<HTML><HEAD> <META charset=US-ASCII http-equiv=Content-Type content="text/html; cha= rset=US-ASCII"> <META content="MSHTML 6.00.2800.1106" name=GENERATOR></HEAD> <BODY style="FONT-SIZE: 10pt; FONT-FAMILY: Arial; BACKGROUND-COLOR: #fffff= f"> <DIV>Matthew writes:</DIV> <BLOCKQUOTE style="PADDING-LEFT: 5px; MARGIN-LEFT: 5px; BORDER-LEFT: blue = 2px solid"> <DIV><FONT face=Arial>When I tune the temperament octave (A3-A4), it needs= to be a 4:2 octave, correct?  </FONT></DIV></BLOCKQUOTE> <DIV>No. Read the many replies which said that it should usually be wider th= an that.</DIV> <BLOCKQUOTE style="PADDING-LEFT: 5px; MARGIN-LEFT: 5px; BORDER-LEFT: blue = 2px solid"> <DIV><FONT face=Arial>And one way to test this octave is to play the A two= octaves above the lower note as the test key, to hear the partials in the o= ctave, am I right?</FONT></DIV></BLOCKQUOTE> <DIV>Not exactly, but read Don Rose's comments on ghosting.</DIV> <BLOCKQUOTE style="PADDING-LEFT: 5px; MARGIN-LEFT: 5px; BORDER-LEFT: blue = 2px solid"> <DIV><FONT face=Arial>  If the octave you are testing has no beat wha= tsoever, you have a perfect temperament octave, is this true?</FONT></DIV></= BLOCKQUOTE> <DIV>No. There is no such thing as a beatless octave. An octave which is not= beating at one level, such as 4:2, will be beating at all other coincident = partials, such as 2:1, 6:3, 8:4. The higher the beatless coincident is in th= e chain, the wider the octave. A good compromise octave is usually pretty qu= iet, though, </DIV> <DIV> </DIV> <DIV>Matthew,</DIV> <DIV> </DIV> <DIV>If you have kept this trail of posts on octave tuning, please go back a= nd re-read it, and the links to which you were referred, including the ones = to the AccuTuner manual Appendices F and H. People are happy to spend time h= elping you, but you've got to do your homework and read the replies. At the = risk of repetition, I include, directly below, a copy of my post from last w= eek on this subject:</DIV> <DIV>Bob Davis</DIV> <DIV>-------------</DIV> <DIV>Matthew's original question was how to tune a 4:2 octave. Several peopl= e, myself included, sent the tests, aural and visual. Whether that [meaning = 4:2] is appropriate for the temperament octave on a particular piano is a se= cond question. Tuning so that "the 10th is just noticeably faster than the t= hird" might produce a good width of octave, but it is NOT a 4:2. [It's wider= ] <DIV> </DIV> <DIV>A clean 4:2 octave IS wide at 2:1, and narrow at 6:3. Most aural tuners= naturally gravitate towards a temperament octave that is very slightly wide= of 4:2 ("the 10th is just noticeably faster than the third"), which will be= substantially wide of 2:1 and a tiny narrow of 6:3. This gives an octave th= at is pretty clean-sounding, and produces fifths which are pretty clean and = fourths that aren't too trashy. Any octave size can be divided into 12 equal= half steps. A true 4:2 octave will produce cleaner fourths and more mo= vement in the fifths, and on most pianos will be unnecessarily narow. Howeve= r, on some pianos with high inharmonicity, a wide temperament octave ad= ded to a clean octave below, will produce a double octave that is too n= oisy. It's a balancing act. </DIV></DIV></BODY></HTML>