum node excitation is achieved at the 25% and 75% positions (Sound Reproduction, page 221. So surely achieving the most satisfying sound possible is more complex than this simple rule?

    1. Indeed you’re correct. However, I take a different approach. The disappearing act to me is critical over everything else. So I start with the rule of thirds, I tweak to get it to work well with a balanced collection of music and get it as tonally correct as I can (the methods of which I’ll cover in an upcoming post). But room modes will almost always kill the bass, even with a “full range” loudspeaker (of which there aren’t really any).

      The trick, if you’re all in, is to add subwoofers. Subwoofers are separate boxes and they can be placed where it’s best for bass, allowing the main speakers to be placed where it’s best for imaging. One without the other isn’t interesting to me personally.

  3. I’m in total agreement with you Paul about using subwoofers to allow optimizing bass separately from other parameters. The challenge then becomes achieving seamless integration between the subs and main speakers. I wasn’t satisfied with the results I was able to achieve until I started using digital crossovers and digital room correction. This opens up options like 100th order crossovers and (pretty much) perfect phase and output alignment. Add to that the attenuation of many room nodes and substantial control over frequency response and I’d never go back. Between using subwoofers and forsaking bit-perfect playback, I imagine my audiophile card has been canceled. Fine by me.

    1. In my experience, only servo based subwoofers (of which there are very few) can integrate seamlessly. I think Velodyne (some models) and Martin Logan Descent’s have servos and not sure what else does. Without an active servo, the woofer in the sub can never match the upper drivers and even if it could, the overhang of the driver’s mass would still be a problem.

      1. I’ll seek out some servo-controlled subs to try. The last ones I’ve heard were the Entecs way back in the 80s – I remember being impressed. Looking at the Velodyne DD series, I see they have servos, room correction features, fine-grained phase adjustment, and relatively steep (8th order) crossovers. They seem to address the features we both have found useful in achieving good integration.

      2. Any good sub (of which there aren’t many around it has to be said) can be integrated well. Not just ‘servo’ subs. I don’t believe the technology has much to do with it.

        Integration is difficult though just doing things by ear, especially in a two channel setup. Where do you put the crossover? What slope do you use? These two questions can be answered much more easily if you have a good measurement system.

        I think most of the reason people shy away from subwoofers is that the integration is hard to achieve without the right tools.

        Most if not all of the systems I have calibrated have benefited from a subwoofer. Often times they can be used in a mode canceling arrangement to remove a peak that could otherwise only be removed by EQ.

        1. I’ll agree with you on most of this but not on the servo part. The toughest integration is between the top end of the sub and the low end of the speaker when the speaker is a quick one (like a panel or electrostat of some kind). This is because the large mass of the subwoofer cone can’t keep up with the speed of the panel woofer. A servo fixes all that.

  4. Thank you, Paul. I look forward to reading your posts when I arrive at work each morning.

    Have you heard REL subwoofers? In my experience, these subwoofers integrate far better than any subwoofers I’ve heard. They are the most musical as well. Perhaps they are servo based, and I’m just not aware of it. When using the high level input, they can be dialed in to blend seamlessly with any speaker.

    Do you have experience with REL? I’m extremely interested in your opinion of them.

  5. George Cardas disagrees with the rule of thirds:

    Setting Up Speakers In A Rectangular Room

    by George Cardas

    Very precise speaker placement can open up a whole new dimension in listening, so I will outline the system that is becoming the standard of the industry. This standardized listening room is a Golden Cuboid and is the model for the math used in this system. This method will work with any box speaker, in any reasonably sized rectangular room. You may find that you have already positioned your speakers this way by ear.

    Active nodes are the main concern when placing speakers in a rectangular room. A node, or the frequency where speakers and parallel walls interact, is proportional to the speaker to the wall distance.

    The three most importance nodes, in order of importance, are proportional to the distance between the speaker and:

    1. The side wall nearest the speaker
    2. The rear wall
    3. The side wall across from the speaker

    A secondary factor is the speaker-to-speaker time constant.

    When you use this Golden Ratio method to set your room up, the speakers are placed so the three nodes progress or differ from one another in Golden Ratio. This eliminates any unison or near unison resonance in the nodes.

    Panel or dipole speakers such as Apogees and Magnepans cancel their side waves, so a formula of .618 x the ceiling height can be used for determining placement from the rear wall. Most box speakers radiate low frequencies in all directions thus a formula that places the speaker to rear wall distance at 1.618 the side wall distance should be used.

    Speaker placement, simply stated

    The distance from the center of the woofer face to the side walls is:

    Room Width times .276 (RW x .276)

    The distance from the center of the woofer face to the wall behind the speaker is:

    Room Width times .447 (RW x .447)

    This is all you need to know to place speakers in a symmetrical, rectangular room!

    Diagram A
    Distance Percentage
    Speaker to side wall: RW x .276
    Speaker to rear wall: RW x .447
    Speaker to opposite side wall: RW x .724
    Speaker to speaker: RW x .447

    For those who must know more…
    CAUTION: THIS MAY MAKE YOUR HEAD EXPLODE!

    Room Setup Guide PDF (196Kb File)

    There is even a calculator for di-pole set up at the bottom ot the page, here:

    http://www.cardas.com/content.php?area=insights&content_id=26&pagestring

Leave a Reply

Stop by for a tour:
Mon-Fri, 8:30am-5pm MST

4865 Sterling Dr.
Boulder, CO 80301
1-800-PSAUDIO

Join the hi-fi family

Stop by for a tour:
4865 Sterling Dr.
Boulder, CO 80301

Join the hi-fi family

linkedin facebook pinterest youtube rss twitter instagram facebook-blank rss-blank linkedin-blank pinterest youtube twitter instagram