Re: [PD] can a [clone] instance determine the total number of clones?

2021-03-03 Thread IOhannes m zmoelnig 

On 3/4/21 12:23 AM, William Huston wrote:

Ideally, it would be nice to change N in*one place*. I think this might be
possible if the first feature was implemented.


hoiw about your own little abstraction wrapper?


~~~
[inlet~]
|
[clone foo $1 $1 $0-]
|
[outlet]

[array define $0-control $1]

[array define $0-data]

[loadbang]
|
[float $1]
|
[* 10]
|
[resize $1(
|
[s $0-data]
~~~


so you only need to change [clonefoo 10] to [clonefoo 39] and it will 
create 39 clones for you, each of them knowing the total number of clones.
it will also  make sure that the "control" array has the same size as 
there are clones, and that the "data" array has 10* that size.
as you can see, "resizing" the "control" array is trivial; resizing to a 
size different from the number of elements requries a bit of code.



> "If I am instance #1, and the number of instances has changed since
> last time (stored in a global variable), then resize my arrays".

that's probably a misunderstanding here, how [clone] works.
1. you cannot change the number of clones dynamically.
2. if you change [clone foo 10] to [clone foo 20], this will delete all 
10 instances of [foo] and create 20 *new* instances.
the new instance #1 has no relationship with the old instance #1 (apart 
from being read from the same abstraction and sharing the same 
arguments). esp. it has no knowledge about anything that might have 
happened before the [clone] object was changed. so there is no "since 
last time".


mdgfasr
IOhannes


___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

Re: [PD] can a [clone] instance determine the total number of clones?

2021-03-03 Thread William Huston 
On Wed, Mar 3, 2021, 3:37 PM Christof Ressi  wrote:

> I find myself passing in the number of total clones as a parameter.
>
> Yes, that's the only way.
>


Thanks Christof.

> I would like to avoid this if possible.
>
> Why? Is it only for cosmetic reasons, or for technical reasons?
>

Yes, mostly aesthetics. It seems redundant. It creates extra work. It would
simply my development workflow.

I'm communicating to my clones via arrays.
Each clone reads values from certain arrays, and writes values to other
arrays.

(In the output case, the clone instance number always corresponds to the
array element number, so I can see what each clone is doing. On the input
case, I'm reading at audio rate

So when I change the number of clones, I must change this in the 2 places
you've mentioned. I also have to resize my arrays, which requires changing
at least 1 message. (I imagine that I can probably change all my arrays
using a single compound message, but haven't tried this yet).

Ideally, it would be nice to change N in *one place*. I think this might be
possible if the first feature was implemented.

e.g., "If I am instance #1, and the number of instances has changed since
last time (stored in a global variable), then resize my arrays".

Yes, [clone foo 20 20] might look ugly, but it does the job. Personally, I
> haven't ever needed the number of cloned instances inside an instance...
>
I am still working on converting my 12 band flanger/auto panner to an
N-band Flanger using clone.

If we think about the clones spreading out and creating an equal phase
displacement across all N bands, and if we let S be the instance number,
then we can calculate the phase displacement for an arbitrary instance like
this:

PhaseDisplacement(S) = (2π / N) * S
So each clone needs to know the value of both S ($1) and N.

but if you need it frequently, you can make a feature request on GitHub.
>
I think I will, thanks!

BH


Christof
> On 03.03.2021 20:12, William Huston wrote:
>
> Is there a way for a clone instance to determine the total number of
> clones?
>
> Because often the behavior of a clone depends on its relative position
> within a chain of instances.
>
> "I am instance 12 of 100, therefore, I should behave like so"
>
> I find myself passing in the number of total clones as a parameter.
> I would like to avoid this if possible.
>
> Thanks,
> BH
>
>
>
>
>
>
>
>
>
>
> --
> William Huston:  williamahus...@gmail.com
> Binghamton NY
>
> *Public Service Mapping / Videography / Research / Education / Safety
> Advocacy*
> Blog  -- Facebook
>  -- Twitter
> -- Youtube
> 
> * -- Podcast Blog  *
> *Document collections*: VirtualPipelines
>  -- BHDCSDimockArchive
> 
> *Please support my work! -- *TinyURL.com/DonateToBillHuston
>
>
>
>
> ___pd-l...@lists.iem.at mailing 
> list
> UNSUBSCRIBE and account-management -> 
> https://lists.puredata.info/listinfo/pd-list
>
> ___
> Pd-list@lists.iem.at mailing list
> UNSUBSCRIBE and account-management ->
> https://lists.puredata.info/listinfo/pd-list
>
___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

Re: [PD] can a [clone] instance determine the total number of clones?

2021-03-03 Thread Alexandre Torres Porres 
just saying I usually have to send the total number of clones as an
argument.

Em qua., 3 de mar. de 2021 às 19:53, Christof Ressi 
escreveu:

> I can imagine adding a flag to [clone] to automatically pass the number of
> instances as the second argument:
>
> [clone -n foo 20] == [clone foo 20 20]
>
> but I'm not sure it's worth it...
>
nah, it's ugly, you need to do something weird instead of just sending the
argument. It'd be good if this was retrievable as an argument in the first
design, now we have to live with it or break compatibility

__
;
pd compatibility-flag 0.51 <
__

:facepalm:
___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

Re: [PD] can a [clone] instance determine the total number of clones?

2021-03-03 Thread Christof Ressi 
I can imagine adding a flag to [clone] to automatically pass the number 
of instances as the second argument:


[clone -n foo 20] == [clone foo 20 20]

but I'm not sure it's worth it...

On 03.03.2021 22:11, IOhannes m zmölnig wrote:

On 3/3/21 9:34 PM, Christof Ressi wrote:

Yes, [clone foo 20 20] might look ugly,


well.
otoh it looks quite elegant inside the cloned abstraction:
$1 - id of current instance
$2 - number of instances
$3... - other args

i can't think of an interface to get the total number of instances 
inside the abstraction that is not much, much uglier.


somethiung like

[clones(
|
[pdcontrol]
|

:facepalm:

gfmards
IOhannes


___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list
___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

Re: [PD] can a [clone] instance determine the total number of clones?

2021-03-03 Thread IOhannes m zmölnig 

On 3/3/21 9:34 PM, Christof Ressi wrote:

Yes, [clone foo 20 20] might look ugly,


well.
otoh it looks quite elegant inside the cloned abstraction:
$1 - id of current instance
$2 - number of instances
$3... - other args

i can't think of an interface to get the total number of instances 
inside the abstraction that is not much, much uglier.


somethiung like

[clones(
|
[pdcontrol]
|

:facepalm:

gfmards
IOhannes



OpenPGP_signature
Description: OpenPGP digital signature
___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

Re: [PD] can a [clone] instance determine the total number of clones?

2021-03-03 Thread IOhannes m zmölnig 

On 3/3/21 9:51 PM, hans w. koch wrote:

i sometimes used the attached idiom, which works in certain circumstances.
(using the [max] object).


for such counting i typically use [value].
as in:

[loadbang]
|
[t b b]
| [value foo]
| [+1]
| [value foo]
|
[delay 0]
[value foo]
|

with "foo" being replaced by some unique symbol shared among all instances.

this is of course more useful outside of [clone] (where christof's 
suggestions is much nicer), e.g. when dealing with multiple 
GOP-abstractions.


masfd
IOhannes



OpenPGP_signature
Description: OpenPGP digital signature
___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

Re: [PD] can a [clone] instance determine the total number of clones?

2021-03-03 Thread hans w. koch 
i sometimes used the attached idiom, which works in certain circumstances.
(using the [max] object).
christofs solultion is more sane though…

hans

<>


> Am 03.03.2021 um 21:34 schrieb Christof Ressi :
> 
> 
>> I find myself passing in the number of total clones as a parameter.
> Yes, that's the only way.
> 
> 
>> I would like to avoid this if possible.
> Why? Is it only for cosmetic reasons, or for technical reasons?
> 
> Yes, [clone foo 20 20] might look ugly, but it does the job. Personally, I 
> haven't ever needed the number of cloned instances inside an instance... but 
> if you need it frequently, you can make a feature request on GitHub.
> 
> Christof
> 
> On 03.03.2021 20:12, William Huston wrote:
>> Is there a way for a clone instance to determine the total number of clones?
>> 
>> Because often the behavior of a clone depends on its relative position
>> within a chain of instances. 
>> 
>> "I am instance 12 of 100, therefore, I should behave like so"
>> 
>> I find myself passing in the number of total clones as a parameter. 
>> I would like to avoid this if possible. 
>> 
>> Thanks,
>> BH
>> 
>> 
>> 
>> 
>> 
>> 
>> 
>> 
>> 
>> 
>> --
>> William Huston:  williamahus...@gmail.com
>> Binghamton NY
>> 
>> Public Service Mapping / Videography / Research / Education / Safety Advocacy
>> Blog -- Facebook -- Twitter  -- Youtube -- Podcast Blog
>> Document collections: VirtualPipelines -- BHDCSDimockArchive
>> Please support my work! -- TinyURL.com/DonateToBillHuston
>> 
>> 
>> 
>> 
>> 
>> 
>> ___
>> 
>> Pd-list@lists.iem.at
>>  mailing list
>> UNSUBSCRIBE and account-management -> 
>> https://lists.puredata.info/listinfo/pd-list
> ___
> Pd-list@lists.iem.at mailing list
> UNSUBSCRIBE and account-management -> 
> https://lists.puredata.info/listinfo/pd-list

___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

Re: [PD] can a [clone] instance determine the total number of clones?

2021-03-03 Thread Christof Ressi 
I find myself passing in the number of total clones as a parameter. 

Yes, that's the only way.

I would like to avoid this if possible. 

Why? Is it only for cosmetic reasons, or for technical reasons?

Yes, [clone foo 20 20] might look ugly, but it does the job. Personally, 
I haven't ever needed the number of cloned instances inside an 
instance... but if you need it frequently, you can make a feature 
request on GitHub.


Christof

On 03.03.2021 20:12, William Huston wrote:
Is there a way for a clone instance to determine the total number of 
clones?


Because often the behavior of a clone depends on its relative position
within a chain of instances.

"I am instance 12 of 100, therefore, I should behave like so"

I find myself passing in the number of total clones as a parameter.
I would like to avoid this if possible.

Thanks,
BH










--
William Huston: williamahus...@gmail.com 
Binghamton NY

*Public Service Mapping / Videography / Research / Education / Safety 
Advocacy*
Blog  -- Facebook 
 -- Twitter 
-- Youtube 
*-- 
Podcast Blog 

*
*Document collections*: VirtualPipelines 
 -- BHDCSDimockArchive 

*Please support my work! -- *TinyURL.com/DonateToBillHuston 



**



___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list
___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list

[PD] can a [clone] instance determine the total number of clones?

2021-03-03 Thread William Huston 
Is there a way for a clone instance to determine the total number of clones?

Because often the behavior of a clone depends on its relative position
within a chain of instances.

"I am instance 12 of 100, therefore, I should behave like so"

I find myself passing in the number of total clones as a parameter.
I would like to avoid this if possible.

Thanks,
BH










--
William Huston:  williamahus...@gmail.com
Binghamton NY

*Public Service Mapping / Videography / Research / Education / Safety
Advocacy*
Blog  -- Facebook
 -- Twitter
-- Youtube

* -- Podcast Blog *
*Document collections*: VirtualPipelines
 -- BHDCSDimockArchive

*Please support my work! -- *TinyURL.com/DonateToBillHuston
___
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> 
https://lists.puredata.info/listinfo/pd-list