Aha. Thank you. A stupid mistake.
I was not reading the examples closely.

-----Original Message-----
From: Richard Dobson [mailto:rwd@cableinet.co.uk]
Sent: Friday, January 08, 1999 10:41 AM
To: csound@maths.ex.ac.uk
Subject: Re: Carry problems in PPC


The format is misleading.
In the second line:
i  .  + .   < < < <

the first dot applies to the first pfield, which is the instr number
so your input score really is:
i2 0  1 100 5 4 0 1
i. +  .   < < < <
i. +  .   7 6 1 0
e

and the sorted output is exactly as specified.


Richard Dobson


Grant Covell wrote:
> 
> I've been trying to use the carry features in the Mills PPC (MIT Csound:
> 3.493 (Nov 23 1998)
> (Mills/PPC: 4.9.3)).
> 
> A score with:
> 
> i2 0  1 100 5 4 0 1
> i  .  + .   < < < <
> i  .  + .   7 6 1 0
> e
> 
> sorts as:
> 
> w 0 60
> i 2 0 0 1 1 100 5 4 0 1
> i 2 1 1 1 1 53.499999 5.5 2.5 0 1
> i 2 2 2 1 1 7 6 1 0 1
> e
> 
> while I expected:
> 
> i2 0  1 100 5 4 0 1
> i  1  1 100 6 5 .5 .5
> i  2  1 100 7 6 1 0