I had the thought that rather than add something new to the score, it
would be possible to add a commandline flag such as
--score-seed-random=x where x could be a given value or 0 to seed with
current time.

On Sat, Apr 19, 2014 at 11:14 AM,   wrote:
> Sounds reasonable request to me.  Will consider.
>
> Quoting Richard Dobson :
>
>> On 17/04/2014 09:42, Victor Lazzarini wrote:
>>>
>>> Sounds doable.
>>>
>>> May I suggest something to you? You probably know this, but you can
>>> ignore the score completely and do all sorts of algorithmic event
>>> generation from the instr0 in the orchestra, where you would have
>>> access to all of its random number generators (and loops etc.).
>>>
>>
>> Indeed, and that is certainly the next step, but given there already is a
>> random facility in the score, I think a seed option would find a number of
>> uses and, so to speak, complete the facility. To do a run multiple times and
>> choose the one I like the most, etc. I am not doing a mega composition
>> project as such, just a short test/demo sound which the score solution fits
>> perfectly (my sound will end up as an embedded track in a web page), save
>> for the seed issue.
>>
>> Richard Dobson
>>
>>
>>
>> Send bugs reports to
>>        https://github.com/csound/csound/issues
>> Discussions of bugs and features can be posted here
>> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
>> csound"
>
>
>
>
>
>
> Send bugs reports to
>        https://github.com/csound/csound/issues
> Discussions of bugs and features can be posted here
> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
> csound"
>
>
>

That would do; but  ideally I would really like to have a version of the 
seed opcode which returns the new (score) seed value (unless it can be 
arranged to display it directly from the score).  I have done this many 
times in other situations - run a process with a clock-sourced new seed 
each time, but if/when a particularly "good" run manifests, it is rather 
handy to know what the seed is so I can then replicate it.  This is (in 
my case) for experimental R&D, not for guaranteeing fresh results in 
live performance.

Richard Dobson

On 19/04/2014 16:23, Steven Yi wrote:
> I had the thought that rather than add something new to the score, it
> would be possible to add a commandline flag such as
> --score-seed-random=x where x could be a given value or 0 to seed with
> current time.
>
> On Sat, Apr 19, 2014 at 11:14 AM,   wrote:
>> Sounds reasonable request to me.  Will consider.
>>