13.0 Listening Rooms and Houses

13.1 How should I place speakers in my room? What size room is best?
        You are after two important, distinct goals: flat frequency
        response and good three-dimensional image. At your disposal is
        the room size, the room shape, speaker height, speaker
        placement, listening position, and room treatments. Even though
        good speakers are essential to good sound, room effects are also
        extremely important. In many cases, the differences in room
        effects will be more noticeable than spending twice as much on
        speakers!

        Here are some generally-accepted-as-good guidelines for good
        sound.  If you use these as a starting point, you will be far
        ahead in terms of getting good sound from your speakers and
        room.  But these are just a guide.  Each room and each speaker
        is a little different.  Experiment to see if a change will
        help.  Also, if the manufacturer recommends something
        different, give that a try, too.  Then use what sounds best to
        you.

        For smoothest bass response, a listening room should be as large
        as possible, have dimensions as unrelated as possible, and
        should be optimally damped. Although nothing is ever ideal,
        there are a few room dimension ratios that are better for
        listening rooms:
                Height          Width           Length
                1               1.14            1.39
                1               1.28            1.54
                1               1.6             2.33
        If your room isn't shaped like that, don't worry. These
        effects are not major.

        Also for smooth bass response, woofers should be at distances
        from the nearest three room boundaries that are as different as
        possible. In some cases, the line dividing the listening room
        into left and right halves must be considered a room boundary.
        Also, for smooth bass response, the listener's ears should be
        at distances from the nearest three room boundaries that are
        as different as possible.

        All of this is essential because a wall near a speaker boosts
        the bass from that speaker at some frequencies. If a speaker
        is the same distance from three walls, then some frequencies
        will be emphasized much more than others, rather than slightly
        more.

        For best three-dimensional image, a listening room should have
        good symmetry about the plane between the two speakers. This
        means that if one speaker is in a corner, the other speaker
        must be in a corner. If this symmetry is not right, the first
        reflection from the wall behind one speaker will be different
        from the first reflection from the wall behind the other speaker
        and critical parts of the stereo signal will be damaged.

        Also, no large object should block the path from speakers to
        listener or from speaker to speaker. Speakers should be
        elevated so that tweeters are at listener ear height. The
        distance between speakers should be no greater than the distance
        from each speaker to the listener. Finally, the tweeters should
        be aimed at the listeners.

        A normal box-shaped listening room with bare walls will have
        "slap echo" which will reduce intelligibility. A good cure is
        randomly-placed wall hangings consisting of small rugs spaced
        an inch or so away from the wall to increase sound absorption.
        Another cure is convex-shaped art objects on the walls to
        disperse harmful reflections. If money is available, commercial
        room treatments such as "Tube Traps" and "RPG Diffusers" are
        also valuable, but many of the benefits of these exotic devices
        are available with simpler techniques.

        As a general rule, in a good room, speakers and listener can be
        close to room boundaries with minimal adverse effects. In a bad
        room, a good strategy is to place both speakers and listener as
        far away from room boundaries as possible.

        An excellent starting point for speaker placement is to measure
        the listening room diagonal dimensions. Divide that measurement
        by three. Put each speaker that distance from a corner, on the
        room diagonals.

                I----------------------------------I
                I                                  I
                I                L                 I
                I                                  I
                I       S                 S        I
                I                                  I
                I----------------------------------I

        Place your listening position midway between the two speakers
        and approximately half way from the speakers to the wall. Be
        sure that there is nothing in the "triangle" formed by the
        listening position and the speakers.

        Try this and then move things 12" (30cm) at a time to see if
        you can improve the sound. Your ears will be a better guide
        than any commonly-available instruments. To keep track of
        what you are doing, take notes. To remember exactly where
        you put the speaker on the floor, a practical trick is to
        mark the floor with a sewing needle and thread.

        Some speakers want to be aimed right at the listener (toed in)
        while others work best pointed straight ahead.  Experiment.

13.2 How do I wire a house for sound?
        A fundamental principle of physics is that the farther a signal
        travels, the more the signal will be degraded. Translate this
        to mean that the shorter the wire, the better. Understanding
        this, the idea of running speaker cable between every room of
        the house isn't as attractive as it first seems.        

        If you still decide to wire your house for sound, you should do
        it at the same time you're wiring for telephone and electricity.
        It is possible to wire a house after the walls are closed, but
        it becomes very difficult.

        It is economical to use common house wire (Romex, UF, NM, etc)
        for speaker wire in the walls, but this may violate building
        codes. Check with an electrician or inspector first. It will
        also confuse future electricians, so label the wire clearly, all
        along its length.

        If you want to make your house like a recording studio, it is
        best to use the techniques of recording studios. When studios
        run long lengths of sound cable from one room to another, they
        drive the cable with 600 ohm line amplifiers. They also use
        shielded, twisted-pair cable. They only connect the shield at
        one end of the cable. Finally, they use balanced inputs at the
        other end of the cable.

13.3 Where can I read more about listening room construction and tuning?
        "Building a Recording Studio" by Jeff Cooper
                Mix Bookshelf
        "Handbook for Sound Engineers"
        "The Master Handbook of Acoustics" by F Alton Everest
        "Sound Engineering 2nd Edition" by Don and Carolyn Davis;
                Howard W. Sams & Co. (C) 1990
        "Good Sound" by Laura Dearborn
                Introductory, but clear and accurate
        "Sound Recording Handbook" by John M. Woram
                Howard W. Sams & Co. #22583
                Excellent General Reference
        "Audio Technology Fundamentals" by Alan A. Cohen
                Howard W. Sams & Co. #22678
                Overview of Audio Theory
        "Introduction to Professional Recording Techniques"
                by Bruce Bartlett
                Howard W. Sams & Co. #22574
        "Modern Recording Techniques" by Hubar and Runstein
                Howard W. Sams & Co. #22682
        "Sound Studio Production Techniques"
                by Dennis N. Nardantonio
                Tab Books
        "The Uneasy Truce Between Music and the Room"
                F. Alton Everest
                Audio, February 1993, Pgs. 36-42
        "Coloration of Room Sound by Reflections"
                F. Alton Everest
                Audio, March 1993, pgs. 30-37

13.4 What is white noise? What is pink noise?
        "White noise" is characterized by the fact that its value
        at any two different moments in time are uncorrelated.
        This leads to such noise having a flat power spectral
        density (in signal power per hertz of bandwidth), and is
        loosely analogous to "white light" which has a flat power
        spectral density with respect to wavelength.

        Pink noise has flat power spectral density per PERCENTAGE
        of bandwidth, which leads to a rolloff of -3 dB/octave
        compared with white noise.

        There are many reasons for using pink noise in audio testing.
        One is that music has an average spectral content much closer
        to pink noise than white noise. Another is that pink noise
        can be readily measured with constant Q bandpass filters and
        naturally leads to flat plots on logarithmic frequency scales
        - which correspond to the equally tempered musical scale.

        Pink noise is often used with 1/3 octave band filters to
        measure room acoustics. This idea has merit since 1/3 octave
        is a convenient number near the limit of our ears ability to
        detect frequency response irregularities, and because
        averaging measurements over 1/3 octave bands smooths out the
        numerous very narrow peaks and dips that arise due to
        standing waves in rooms.

        Another term you'll hear about is Gaussian noise - this is
        noise with a Gaussian amplitude probability density.
        Gaussian noise has the amazing property that linearly
        filtering it preserves its Gaussian amplitude density and
        that sums of Gaussian random variables are again Gaussian.
        The two terms shouldn't be confused. It is possible to have
        Gaussian white or pink noise.

COPYRIGHT NOTICE
The information contained here is collectively copyrighted by the
authors. The right to reproduce this is hereby given, provided it is
copied intact, with the text of sections 1 through 8, inclusive.
However, the authors explicitly prohibit selling this document, any
of its parts, or any document which contains parts of this document.

--
Bob Neidorff; Texas Instruments     |  Internet: neido...@ti.com
50 Phillippe Cote St.               |  Voice   : (US) 603-222-8541
Manchester, NH  03101 USA

Note: Texas Instruments has openings for Analog and Mixed
Signal Design Engineers in Manchester, New Hampshire.  If
interested, please send resume in confidence to address above.

14.0 Recording
        There are more different recording systems available today than
        ever before. Digital and analog are both available to the
        consumer. With the advent of consumer digital recorders, used
        pro analog recorders are becoming available for surprisingly low
        prices. Now may be the time for you to buy a microphone and
        recorder and make your first!

14.1 What is DAT? What is its status today?
        DAT (Digital Audio Tape) is currently the standard professional
        digital format for 2-track digital recording. DAT had a
        short-lived consumer presence, but never "made it". As digital
        recorders have no tolerance for clipping, using a DAT recorder
        takes a slightly different knack. The results can be worth it,
        however, as DAT format offers the same resolution and dynamic
        range as CDs. DATs record for up to 3 hours on a tape, and can
        run at three different sampling rates: 32 kHz, 44.1 kHz
        (for CD), and 48 kHz (the DAT standard).  Longplay mode cuts
        frequency response to 14kHz but adds even more recording time.

14.2 What is DCC? What is its status today?
        DCC is Philips' attempt to modernize the regular cassette. DCC
        decks can play analog cassettes, and can record new Digital
        Compact Cassettes. They use stationary heads (DATs use rotary
        heads as do VCR's), and although they are digital, they use
        lossy compression to fit all the data on the cassette. Although
        DCC sound quality is far better than the 1960 standard cassette,
        the DCC does not have the sound quality present in DAT or CD.
        DCC may be a good choice for consumers who want to assemble mix
        tapes for cars or walkmans, but is not suitable for any
        professional applications.

        As of October 1996, DCC is quite affordable in price.  Some
        DCC home recorders are under $200.  However, blank DCC tapes
        are still hard to find and fairly expensive ($10 each for 90
        minute lengths).  Also, DCC manufacturers are dropping DCC
        from their lines, indicating that it is either on the way
        out or never made it in.

        Although the ability to play analog cassettes is a strong
        advantage of DCC, many people have had trouble with oxide
        particles falling off analog cassettes and clogging the gap
        of the DCC head.  This may be due to the extremely low
        quality of some analog cassette tapes and may be due to the
        very tiny gap of DCC heads.

        Caution: NEVER demagnetize DCC heads.  This will permanently
        damage the heads.

        As of May 1997, Philips has announced plans to discontinue DCC.

14.3 What about writable compact discs? What is the status today?
        Recordable and rewritable CD recorders and discs are available,
        and costs are dropping.  As of Dec 2003, recorders have shown
        up for <$30 and blank disks are advertised as low as $0.25
        each in bulk.  Many people report destroying many disks before
        getting their machine working correctly, but once people learn
        the software and hardware steps, archival CDs can be made
        inexpensively and routinely.  There is definitely a difference
        in discs and a difference in recorders.  However, it is tough
        to generalize on which are better or worse other than to say
        that name brand discs are a safer bet than off-brand discs.
        For more on CD-R read this excellent document:
        http://www.fadden.com/cdrfaq/

14.4 What are Dolby B, C, and S, HX Pro, and DBX? Are they compatible?
        Dolby B, C, S, and DBX are techniques for increasing the
        signal/noise ratio of recordings. All work in similar ways:
        they compress the dynamic range of the sound during recording,
        then expand it back upon playback. As much as we would like
        it to be otherwise, you only get correct reproduction if you
        use Dolby B to play back a Dolby B tape. Same for Dolby C,
        Dolby S, and DBX. Dolby HX Pro is the exception.

        Dolby B works mostly with higher frequencies; it increases
        their levels during recording and decreases their levels, and
        the levels of high-frequency noise such as tape hiss, during
        playback.

        Dolby B tapes can be played back without Dolby B processing,
        but high frequencies are over-emphasized and the sound will
        be excessively bright. This can be compensated for to some
        extent by turning down the treble control. Audio novices
        often remark that commercially recorded tapes recorded using
        Dolby B sound dull when played back with Dolby B; this is
        because they are accustomed to the boosted high frequencies
        they hear when playing these tapes without Dolby.

        Dolby C achieves greater noise reduction (about 8-10 db) than
        Dolby B by working with a greater range of frequencies and
        altering relative levels more; this means that playing Dolby C
        tapes back with no Dolby processing or with Dolby B, leads to
        very bad frequency response and a sound that most people find
        unpleasent. Dolby C may also be more sensitive to variations
        among decks in exact frequency response, alignment, etc. Some
        people find that tapes recorded using Dolby C sound best only
        when played back on the deck on which they were recorded.

        Dolby S works with an even broader range of frequencies than
        Dolby C, and achieves slightly greater noise reduction. Its
        has three advantages over Dolby C: (1) many people find that
        tapes recorded and played back using Dolby S sound closer to
        the original than tapes done using Dolby C; (2) tapes recorded
        using Dolby S don't sound awful if played back on Dolby B decks,
        and (3) Dolby S seems to be less sensitive to variations among
        decks.

        DBX is similar to Dolby B, C, and S, but uses the same compression
        on all frequencies, high and low. However, DBX is mostly used
        in the professional market. Very little home DBX equipment is
        available, and some of that home equipment is no better than
        comparable Dolby B home systems. All DBX systems are compatible
        with all other DBX systems, but incompatible with Dolby. A DBX
        tape will sound terrible without DBX processing during playback.

        All compression/expansion systems suffer two problems. One is due
        to the fact that compressors can't compress a loud signal before
        they have heard a bit of it, so that little bit of loud signal
        will get through uncompressed. Likewise, quiet passages will not
        be expanded until after they are detected. These delays give rise
        to an audible problem often called "breathing".

        The other problem inherent in all compression/expansion systems
        is that if there are any frequency response errors in the tape
        recorder, they will be made worse by the compression/expansion.
        For example, if there is a 2dB dip in frequency response at 1kHz
        in the tape recorder, this will be accentuated to a 4dB dip if
        the compressor is using a 2:1 ratio. So compression/expansion
        trades noise for frequency response error. For that reason and
        the previously mentioned breathing, some people prefer to use
        their recorder without any noise reduction at all. They prefer
        a bit of noise to the other errors.

        Dolby HX Pro is not noise reduction and does not use
        compression or expansion. HX Pro is a technique developed by
        Dolby Labs to increase tape headroom by decreasing the bias
        when recording signals with a large high frequency component.
        This allows better transient response, particularly on less
        expensive tapes, and requires no processing when the tape is
        played back. Dolby HX tapes can be played back on any system
        with no decrease in quality.

        Dolby Corporation has developed other techniques and other
        acronyms for products related to surround sound.  The phrase
        "contains Dolby" isn't as meaningful today as it used to be.

14.5 What is the best cassette deck under $400?

14.6 What is PASC? Can I hear the effects?
        PASC (Perceptual Audio Sub-band Coding) is a data-compression
        algorithm. It increases the length of recording that can be
        stored in a given number of data bits by eliminating sounds that
        the developers' research claims can not be perceived by human
        listeners. Its most important component is the omission of
        quiet sounds that occur at the same time and near the frequency
        of louder sounds. It provides up to a 4x increase in the length
        of recordings a given digital medium can hold; this is essential
        to allow full-length digital recordings on DCC (and on MD, which
        uses a different compression technique). It is not necessary
        to translate CD data to analog before compressing it using PASC,
        nor the reverse.

        It is very difficult to hear any degradation from PASC, but it
        is possible, depending on the source and listener.  The effect
        is not a distinctive noise (like a hiss) nor a consistent
        diminution (like a notch in a speaker's response), but a broad,
        uncorrelated dropout in a changing collection of sounds that
        are masked by sounds that you can hear very easily.

        Since it is lossy, repeated PASC recording will cause
        progressive loss, and this signal damage may become easily
        noticeable. This is a side effect that recording companies
        hope will have the effect of discouraging piracy via DCC.
        DCC recorders do have digital inputs so can make one perfect
        copy of a master, but copy protection prevents digital
        duplication of a copy.

        For more information on audio compression, consult these
        articles (courtesy of Jonas Palm):

        R. Veldhuis, M. Breeuwer, R. van der Waal, "Subband Coding of
        Digital Audio Signals  Without Loss of Quality,"  IEEE ICASSP,
        1989, pp. 2009-2012.

        J. Johnston, "Perceptual Transform Coding of Wideband Stereo
        Signals," IEEE ICASSP, 1989, pp. 1993-1996.

        G. Davidson, L. Fielder, M. Antill, "High-Quality Audio Transform
        Coding at 128 kbits/s," IEEE ICASSP, 1990, pp. 1117-1120.

        J. Princen, A. Bradley, "Analysis/Synthesis Filter Bank Design
        Based on Time Domain Aliasing Cancellation," IEEE Trans ASSP,
        Oct. 1986, v. 34 n. 5, pp. 2161-2164.

        P. Duhamel, Y. Mahieux, J. Petit, "A Fast Algorithm for the
        Implementation of Filter Banks Based On 'Time Domain Aliasing
...

read more »