Hi all,

 From the commit log:

2005-03-13    Istvan Varga    
    * Engine/fgens.c:
    * H/ftgen.h:
      added GEN51
    * Opcodes/reverbsc.c:
      minor code improvements

2005-03-13  John ffitch  
    * Opcodes/ftest.c (gentune): Added Gabriel's gen to the loadable
    gens under the name gentune


Are we going to have both the loadable op and the Gen51?

steven


-------------------------------------------------------
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now.
http://ads.osdn.com/?ad_id=6595&alloc_id=14396&op=click
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

steven yi wrote:

> Are we going to have both the loadable op and the Gen51?

I added the numbered version before noticing that there is already
the named gen. I do not object to removing GEN51 and keeping
gentune only, although having GEN51 allows for better compatibility
with other Csound versions that do not support named loadable GEN
routines.


-------------------------------------------------------
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now.
http://ads.osdn.com/?ad_id=6595&alloc_id=14396&op=click
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

Makes sense to me to keep the numbered version for the sake of greater 
compatibility.  I don't like the idea of the redundant code, so any 
objections to removing the loadable version?

steven

Istvan Varga wrote:

> steven yi wrote:
>
>> Are we going to have both the loadable op and the Gen51?
>
>
> I added the numbered version before noticing that there is already
> the named gen. I do not object to removing GEN51 and keeping
> gentune only, although having GEN51 allows for better compatibility
> with other Csound versions that do not support named loadable GEN
> routines.




-------------------------------------------------------
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now.
http://ads.osdn.com/?ad_id=6595&alloc_id=14396&op=click
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

Yes; compatibility with code that does not exist yet is not a
reasonable stance.  The aim has been to separate the engine from the
details, and the loading of NEW ftable definitions fits that model.
Otherwise we will be forever be making the system compatible with the
latest idea that someone puts in Csound4 (which as far as I know
remains frozen).
==John ffitch

>>>>> "steven" == steven yi  writes:

 steven> Makes sense to me to keep the numbered version for the sake of greater 
 steven> compatibility.  I don't like the idea of the redundant code, so any 
 steven> objections to removing the loadable version?



-------------------------------------------------------
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now.
http://ads.osdn.com/?ad_id=6595&alloc_id=14396&op=click
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

Well, it seems that it exists since Gabriel wrote it and sent it to the 
list and it is the same code as in CS5 CVS.  I assume that he will put 
it in as gen 51 in CsoundAV as that is what he numbered it and what he 
announced it as.  Now, if you want to ignore that, that's fine, but it 
seems not very pragmatic.  Don't get me wrong, I wish everyone would put 
all new code into CS5 myself and I think that having to deal with new 
CS4 compatibility issues that rise out of not respecting that is not 
very sane, but since Gabriel may very well be addressing musical issues 
and has a very real need for gen51 in his music and in CsoundAV, and 
since there is no labelled fgen in CsoundAV, it seems like a very minor 
inconvenience to allow gen51 in for compatibility versus rejecting the 
notion.  I think this is reasonable, unless Gabriel will be syncing up 
CsoundAV with CS5 sooner than later and the gen routine can be 
introduced with the named, loadable version and the numbered version 
dropped.

steven


jpff@codemist.co.uk wrote:

>Yes; compatibility with code that does not exist yet is not a
>reasonable stance.  The aim has been to separate the engine from the
>details, and the loading of NEW ftable definitions fits that model.
>Otherwise we will be forever be making the system compatible with the
>latest idea that someone puts in Csound4 (which as far as I know
>remains frozen).
>==John ffitch
>
>  
>
>>>>>>"steven" == steven yi  writes:
>>>>>>            
>>>>>>
>
> steven> Makes sense to me to keep the numbered version for the sake of greater 
> steven> compatibility.  I don't like the idea of the redundant code, so any 
> steven> objections to removing the loadable version?
>
>
>
>-------------------------------------------------------
>SF email is sponsored by - The IT Product Guide
>Read honest & candid reviews on hundreds of IT Products from real users.
>Discover which products truly live up to the hype. Start reading now.
>http://ads.osdn.com/?ad_id=6595&alloc_id=14396&op=click
>_______________________________________________
>Csound-devel mailing list
>Csound-devel@lists.sourceforge.net
>https://lists.sourceforge.net/lists/listinfo/csound-devel
>
>
>  
>



-------------------------------------------------------
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now.
http://ads.osdn.com/?ad_id=6595&alloc_id=14396&op=click
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

And this argument applies to GEN52 as well does it?
==John ffitch


-------------------------------------------------------
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now.
http://ads.osdn.com/?ad_id=6595&alloc_id=14396&op=click
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

I think the real solution would be to improve the fgen system to
allow for a routine to have both a name and a number. An API function
for adding a new gen would take four arguments: the ENVIRON* pointer,
the suggested number of the routine (of type 'int', with zero or
negative numbers meaning that a number should be automatically assigned),
the name (of type const char*, NULL or empty string means no name),
and the function pointer.
The return value would be zero on success, while possible error cases
include an already used number or name, an invalid number, or a memory
allocation failure.
An i-time only opcode that returns the number for a particular gen name
would also be useful.

steven yi wrote:

> Well, it seems that it exists since Gabriel wrote it and sent it to the 
> list and it is the same code as in CS5 CVS.  I assume that he will put 
> it in as gen 51 in CsoundAV as that is what he numbered it and what he 
> announced it as.  Now, if you want to ignore that, that's fine, but it 
> seems not very pragmatic.  Don't get me wrong, I wish everyone would put 
> all new code into CS5 myself and I think that having to deal with new 
> CS4 compatibility issues that rise out of not respecting that is not 
> very sane, but since Gabriel may very well be addressing musical issues 
> and has a very real need for gen51 in his music and in CsoundAV, and 
> since there is no labelled fgen in CsoundAV, it seems like a very minor 
> inconvenience to allow gen51 in for compatibility versus rejecting the 
> notion.  I think this is reasonable, unless Gabriel will be syncing up 
> CsoundAV with CS5 sooner than later and the gen routine can be 
> introduced with the named, loadable version and the numbered version 
> dropped.


-------------------------------------------------------
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now.
http://ads.osdn.com/?ad_id=6595&alloc_id=14396&op=click
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

Hi John,

Yes, I'd say, if the situation was similar to the current one. Like I 
said, I think it's purely pragmatic, and I don't particularly care for 
supporting "new" CS4 features either. Adding a new numbered gen routine 
that doesn't impact anything in the engine and introduces no new 
dependencies doesn't seem very much to accomodate, while gaining a 
little bit of compatibility.

Well, I've said all I have to say on this issue. I maintain my request 
that the loadable version be removed in favor of the numbered version.  
If that's asking too much, then we'll simply disagree and either way 
we'll be done with this issue.

Thanks,
steven




jpff@codemist.co.uk wrote:

>And this argument applies to GEN52 as well does it?
>==John ffitch
>
>
>-------------------------------------------------------
>SF email is sponsored by - The IT Product Guide
>Read honest & candid reviews on hundreds of IT Products from real users.
>Discover which products truly live up to the hype. Start reading now.
>http://ads.osdn.com/?ad_id=6595&alloc_id=14396&op=click
>_______________________________________________
>Csound-devel mailing list
>Csound-devel@lists.sourceforge.net
>https://lists.sourceforge.net/lists/listinfo/csound-devel
>
>
>  
>



-------------------------------------------------------
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now.
http://ads.osdn.com/?ad_id=6595&alloc_id=14396&op=click
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

Hi Istvan,

I'm not sure this would work well.  Users depend on a gen routine to be 
named or numbered an exact way for them to use it in their scripts. The 
request for a number that has the possibility of failure means that it 
is not guaranteed, so a user's csound project may or may not render 
correctly.  (I'm having a hard time understanding how the requested 
number would be usable if there's the possibility of failure.)

I can imagine using the i-time opcode for caching the number assigned to 
a gen name so that the lookup doesn't have to happen everywhere else in 
the code (which performance wise seems only slight), but is there any 
other use-case scenarios for an opcode such as that?  I imagine most 
people will just use the named version.

It seems to me after CS5 is out of beta and everyone's personal Csound 
projects are based on CS5, there won't be any more numbered gen routines 
and only dynamically loaded, named gen routines.

steven


> I think the real solution would be to improve the fgen system to
 >allow for a routine to have both a name and a number. An API function
 >for adding a new gen would take four arguments: the ENVIRON* pointer,
 >the suggested number of the routine (of type 'int', with zero or
 >negative numbers meaning that a number should be automatically >assigned),
 >the name (of type const char*, NULL or empty string means no name),
 >and the function pointer.
 >The return value would be zero on success, while possible error cases
 >include an already used number or name, an invalid number, or a memory
 >allocation failure.
 >An i-time only opcode that returns the number for a particular gen name
 >would also be useful.


-------------------------------------------------------
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now.
http://ads.osdn.com/?ad_id=6595&alloc_id=14396&op=click
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net