Oooh. I like it!

-pbs



--- jpff@cs.bath.ac.uk wrote:

From: jpff@cs.bath.ac.uk
To: csound@lists.bath.ac.uk
Subject: Re: [Csnd] vbap suggestions
Date: Fri, 13 Jul 2012 22:35:39 +0100

OK; wull look into it

too late tonight....



> hi ben, hi john -
> i agree. i also never really understood the *move variants, and i do
> think that a new variant (like vbap1) which refers to a handle argument
> would be the best.
> thanks for looking into this, john!
> best -
> 	joachim
>
>
> Am 13.07.2012 21:44, schrieb ben hackbarth:
>> hi john,
>>
>> i agree with joachim; there are definately situations where one would
>> want to use multiple vbap setups in a single orchestra run.  and i
>> think that the idea having vbaplsinit return a handle that is then
>> supplied to the vbap opcodes is a good solution.
>>
>> regarding the vbap*move opcodes, when i first looked at vbap years
>> ago, their existence made sense since the regular (non-moving) vbap
>> opcodes didn't take k-rate arguments.  however, since they now do, i'm
>> not even sure how useful the *move opcodes are anymore.  i get that
>> backwards compatibility is important, but i'm not sure that a next
>> generation of move opcodes is needed.... unless there are those out
>> there who use them, of course!
>>
>> what about using "vbap1" as the next-generation vbap and adding the
>> speaker-setup init argument to it?
>>
>> --  ben
>>
>>
>> On Fri, Jul 13, 2012 at 10:37 AM, john ffitch 
>> wrote:
>>> I have been looking at the vbap/vbaplsinit issue.  As currently
>>> implemented so there is only one call to vbaplsinit, and the result is
>>> held in an internal global.  As I understand recent comments, some
>>> people would like to have more than one speaker layout in one run.
>>>
>>> I _think_ this could achieved if
>>> 1) vbaplsinit returned a value that indicates the speaker layout
>>> 2) add another (optional) argument to vbap opcodes to say which layout
>>> to use
>>>
>>> This is compatible with current code but at present I cannot see how
>>> to make vbap*move opcodes work this way without breaking backwards
>>> compatability.  One could create a vbap*moveX family opcode with the
>>> extra argument but that seems odd.
>>>
>>> Any suggestions?
>>>
>>> ==John ffitch
>>>
>>>
>>> Send bugs reports to the Sourceforge bug tracker
>>>             https://sourceforge.net/tracker/?group_id=81968&atid=564599
>>> 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 the Sourceforge bug tracker
>>             https://sourceforge.net/tracker/?group_id=81968&atid=564599
>> 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 the Sourceforge bug tracker
>             https://sourceforge.net/tracker/?group_id=81968&atid=564599
> 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 the Sourceforge bug tracker
            https://sourceforge.net/tracker/?group_id=81968&atid=564599
Discussions of bugs and features can be posted here
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"





_____________________________________________________________
Netscape.  Just the Net You Need.

> Oooh. I like it!
>
> -pbs
>
>

In CS6 the vpab, vbap1 and vbapz opcodes now take an optional extra
argument, defaulting to zero, to say which speaker layout is meant. (in
git now)

vbaplsinit on the other hand still only creates layout zero.  I think the
only problem is to specify the layout.  In vbaplsinit I cannot see how to
modify it in a compatible way.  I could create a vbaplsinit1 with an
additional argument, but rather code-bloating.....

My earlier belief that it could be with an answer seems unlikely if one
wants to control things.

Any advice?

==John ff

I ike it -- compable with current opcode, but nees to decide on max number
of layouts; 10, 100, 1000,.... according to number of digits one keeps.

I have a vbaplsinit1 at present but would be easy to modify

==John ff

> An idea is a new opcode
>
>   vbaplayouts inum
>
> that allocates an array of ls-instances and assigns it to the
> global variable `vbap_layout_array'. Then the fractional part of
> `idim' in
>
>   vbaplsinit idim, ilsnum [, idir1] [, idir2] [...] [, idir32]
>
> is the number of the layout. Example:
>
>   vbaplayouts 4
>   vbaplsinit 2.01, 8, 0, 45, 90, 135, 180, 225, 270, 315
>   vbaplsinit 2.02, 8, 0, 30, 80, 140, 150, 200, 290, 330
>   vbaplsinit 2.03, 8, 0, ...
>   vbaplsinit 2.04, 8, 0, ...
>
> If the frac(idim) is 0, the ls-instance is created with the
> classic `create_ls_table'.
>
> tito
>
>
> Send bugs reports to the Sourceforge bug tracker
>             https://sourceforge.net/tracker/?group_id=81968&atid=564599
> Discussions of bugs and features can be posted here
> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
> csound"
>
>
>
>

I have stuck with your first suggestion -- easier to explain/document. 
Limited to 100 layouts which should be enough.

Still to do
a) arrange to recover space (was not in original either)
b) back-port to cs5

Thanks for the ideas

==John ff

> Another interpretation of `idim' (but I don't know if it confuses) to
> avoid a limit is
>
>   vbaplsinit idim, ilsnum [, idir1] [, idir2] [...] [, idir32]
>
> where `idim' is an "integer" for the old behavior and
>
>   layout.dimension
>
> if there are multiple layouts. Example:
>
>    vbaplsinit 2, ...       (layout =   0, dimension = 2)
>    vbaplsinit 3, ...       (layout =   0, dimension = 3)
>
>    vbaplayouts 1234
>    vbaplsinit 132.2, ...   (layout = 132, dimension = 2)
>
> tito
>
> On Tue, Jul 17, 2012 at 07:28:55PM +0100, jpff@cs.bath.ac.uk wrote:
>> I ike it -- compable with current opcode, but nees to decide on max
>> number
>> of layouts; 10, 100, 1000,.... according to number of digits one keeps.
>>
>> I have a vbaplsinit1 at present but would be easy to modify
>>
>> ==John ff
>>
>> > An idea is a new opcode
>> >
>> >   vbaplayouts inum
>> >
>> > that allocates an array of ls-instances and assigns it to the
>> > global variable `vbap_layout_array'. Then the fractional part of
>> > `idim' in
>> >
>> >   vbaplsinit idim, ilsnum [, idir1] [, idir2] [...] [, idir32]
>> >
>> > is the number of the layout. Example:
>> >
>> >   vbaplayouts 4
>> >   vbaplsinit 2.01, 8, 0, 45, 90, 135, 180, 225, 270, 315
>> >   vbaplsinit 2.02, 8, 0, 30, 80, 140, 150, 200, 290, 330
>> >   vbaplsinit 2.03, 8, 0, ...
>> >   vbaplsinit 2.04, 8, 0, ...
>> >
>> > If the frac(idim) is 0, the ls-instance is created with the
>> > classic `create_ls_table'.
>> >
>> > tito
>
>
> Send bugs reports to the Sourceforge bug tracker
>             https://sourceforge.net/tracker/?group_id=81968&atid=564599
> Discussions of bugs and features can be posted here
> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
> csound"
>
>
>
>