<HTML>In a message dated 9/28/01 4:31:10 AM Central Daylight Time, dnereson@dimensional.com (Dave Nereson) writes: <BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px"><<This is the area of the piano which requires the utmost in precision. Even an error of 1 cent is audible and does affect the quality of harmony. The tolerance for an error on the PTG Exam in this area is 1 cent.    (B. Bremmer)>>    I read somewhere that the smallest pitch difference the average person (including most musicians and some tuners) can hear is about 3 cents, so when I read tuning articles where a few tenths of a cent are being quibbled over, I just have to wince.  Now if all my customers were concert artists, that'd be different.... </BLOCKQUOTE> Do you really believe that you could tune the midrange of a piano with a tolerance of +/- 3 cents on each note and get paid for it?  It just so happens that I work within that range when tuning my own temperament called the EBVT.  It is designed to produce a specific effect but still remain below the threshold of tolerance that nearly anyone would have. A single error of 1 cent may not be noticed but even in the example that Phil Bondi gave where he said that there was a tendency to make the Ab3-C4  3rd a little too fast, this could make a significant difference.  This is the exact description of a late 19th Century Quasi Equal Temperament which can be found in Professor Owen Jorgensen's big red book called "Tuning". If the error is +1 for C4, the key of Ab (4 flats) will have a more vibrant sound, good for playing Romantic era music and the key of C will have a gentler sound, useful in all music from all eras.  If the error is -1 for C4, the effect will be *Reverse Well* which will be contrary and undesirable to virtually all forms of music from any era. Anywhere in the C3-C5 range, a change of pitch of less than 1 cent will affect the speed of the Rapidly Beating Intervals (RBI). While a single error may not be noticeable, it is the *cummulative effect* of compounded errors or of many random errors that may well produce a very undesirable effect. This is the old argument that "temperament doesn't matter".  When the crowd was gathering outside the Temperament Festival at the 1998 Convention, there was an RPT (who has now quit the business) there spouting off his opinion.  My suggestion to him was that next year, he organize the "Temperament Doesn't Matter Festival". I submit that a high degree of accuracy and purposeful construction of temperament and octave stretch, whatever the goal may be is critical in the midrange portion of the piano.  A tolerance of 1 cent is generous.  I aim for 1/10 cent on my concert tunings. Bill Bremmer RPT Madison, Wisconsin</HTML>