[EMAIL PROTECTED],Internet writes:
>From: Eric Laeuffer <[EMAIL PROTECTED]>
>Subject: Re: star's parameters
>Date: Wed, 03 Mar 1999 15:22:03 +0100
>Eric Laeuffer wrote:
<snip>
>>
>> This solution need to use a delay to avoid deadlock.
>> With this delay the control is delayed by one sample.
>>
>> I would like to avoid this delay.
>> So I think about an other solution.
The delay is not a restriction imposed by Ptolemy. It is mathematically
required in a discrete-time feedback system. There is no way to
avoid the delay. There are a few tricks you can play, but they all
result in something that is not a true closed-loop system.
>>
>> My first star would have just one entry which is the flow of data. The
>> control would be a settable state of this star.
>> And then the second star would change the state of the second one
>after
>> process of the output of the first one.
>>
Even if you did this, there is still a delay in the loop. The settable
state of
the first star is exactly a z^-1 operator. The first star uses the
initial value
of this state for the first sample, then the second star sets the state
based
on those results. The new value of the state isn't used until the
second sample.
If you want to avoid the delay icon for other reasons (e.g. it is buggy
in
your universe), I apologize for telling you what you already know. I'm
afraid
I don't have much help.
---------------------------------------
Telos Systems -- 2101 Superior Avenue, Cleveland OH 44114
Phone: 216.241.7225 Fax: 216.241.4103
http://www.zephyr.com http://www.audioactive.com
----------------------------------------------------------------------------
Posted to the ptolemy-hackers mailing list. Please send administrative
mail for this list to: [EMAIL PROTECTED]
