In my compositions I often use an instrument to generate line segments that are sent via zak to another instrument.&nbsp; Usually I call this instrument several times in succession to create a more complex signal, basically like a break point line that is being defined with score statements.&nbsp; These instruments are very simple, generally in the form of something like:<br>
<br>intsr 2<br>kout linseg p4, p3, p5<br>zkw kout, 0<br>endin<br><br>instr 1<br>kin zkr 0<br>;do something here<br>endin<br><br>That has been working well for me in the past where I use this technique to make slower moving signals.&nbsp; However, in a recent composition, one with an upbeat tempo and short line segments, I found I was getting discontinuities between the individual score events with this setup.&nbsp; More specifically, the signal drops to 0, only a few samples in length at maximum.&nbsp; Strangely, these discontinuities are not always present depending on where the line segments are started.&nbsp; Using the same events with the same lengths and relative positions but shifted to a different spot gives different results.<br>
<br>My suspicion is that there is a limit the how precise the score statements can be causing rounding errors.&nbsp; Right now I am using positions and lengths with 3 numbers after the decimal point at about 130 bpm.&nbsp; Statements such as the following are typical:<br>
<br>i2 &nbsp;&nbsp; 8.125&nbsp;&nbsp;&nbsp; .625&nbsp;&nbsp;&nbsp; 0&nbsp;&nbsp;&nbsp; .3<br>i2 &nbsp;&nbsp; 8.75&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; .125&nbsp;&nbsp;&nbsp; .3&nbsp;&nbsp; .93<br>i2 &nbsp;&nbsp; 8.875&nbsp;&nbsp;&nbsp; .25&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; .93 1<br><br>If it makes any difference, I am not currently making use of + and . score elements with these particular line generating instruments (I am using another piece of software to generate the events) but the positions and lengths do add up to being back to back with no space in between them.<br>
<br>So, is there a maximum precision for score events as i am suspecting that would be within the ranges I indicated above or should I be looking for a different culprit?<br><br>Thanks.<br clear="all"><br>-- <br>Electronically,<br>
Jeff Taylor