Re: audio visualized

[olivier jeannel]

https://www.dropbox.com/s/33h9t5bzm6hj016/xport_SoundWave.rar

Here's a link to an equalizer thing made with wave to spectrum.
It's 5Mb because there is a wav included. Note you'll have to manualy 
input the audio file path in the Wave2Spectrum compound (file is located 
in "Audio" directory)


I gave Claude original sample scene a try and couldn't get it to work 
until I replace the wave2Spectrum compound in the Icetree by the one I 
had installed on my laptop.
I suspect the compound in the scene sample contains an old version of 
wave2Spectrum. Or my laptop is to weak maybe.


I made pretty funny stuf with it. I should post some samples if the 
project allows it.


Take care,

Olivier






Le 03/06/2013 22:56, olivier jeannel a écrit :

Sorry was claude vervoort (wave2spectrum)
Le 03/06/2013 22:34, royston michaels a écrit :

That would be awesome if you could.
Thanks, do you mean Claude or Paul??



On 6/3/13, olivier jeannel  wrote:

Yes I used them both, and also the python addon from Guy Rabiller (very
nice and straight forward).
I like the one from Paul better because you can have several audio
sources, while the one from Stephen you have to cache the file and you
can only have one audio cached file per scene (unless I'm completly 
wrong).

Best advice is dowload Paul demo scene on github
https://github.com/claudevervoort-perso/xsi-audio-spectrum/blob/master/scenes/sine_demo_scene.zip 


It shows how to do setup properly.
Unfortunatly I'll be out of the office tomorrow. Let me know if you 
need

it, I might be able to send a file in the morning.

Le 03/06/2013 21:35, royston michaels a écrit :

Hi Olivier,
have you tried Stephan Woermann's plugin?
I'm finding wave2spectrum not working for
me, although the test on vimeo is exactly what
I need with a few tweaks.
Do you have any tips using wave2spectrum?
I'm still an ice noob.




On 6/3/13, royston michaels  wrote:

Thanks so much Olivier,

Will dig into it.

On 6/3/13, olivier jeannel  wrote:

WaveToSpectrum from Claude Vervoort
Does it exactly.
Just been using it last week ;)

https://github.com/claudevervoort-perso/xsi-audio-spectrum

http://www.si-community.com/community/viewtopic.php?f=15&p=19630


Le 03/06/2013 17:32, royston michaels a écrit :

Hi list,
I need some help creating the following,
https://www.youtube.com/watch?NR=1&feature=fvwp&v=c1I0XjSaVmg

I can do this manually but it needs to animate with an audio file
supplied
and I'd also like to have the height drive the color value. I'm 
sure

this is
very easy with ice.

Any ideas or links I can check out is much appreciated.

TIA
Royston

willpresto...@gmail.com

[Will Preston]

[no subject]

[Will Preston]

willpresto...@gmail.com

Re: audio visualized

[royston michaels]

Thanks Olivier,
Much appreciated,

On 6/4/13, olivier jeannel  wrote:
> https://www.dropbox.com/s/33h9t5bzm6hj016/xport_SoundWave.rar
>
> Here's a link to an equalizer thing made with wave to spectrum.
> It's 5Mb because there is a wav included. Note you'll have to manualy
> input the audio file path in the Wave2Spectrum compound (file is located
> in "Audio" directory)
>
> I gave Claude original sample scene a try and couldn't get it to work
> until I replace the wave2Spectrum compound in the Icetree by the one I
> had installed on my laptop.
> I suspect the compound in the scene sample contains an old version of
> wave2Spectrum. Or my laptop is to weak maybe.
>
> I made pretty funny stuf with it. I should post some samples if the
> project allows it.
>
> Take care,
>
> Olivier
>
>
>
>
>
>
> Le 03/06/2013 22:56, olivier jeannel a écrit :
>> Sorry was claude vervoort (wave2spectrum)
>> Le 03/06/2013 22:34, royston michaels a écrit :
>>> That would be awesome if you could.
>>> Thanks, do you mean Claude or Paul??
>>>
>>>
>>>
>>> On 6/3/13, olivier jeannel  wrote:
 Yes I used them both, and also the python addon from Guy Rabiller (very
 nice and straight forward).
 I like the one from Paul better because you can have several audio
 sources, while the one from Stephen you have to cache the file and you
 can only have one audio cached file per scene (unless I'm completly
 wrong).
 Best advice is dowload Paul demo scene on github
 https://github.com/claudevervoort-perso/xsi-audio-spectrum/blob/master/scenes/sine_demo_scene.zip


 It shows how to do setup properly.
 Unfortunatly I'll be out of the office tomorrow. Let me know if you
 need
 it, I might be able to send a file in the morning.

 Le 03/06/2013 21:35, royston michaels a écrit :
> Hi Olivier,
> have you tried Stephan Woermann's plugin?
> I'm finding wave2spectrum not working for
> me, although the test on vimeo is exactly what
> I need with a few tweaks.
> Do you have any tips using wave2spectrum?
> I'm still an ice noob.
>
>
>
>
> On 6/3/13, royston michaels  wrote:
>> Thanks so much Olivier,
>>
>> Will dig into it.
>>
>> On 6/3/13, olivier jeannel  wrote:
>>> WaveToSpectrum from Claude Vervoort
>>> Does it exactly.
>>> Just been using it last week ;)
>>>
>>> https://github.com/claudevervoort-perso/xsi-audio-spectrum
>>>
>>> http://www.si-community.com/community/viewtopic.php?f=15&p=19630
>>>
>>>
>>> Le 03/06/2013 17:32, royston michaels a écrit :
 Hi list,
 I need some help creating the following,
 https://www.youtube.com/watch?NR=1&feature=fvwp&v=c1I0XjSaVmg

 I can do this manually but it needs to animate with an audio file
 supplied
 and I'd also like to have the height drive the color value. I'm
 sure
 this is
 very easy with ice.

 Any ideas or links I can check out is much appreciated.

 TIA
 Royston

>

>>>
>>>
>>
>>
>>
>
>

Persistent objects ids

[Michal Doniec]

Hi,

Is there any obvious way I am missing which would allow me to identify
objects in the scene, but not by name?
Something like ObjectID, but persistent between scene saves and loads. I
can generate UUID or something similar using "ObjectAdded" event and store
custom property but I'd rather avoid it.

I want to be able to track objects regardless if they were renamed or not.

-- 
--
Michal
http://uk.linkedin.com/in/mdoniec

Re: Persistent objects ids

[Sandy Sutherland]

Create an annotation and write a ID into that - we used that in our 
asset system, worked fine for us.


S.

On 2013/06/04 12:49 PM, Michal Doniec wrote:

Hi,

Is there any obvious way I am missing which would allow me to identify 
objects in the scene, but not by name?
Something like ObjectID, but persistent between scene saves and 
loads. I can generate UUID or something similar using "ObjectAdded" 
event and store custom property but I'd rather avoid it.


I want to be able to track objects regardless if they were renamed or not.

--
--
Michal
http://uk.linkedin.com/in/mdoniec

Re: Persistent objects ids

[Michal Doniec]

That's what I've been doing (I use properties instead of annotation but the
principle is the same), maybe there is no need to change after all :)


On 4 June 2013 12:52, Sandy Sutherland  wrote:

> Create an annotation and write a ID into that - we used that in our asset
> system, worked fine for us.
>
> S.
>
>
> On 2013/06/04 12:49 PM, Michal Doniec wrote:
>
>> Hi,
>>
>> Is there any obvious way I am missing which would allow me to identify
>> objects in the scene, but not by name?
>> Something like ObjectID, but persistent between scene saves and loads. I
>> can generate UUID or something similar using "ObjectAdded" event and store
>> custom property but I'd rather avoid it.
>>
>> I want to be able to track objects regardless if they were renamed or not.
>>
>> --
>> --
>> Michal
>> http://uk.linkedin.com/in/**mdoniec 
>>
>
>


-- 
--
Michal
http://uk.linkedin.com/in/mdoniec

Re: audio visualized

[Claude Vervoort]

Just stumbled on that one here. Thanks Olivier for pointing to my plugin :)
I understand now where it comes from ;)

I tried to help Royston by annotating the ICE tree, i'll upload it later
today.

Cheers! (and yes i'm curious to see the funny stuff !!!).

Claude



On Tue, Jun 4, 2013 at 5:52 AM, royston michaels wrote:

> Thanks Olivier,
> Much appreciated,
>
> On 6/4/13, olivier jeannel  wrote:
> > https://www.dropbox.com/s/33h9t5bzm6hj016/xport_SoundWave.rar
> >
> > Here's a link to an equalizer thing made with wave to spectrum.
> > It's 5Mb because there is a wav included. Note you'll have to manualy
> > input the audio file path in the Wave2Spectrum compound (file is located
> > in "Audio" directory)
> >
> > I gave Claude original sample scene a try and couldn't get it to work
> > until I replace the wave2Spectrum compound in the Icetree by the one I
> > had installed on my laptop.
> > I suspect the compound in the scene sample contains an old version of
> > wave2Spectrum. Or my laptop is to weak maybe.
> >
> > I made pretty funny stuf with it. I should post some samples if the
> > project allows it.
> >
> > Take care,
> >
> > Olivier
> >
> >
> >
> >
> >
> >
> > Le 03/06/2013 22:56, olivier jeannel a écrit :
> >> Sorry was claude vervoort (wave2spectrum)
> >> Le 03/06/2013 22:34, royston michaels a écrit :
> >>> That would be awesome if you could.
> >>> Thanks, do you mean Claude or Paul??
> >>>
> >>>
> >>>
> >>> On 6/3/13, olivier jeannel  wrote:
>  Yes I used them both, and also the python addon from Guy Rabiller
> (very
>  nice and straight forward).
>  I like the one from Paul better because you can have several audio
>  sources, while the one from Stephen you have to cache the file and you
>  can only have one audio cached file per scene (unless I'm completly
>  wrong).
>  Best advice is dowload Paul demo scene on github
> 
> https://github.com/claudevervoort-perso/xsi-audio-spectrum/blob/master/scenes/sine_demo_scene.zip
> 
> 
>  It shows how to do setup properly.
>  Unfortunatly I'll be out of the office tomorrow. Let me know if you
>  need
>  it, I might be able to send a file in the morning.
> 
>  Le 03/06/2013 21:35, royston michaels a écrit :
> > Hi Olivier,
> > have you tried Stephan Woermann's plugin?
> > I'm finding wave2spectrum not working for
> > me, although the test on vimeo is exactly what
> > I need with a few tweaks.
> > Do you have any tips using wave2spectrum?
> > I'm still an ice noob.
> >
> >
> >
> >
> > On 6/3/13, royston michaels  wrote:
> >> Thanks so much Olivier,
> >>
> >> Will dig into it.
> >>
> >> On 6/3/13, olivier jeannel  wrote:
> >>> WaveToSpectrum from Claude Vervoort
> >>> Does it exactly.
> >>> Just been using it last week ;)
> >>>
> >>> https://github.com/claudevervoort-perso/xsi-audio-spectrum
> >>>
> >>> http://www.si-community.com/community/viewtopic.php?f=15&p=19630
> >>>
> >>>
> >>> Le 03/06/2013 17:32, royston michaels a écrit :
>  Hi list,
>  I need some help creating the following,
>  https://www.youtube.com/watch?NR=1&feature=fvwp&v=c1I0XjSaVmg
> 
>  I can do this manually but it needs to animate with an audio file
>  supplied
>  and I'd also like to have the height drive the color value. I'm
>  sure
>  this is
>  very easy with ice.
> 
>  Any ideas or links I can check out is much appreciated.
> 
>  TIA
>  Royston
> 
> >
> 
> >>>
> >>>
> >>
> >>
> >>
> >
> >
>
>

Re: unloading a plugin, recompiling and reloading it in softimage.

[Luc-Eric Rousseau]

see if New Scene clears it up, it flushes the undo stack
Le 2013-06-03 21:19, "Raffaele Fragapane"  a
écrit :

> In Maya, with nodes particularly, this is very common (a mandatory step
> practically), in Soft I only had it once or twice, but some times with
> commands you have to flush the undo cache before you unload the plugin if
> you have an undo implementation.
> Other than that, and we're talking fluke chance levels, I never had the
> issue in Linux or windows.
>
>
> On Tue, Jun 4, 2013 at 10:18 AM, Ahmidou Lyazidi wrote:
>
>> I know it's not helping, but I did a few linux and windows c++ plugins
>> lately, and never had those issues. That's pretty weird.
>>
>> ---
>> Ahmidou Lyazidi
>> Director | TD | CG artist
>> http://vimeo.com/ahmidou/videos
>> http://www.cappuccino-films.com
>>
>>
>> 2013/6/4 ran sariel 
>>
>>> No Mat, there are no other copies, and when the plugin is unloaded. soft
>>> would not run the command,
>>>
>>> Cheers
>>> Ran
>>>
>>>
>>> On Mon, Jun 3, 2013 at 4:54 PM, Matt Lind wrote:
>>>
 Sounds like you may have multiple copies installed in different
 locations and may not be aware of it.

 ** **

 Matt

 ** **

 ** **

 ** **

 *From:* softimage-boun...@listproc.autodesk.com [mailto:
 softimage-boun...@listproc.autodesk.com] *On Behalf Of *ran sariel
 *Sent:* Monday, June 03, 2013 4:53 PM
 *To:* softimage

 *Subject:* Re: unloading a plugin, recompiling and reloading it in
 softimage.

 ** **

 the plugin registers a command and a couple of custom operators, that
 are created only when necessary. (not in the case I'm testing at the 
 moment)
 

 even when starting a new scene (so there's no active nodes related to
 the plugin), unloading the plugin recompiling and reloading, it is still
 reverting to the old plugin code. 

 ** **

  

 ** **

 ** **

 On Mon, Jun 3, 2013 at 2:44 PM, Stephan Woermann <
 swoerman...@googlemail.com> wrote:

 On a custom render property, it doesn´t work, as example.

 Maybe because some of the render parameters are still active in some
 parts of the Softimage internal structure.


 With custom shaders, sometimes you can´t recompile the dll. You get an
 write error. The same background as above.


 Custom ICE-Nodes are also a little bit tricky. When you unload the
 plugin, an existing Node will become a red status,

 Recompile and reload the plugin, will make the Node sometimes wired,
 some portsnames are double.

 Or if you delete a Node without the plugin is loaded, you get a crash.*
 ***

 ** **

 I would prefer the way to test if the plugin can be reloaded and
 refreshed how it should and when not, restart Softimage.

 I think it depend on what type of plugin you code and how deep the
 plugin is currently used in the Softimage Scene.

 Sometimes you can recompile and refresh and sometimes you can´t...

 Stephan

 ** **

 2013/6/3 Matt Lind 

 If you’re working on a PPG Layout, you’ll need to ‘refresh’ the PPG to
 see changes.

  

 That is, each PPG has a name with a triangle to its left, which if
 clicked, collapses the PPG.  Right-click the darker grey area to the right
 of the collapsible name and choose “refresh”.  This will force the PPG
 Layout logic code to evaluate from scratch.

  

  

 Matt

  

  

  

  

 *From:* softimage-boun...@listproc.autodesk.com [mailto:
 softimage-boun...@listproc.autodesk.com] *On Behalf Of *Stephan
 Woermann
 *Sent:* Monday, June 03, 2013 1:43 PM
 *To:* softimage@listproc.autodesk.com
 *Subject:* Re: unloading a plugin, recompiling and reloading it in
 softimage.

  

 Another problem can be, when some code is still "active" in Softimage.
 Like an open shader PPG of an unloaded Plugin or something else.

 Stephan

  

 2013/6/3 Stephan Woermann 

 Haven´t check the Update All over the Plugin Manager, so the next can
 be wrong.
 For changes in the PPG with new parameters, you must restart Softimage.
 

 For changes in the logic or others, unload/reload should work.

 Stephan

  

 2013/6/3 ran sariel 

 hope it was that simple, this just doesn't work on my setup.

 the plugin is definitely unloaded before the recompile, and I wonder
 where on earth does soft cache it.

  

 On Mon, Jun 3, 2013 at 10:42 AM, Vincent Ullmann <
 vincent.ullm...@googlem

Re:

[Alan Fregtman]

Trying to register another address? :p



On Tue, Jun 4, 2013 at 5:36 AM, Will Preston
wrote:

> willpresto...@gmail.com
>

Re: Persistent objects ids

[Alan Fregtman]

At work I use a topology hash which I base off a string composed of either
boundingbox and component counts, or if there's collisions from similar
objects, the whole position array in local space.

I use SHA1 for the hashing as it's a teensy bit faster to compute than MD5,
but both of those have known to have had collision flaws, so ideally one
should use something fancier.

As long as the topo stays the same, it makes it super easy to track renamed
meshes or whatever.



On Tue, Jun 4, 2013 at 7:57 AM, Michal Doniec  wrote:

> That's what I've been doing (I use properties instead of annotation but
> the principle is the same), maybe there is no need to change after all :)
>
>
> On 4 June 2013 12:52, Sandy Sutherland  wrote:
>
>> Create an annotation and write a ID into that - we used that in our asset
>> system, worked fine for us.
>>
>> S.
>>
>>
>> On 2013/06/04 12:49 PM, Michal Doniec wrote:
>>
>>> Hi,
>>>
>>> Is there any obvious way I am missing which would allow me to identify
>>> objects in the scene, but not by name?
>>> Something like ObjectID, but persistent between scene saves and loads. I
>>> can generate UUID or something similar using "ObjectAdded" event and store
>>> custom property but I'd rather avoid it.
>>>
>>> I want to be able to track objects regardless if they were renamed or
>>> not.
>>>
>>> --
>>> --
>>> Michal
>>> http://uk.linkedin.com/in/**mdoniec 
>>>
>>
>>
>
>
> --
> --
> Michal
> http://uk.linkedin.com/in/mdoniec
>

Re: Persistent objects ids

[Michal Doniec]

Interesting idea Alan - thanks for this. Can't apply it to
this particular thing I am doing now (topo, or anything can change), but
I'll save it for the future.


On 4 June 2013 15:25, Alan Fregtman  wrote:

> At work I use a topology hash which I base off a string composed of either
> boundingbox and component counts, or if there's collisions from similar
> objects, the whole position array in local space.
>
> I use SHA1 for the hashing as it's a teensy bit faster to compute than
> MD5, but both of those have known to have had collision flaws, so ideally
> one should use something fancier.
>
> As long as the topo stays the same, it makes it super easy to track
> renamed meshes or whatever.
>
>
>
> On Tue, Jun 4, 2013 at 7:57 AM, Michal Doniec  wrote:
>
>> That's what I've been doing (I use properties instead of annotation but
>> the principle is the same), maybe there is no need to change after all :)
>>
>>
>> On 4 June 2013 12:52, Sandy Sutherland  wrote:
>>
>>> Create an annotation and write a ID into that - we used that in our
>>> asset system, worked fine for us.
>>>
>>> S.
>>>
>>>
>>> On 2013/06/04 12:49 PM, Michal Doniec wrote:
>>>
 Hi,

 Is there any obvious way I am missing which would allow me to identify
 objects in the scene, but not by name?
 Something like ObjectID, but persistent between scene saves and loads.
 I can generate UUID or something similar using "ObjectAdded" event and
 store custom property but I'd rather avoid it.

 I want to be able to track objects regardless if they were renamed or
 not.

 --
 --
 Michal
 http://uk.linkedin.com/in/**mdoniec 

>>>
>>>
>>
>>
>> --
>> --
>> Michal
>> http://uk.linkedin.com/in/mdoniec
>>
>
>


-- 
--
Michal
http://uk.linkedin.com/in/mdoniec

Re: Persistent objects ids

[Ciaran Moloney]

Hi,
did you try XSIUtils.DataRepository.GetIdentifier() passing the
siObjectGUID argument?
I've used this to keep track of ICE nodes without relying on name strings.
It's persistent between sessions and is preserved when models are exported.
Works like a charm.
The only problem is that it's possibly a bit tricky to get a pointer back
to the object, since you need to know what you're looking for. I use
Application.FindObjects()
to get a collection of objects of this class, then check for the ObjectGUID
I'm looking for.

Ciaran


On Tue, Jun 4, 2013 at 12:49 PM, Michal Doniec  wrote:

> Hi,
>
> Is there any obvious way I am missing which would allow me to identify
> objects in the scene, but not by name?
> Something like ObjectID, but persistent between scene saves and loads. I
> can generate UUID or something similar using "ObjectAdded" event and store
> custom property but I'd rather avoid it.
>
> I want to be able to track objects regardless if they were renamed or not.
>
> --
> --
> Michal
> http://uk.linkedin.com/in/mdoniec
>

Re: Persistent objects ids

[Michal Doniec]

No, i haven't, but this is what I was looking for, thanks a lot!


On 4 June 2013 16:52, Ciaran Moloney  wrote:

> Hi,
> did you try XSIUtils.DataRepository.GetIdentifier() passing the
> siObjectGUID argument?
> I've used this to keep track of ICE nodes without relying on name strings.
> It's persistent between sessions and is preserved when models are exported.
> Works like a charm.
> The only problem is that it's possibly a bit tricky to get a pointer back
> to the object, since you need to know what you're looking for. I use 
> Application.FindObjects()
> to get a collection of objects of this class, then check for the ObjectGUID
> I'm looking for.
>
> Ciaran
>
>
> On Tue, Jun 4, 2013 at 12:49 PM, Michal Doniec  wrote:
>
>> Hi,
>>
>> Is there any obvious way I am missing which would allow me to identify
>> objects in the scene, but not by name?
>> Something like ObjectID, but persistent between scene saves and loads. I
>> can generate UUID or something similar using "ObjectAdded" event and store
>> custom property but I'd rather avoid it.
>>
>> I want to be able to track objects regardless if they were renamed or not.
>>
>> --
>> --
>> Michal
>> http://uk.linkedin.com/in/mdoniec
>>
>
>


-- 
--
Michal
http://uk.linkedin.com/in/mdoniec

Re: Persistent objects ids

[Alan Fregtman]

How about with refmodels? I know ObjectID sometimes changes with refmodels.
Does this ObjectGUID suffer the same fate?



On Tue, Jun 4, 2013 at 11:52 AM, Ciaran Moloney wrote:

> Hi,
> did you try XSIUtils.DataRepository.GetIdentifier() passing the
> siObjectGUID argument?
> I've used this to keep track of ICE nodes without relying on name strings.
> It's persistent between sessions and is preserved when models are exported.
> Works like a charm.
> The only problem is that it's possibly a bit tricky to get a pointer back
> to the object, since you need to know what you're looking for. I use 
> Application.FindObjects()
> to get a collection of objects of this class, then check for the ObjectGUID
> I'm looking for.
>
> Ciaran
>
>
> On Tue, Jun 4, 2013 at 12:49 PM, Michal Doniec  wrote:
>
>> Hi,
>>
>> Is there any obvious way I am missing which would allow me to identify
>> objects in the scene, but not by name?
>> Something like ObjectID, but persistent between scene saves and loads. I
>> can generate UUID or something similar using "ObjectAdded" event and store
>> custom property but I'd rather avoid it.
>>
>> I want to be able to track objects regardless if they were renamed or not.
>>
>> --
>> --
>> Michal
>> http://uk.linkedin.com/in/mdoniec
>>
>
>

RE: Persistent objects ids

[Matt Lind]

This is not safe between sessions as IDs are dynamically generated and assigned 
in the order objects were created in the scene.  If an older object is deleted 
and the scene reopened, everything after it will have a new ID.

Matt



From: softimage-boun...@listproc.autodesk.com 
[mailto:softimage-boun...@listproc.autodesk.com] On Behalf Of Ciaran Moloney
Sent: Tuesday, June 04, 2013 8:52 AM
To: softimage@listproc.autodesk.com
Subject: Re: Persistent objects ids

Hi,
did you try XSIUtils.DataRepository.GetIdentifier() passing the siObjectGUID 
argument?
I've used this to keep track of ICE nodes without relying on name strings. It's 
persistent between sessions and is preserved when models are exported.
Works like a charm.
The only problem is that it's possibly a bit tricky to get a pointer back to 
the object, since you need to know what you're looking for. I use 
Application.FindObjects() to get a collection of objects of this class, then 
check for the ObjectGUID I'm looking for.

Ciaran


On Tue, Jun 4, 2013 at 12:49 PM, Michal Doniec 
mailto:doni...@gmail.com>> wrote:
Hi,

Is there any obvious way I am missing which would allow me to identify objects 
in the scene, but not by name?
Something like ObjectID, but persistent between scene saves and loads. I can 
generate UUID or something similar using "ObjectAdded" event and store custom 
property but I'd rather avoid it.

I want to be able to track objects regardless if they were renamed or not.

--
--
Michal
http://uk.linkedin.com/in/mdoniec

Re: audio visualized

[Matthew Graves]

Very interesting only a few months ago I was making a series of particle
animations based on Stephan Woermann's plugin. Oliver I found the trick was
to make a custom cache from the audio cache then you can have multiple
caches. I will have try out your plugin Calude, there is so much potential
in what you can do from audio input in ICE.


On Tue, Jun 4, 2013 at 1:05 PM, Claude Vervoort
wrote:

> Just stumbled on that one here. Thanks Olivier for pointing to my plugin
> :) I understand now where it comes from ;)
>
> I tried to help Royston by annotating the ICE tree, i'll upload it later
> today.
>
> Cheers! (and yes i'm curious to see the funny stuff !!!).
>
> Claude
>
>
>
> On Tue, Jun 4, 2013 at 5:52 AM, royston michaels wrote:
>
>> Thanks Olivier,
>> Much appreciated,
>>
>> On 6/4/13, olivier jeannel  wrote:
>> > https://www.dropbox.com/s/33h9t5bzm6hj016/xport_SoundWave.rar
>> >
>> > Here's a link to an equalizer thing made with wave to spectrum.
>> > It's 5Mb because there is a wav included. Note you'll have to manualy
>> > input the audio file path in the Wave2Spectrum compound (file is located
>> > in "Audio" directory)
>> >
>> > I gave Claude original sample scene a try and couldn't get it to work
>> > until I replace the wave2Spectrum compound in the Icetree by the one I
>> > had installed on my laptop.
>> > I suspect the compound in the scene sample contains an old version of
>> > wave2Spectrum. Or my laptop is to weak maybe.
>> >
>> > I made pretty funny stuf with it. I should post some samples if the
>> > project allows it.
>> >
>> > Take care,
>> >
>> > Olivier
>> >
>> >
>> >
>> >
>> >
>> >
>> > Le 03/06/2013 22:56, olivier jeannel a écrit :
>> >> Sorry was claude vervoort (wave2spectrum)
>> >> Le 03/06/2013 22:34, royston michaels a écrit :
>> >>> That would be awesome if you could.
>> >>> Thanks, do you mean Claude or Paul??
>> >>>
>> >>>
>> >>>
>> >>> On 6/3/13, olivier jeannel  wrote:
>>  Yes I used them both, and also the python addon from Guy Rabiller
>> (very
>>  nice and straight forward).
>>  I like the one from Paul better because you can have several audio
>>  sources, while the one from Stephen you have to cache the file and
>> you
>>  can only have one audio cached file per scene (unless I'm completly
>>  wrong).
>>  Best advice is dowload Paul demo scene on github
>> 
>> https://github.com/claudevervoort-perso/xsi-audio-spectrum/blob/master/scenes/sine_demo_scene.zip
>> 
>> 
>>  It shows how to do setup properly.
>>  Unfortunatly I'll be out of the office tomorrow. Let me know if you
>>  need
>>  it, I might be able to send a file in the morning.
>> 
>>  Le 03/06/2013 21:35, royston michaels a écrit :
>> > Hi Olivier,
>> > have you tried Stephan Woermann's plugin?
>> > I'm finding wave2spectrum not working for
>> > me, although the test on vimeo is exactly what
>> > I need with a few tweaks.
>> > Do you have any tips using wave2spectrum?
>> > I'm still an ice noob.
>> >
>> >
>> >
>> >
>> > On 6/3/13, royston michaels  wrote:
>> >> Thanks so much Olivier,
>> >>
>> >> Will dig into it.
>> >>
>> >> On 6/3/13, olivier jeannel  wrote:
>> >>> WaveToSpectrum from Claude Vervoort
>> >>> Does it exactly.
>> >>> Just been using it last week ;)
>> >>>
>> >>> https://github.com/claudevervoort-perso/xsi-audio-spectrum
>> >>>
>> >>> http://www.si-community.com/community/viewtopic.php?f=15&p=19630
>> >>>
>> >>>
>> >>> Le 03/06/2013 17:32, royston michaels a écrit :
>>  Hi list,
>>  I need some help creating the following,
>>  https://www.youtube.com/watch?NR=1&feature=fvwp&v=c1I0XjSaVmg
>> 
>>  I can do this manually but it needs to animate with an audio file
>>  supplied
>>  and I'd also like to have the height drive the color value. I'm
>>  sure
>>  this is
>>  very easy with ice.
>> 
>>  Any ideas or links I can check out is much appreciated.
>> 
>>  TIA
>>  Royston
>> 
>> >
>> 
>> >>>
>> >>>
>> >>
>> >>
>> >>
>> >
>> >
>>
>>
>

the secret of animation

[Stephen Davidson]

https://vimeo.com/67501143

-- 

Best Regards,
*  Stephen P. Davidson**
   **(954) 552-7956
*sdavid...@3danimationmagic.com

*Any sufficiently advanced technology is indistinguishable from magic*


 - Arthur C. Clarke

syFlex ICE- length

[David Barosin]

Can anyone confirm whether or not the 'length' parameter works on the ICE
node for syflexISpring?  I'm using 2014

Thanks,
 -Dave

Re: syFlex ICE- length

[David Barosin]

Just to add...  I'm looking to get the rescale behavior that the non ice
operators offer.

Thanks again,
 -Dave


On Tue, Jun 4, 2013 at 5:05 PM, David Barosin  wrote:

> Can anyone confirm whether or not the 'length' parameter works on the ICE
> node for syflexISpring?  I'm using 2014
>
> Thanks,
>  -Dave
>
>
>
>
>
>

Flow along curve speed control

[john clausing]

hello all,..

i have an animated curve (translating) that emits an ICE cloud by position.
the cloud travels by speed "from emit" on the curve with "flow on curve".

it all works fine with the exception of speed control..there is none, even 
when i change the speed in the "flow on curve" node.
i suspect it all has to do with the curve's translation, or with the 
"translating" curve doubling as the emitter.

any thoughts?

thank you as always,

john

Re: particle flow along curve with turbulence

[David Rivera]

Pingo, how did it go along? I think it´s about time that someone nails this 
kind of compound. 
I remember back in SI 7  there was no "flow along curve" then it was given to 
community,
we need a "factory preset" turbulize on curve flow :)

D






 From: Pingo van der Brinkloev 
To: softimage@listproc.autodesk.com 
Sent: Saturday, June 1, 2013 3:59 PM
Subject: Re: particle flow along curve with turbulence
 


Hey thanks guys for all the input(s). I thought this was gonna be real easy. 
Orlando, looks like your compound is doing the trick I'll dive into it. Thanks!

P


On 01/06/2013, at 22.46, Orlando Esponda  wrote:

Pingo, here's a tricky compund, it may work or it may not, but just in case you 
want to give it a go. Just explode the compound and tweak it. A couple of 
comments will tell you what's doing what...    it's pretty simple compound so 
you shouldn't have troubles understanding it.  It should go between your 
"follow curve" nodes and your "strand trails" nodes.
>
>Hope that helps,
>Orlando.
>
>
>
>
>
>
>
>
>
>On Sat, Jun 1, 2013 at 12:34 PM,  wrote:
>
>are you adding the turbulence on the position, like: "get 
>position->turbulence->set position"?
>>If so the turbulence will be very "strong" and mostly override the effect of 
>>any forces.
>>
>>if you add the turbulence as a force (turbulence->add forces), it's influence 
>>will be less radical
>>
>>for a flow along curve, I'd combine three forces / vectors.
>>- subtract pointposition of the closest location on the curve from the 
>>particle's position (or the other way around) to pull the particles towards 
>>the curve
>>- pointtangent of the closest location on the curve for the movement along 
>>the curve
>>- turbulence to randomize
>>
>>you can multiply each of those vectors with a scalar - and by changing the 
>>scalars have fine control over the blending of the different forces.
>>
>>
>>
>>
>>-Original Message- From: Pingo van der Brinkloev
>>Sent: Saturday, June 01, 2013 10:25 AM
>>To: softimage@listproc.autodesk.com
>>Subject: particle flow along curve with turbulence
>>
>>
>>Hey list, I have some particles flowing along a curve nicely (with strands). 
>>I would like to add turbulence to the particles, so they wiggle, but still 
>>follow the curve(!) How do I do this without the particles going a-walk on 
>>me? Whichever way I try it it, the turbulence always cumulate.
>>
>>Feel like there's some memo on turbulence I didn't get.
>>
>>Cheers! 
>>
>
>--
>IMPRESSUM:
>PiXABLE STUDIOS GmbH & Co.KG, Sitz: Dresden, Amtsgericht: Dresden, HRA 6857,
>Komplementärin: Lenhard & Barth Verwaltungsgesellschaft mbH, Sitz: Dresden,
>Amtsgericht: Dresden, HRB 26501, Geschäftsführer: Frank Lenhard, Tino Barth
>
>IMPRINT:
>PiXABLE STUDIOS GmbH & Co.KG, Domicile: Dresden, Court of Registery: Dresden,
>Company Registration Number: HRA 6857, General Partner: Lenhard & Barth
>Verwaltungsgesellschaft mbH, Domicile: Dresden, Court of Registery: Dresden, 
>Company
>Registration Number: HRB 26501, Chief Executive Officers: Frank Lenhard, Tino 
>Barth
>
>
>--
>Diese E-Mail enthält vertrauliche und/oder rechtlich geschützte Informationen. 
>Wenn Sie nicht
>der richtige Adressat sind oder diese E-Mail irrtümlich erhalten haben, 
>informieren Sie bitte
>sofort den Absender und vernichten Sie diese Mail. Das unerlaubte Kopieren 
>sowie die
>unbefugte Weitergabe dieser Mail ist nicht gestattet.
>
>This e-mail may contain confidential and/or privileged information. If you are 
>not the intended
>recipient (or have received this e-mail in error) please notify the sender 
>immediately and destroy
>this e-mail. Any unauthorized copying, disclosure or distribution of the 
>material in this e-mail is
>strictly forbidden. 

Re: Persistent objects ids

[Ciaran Moloney]

Looks like it'll keep the same GUID as when exported as long as it's unique
in the scene. If you have two references of a refmodel in a scene, the
contents of the second model will get new GUIDs.


On Tue, Jun 4, 2013 at 5:41 PM, Alan Fregtman wrote:

> How about with refmodels? I know ObjectID sometimes changes with
> refmodels. Does this ObjectGUID suffer the same fate?
>
>
>
> On Tue, Jun 4, 2013 at 11:52 AM, Ciaran Moloney 
> wrote:
>
>> Hi,
>> did you try XSIUtils.DataRepository.GetIdentifier() passing the
>> siObjectGUID argument?
>> I've used this to keep track of ICE nodes without relying on name
>> strings. It's persistent between sessions and is preserved when models are
>> exported.
>> Works like a charm.
>> The only problem is that it's possibly a bit tricky to get a pointer back
>> to the object, since you need to know what you're looking for. I use 
>> Application.FindObjects()
>> to get a collection of objects of this class, then check for the ObjectGUID
>> I'm looking for.
>>
>> Ciaran
>>
>>
>> On Tue, Jun 4, 2013 at 12:49 PM, Michal Doniec  wrote:
>>
>>> Hi,
>>>
>>> Is there any obvious way I am missing which would allow me to identify
>>> objects in the scene, but not by name?
>>> Something like ObjectID, but persistent between scene saves and loads. I
>>> can generate UUID or something similar using "ObjectAdded" event and store
>>> custom property but I'd rather avoid it.
>>>
>>> I want to be able to track objects regardless if they were renamed or
>>> not.
>>>
>>> --
>>> --
>>> Michal
>>> http://uk.linkedin.com/in/mdoniec
>>>
>>
>>
>

Re: Persistent objects ids

[Ciaran Moloney]

Yes, you're right...I can reproduce a change of ID when deleting an object
and re-opening a scene. However, I would not dismiss the object GUID, which
appears to persists between sessions despite the change in the object ID
value. In the below snippet the sphere object changed ID but not GUID.

# Name: cylinder
# ID: 114
# GUID: {090F2A9A-D1FC-4112-ACC5-9345BEB48954}
#
# Name: disc
# ID: 92
# GUID: {D03EF191-3962-4540-929D-F5438CE59D79}
#
# Name: torus1
# ID: 68
# GUID: {1AC14CC7-1175-4E30-9CDD-BC0C0C1DE6EA}
#
# Name: sphere
# ID: 80
# GUID: {CC63558A-D5A3-4F55-8621-6D428FE2743A}
# 

# Delete cylinder, save and re-open scene

# Name: disc
# ID: 92
# GUID: {D03EF191-3962-4540-929D-F5438CE59D79}
#
# Name: torus1
# ID: 69
# GUID: {1AC14CC7-1175-4E30-9CDD-BC0C0C1DE6EA}
#
# Name: sphere
# ID: 99
# GUID: {CC63558A-D5A3-4F55-8621-6D428FE2743A}
#
# 




On Tue, Jun 4, 2013 at 6:38 PM, Matt Lind  wrote:

> This is not safe between sessions as IDs are dynamically generated and
> assigned in the order objects were created in the scene.  If an older
> object is deleted and the scene reopened, everything after it will have a
> new ID.
>
> ** **
>
> Matt
>
> ** **
>
> ** **
>
> ** **
>
> *From:* softimage-boun...@listproc.autodesk.com [mailto:
> softimage-boun...@listproc.autodesk.com] *On Behalf Of *Ciaran Moloney
> *Sent:* Tuesday, June 04, 2013 8:52 AM
> *To:* softimage@listproc.autodesk.com
> *Subject:* Re: Persistent objects ids
>
> ** **
>
> Hi,
>
> did you try XSIUtils.DataRepository.GetIdentifier() passing the
> siObjectGUID argument?
>
> I've used this to keep track of ICE nodes without relying on name strings.
> It's persistent between sessions and is preserved when models are exported.
> 
>
> Works like a charm.
>
> The only problem is that it's possibly a bit tricky to get a pointer back
> to the object, since you need to know what you're looking for. I use 
> Application.FindObjects()
> to get a collection of objects of this class, then check for the ObjectGUID
> I'm looking for.
>
> ** **
>
> Ciaran
>
> ** **
>
> ** **
>
> On Tue, Jun 4, 2013 at 12:49 PM, Michal Doniec  wrote:*
> ***
>
> Hi,
>
> ** **
>
> Is there any obvious way I am missing which would allow me to identify
> objects in the scene, but not by name?
>
> Something like ObjectID, but persistent between scene saves and loads. I
> can generate UUID or something similar using "ObjectAdded" event and store
> custom property but I'd rather avoid it.
>
> ** **
>
> I want to be able to track objects regardless if they were renamed or not.
> 
>
> ** **
>
> --
> --
> Michal
> http://uk.linkedin.com/in/mdoniec 
>
> ** **
>

RE: Persistent objects ids

[Matt Lind]

Object GUID is persistent, but not granular enough.  I've run into cases where 
multiple items get the same GUID.   Same goes for CLSID.

Matt





From: softimage-boun...@listproc.autodesk.com 
[mailto:softimage-boun...@listproc.autodesk.com] On Behalf Of Ciaran Moloney
Sent: Tuesday, June 04, 2013 3:55 PM
To: softimage@listproc.autodesk.com
Subject: Re: Persistent objects ids

Yes, you're right...I can reproduce a change of ID when deleting an object and 
re-opening a scene. However, I would not dismiss the object GUID, which appears 
to persists between sessions despite the change in the object ID value. In the 
below snippet the sphere object changed ID but not GUID.

# Name: cylinder
# ID: 114
# GUID: {090F2A9A-D1FC-4112-ACC5-9345BEB48954}
#
# Name: disc
# ID: 92
# GUID: {D03EF191-3962-4540-929D-F5438CE59D79}
#
# Name: torus1
# ID: 68
# GUID: {1AC14CC7-1175-4E30-9CDD-BC0C0C1DE6EA}
#
# Name: sphere
# ID: 80
# GUID: {CC63558A-D5A3-4F55-8621-6D428FE2743A}
# 

# Delete cylinder, save and re-open scene

# Name: disc
# ID: 92
# GUID: {D03EF191-3962-4540-929D-F5438CE59D79}
#
# Name: torus1
# ID: 69
# GUID: {1AC14CC7-1175-4E30-9CDD-BC0C0C1DE6EA}
#
# Name: sphere
# ID: 99
# GUID: {CC63558A-D5A3-4F55-8621-6D428FE2743A}
#
# 



On Tue, Jun 4, 2013 at 6:38 PM, Matt Lind 
mailto:ml...@carbinestudios.com>> wrote:
This is not safe between sessions as IDs are dynamically generated and assigned 
in the order objects were created in the scene.  If an older object is deleted 
and the scene reopened, everything after it will have a new ID.

Matt



From: 
softimage-boun...@listproc.autodesk.com
 
[mailto:softimage-boun...@listproc.autodesk.com]
 On Behalf Of Ciaran Moloney
Sent: Tuesday, June 04, 2013 8:52 AM
To: softimage@listproc.autodesk.com
Subject: Re: Persistent objects ids

Hi,
did you try XSIUtils.DataRepository.GetIdentifier() passing the siObjectGUID 
argument?
I've used this to keep track of ICE nodes without relying on name strings. It's 
persistent between sessions and is preserved when models are exported.
Works like a charm.
The only problem is that it's possibly a bit tricky to get a pointer back to 
the object, since you need to know what you're looking for. I use 
Application.FindObjects() to get a collection of objects of this class, then 
check for the ObjectGUID I'm looking for.

Ciaran


On Tue, Jun 4, 2013 at 12:49 PM, Michal Doniec 
mailto:doni...@gmail.com>> wrote:
Hi,

Is there any obvious way I am missing which would allow me to identify objects 
in the scene, but not by name?
Something like ObjectID, but persistent between scene saves and loads. I can 
generate UUID or something similar using "ObjectAdded" event and store custom 
property but I'd rather avoid it.

I want to be able to track objects regardless if they were renamed or not.

--
--
Michal
http://uk.linkedin.com/in/mdoniec

Re: Persistent objects ids

[Ciaran Moloney]

So, if the GUID isn't unique, that's a pretty lousy algorithm
Luckily, I haven't observed that so far.

Ciaran

On Tue, Jun 4, 2013 at 11:57 PM, Matt Lind  wrote:

> Object GUID is persistent, but not granular enough.  I’ve run into cases
> where multiple items get the same GUID.   Same goes for CLSID.
>
> ** **
>
> Matt
>
> ** **
>
> ** **
>
> ** **
>
> ** **
>
> ** **
>
> *From:* softimage-boun...@listproc.autodesk.com [mailto:
> softimage-boun...@listproc.autodesk.com] *On Behalf Of *Ciaran Moloney
> *Sent:* Tuesday, June 04, 2013 3:55 PM
>
> *To:* softimage@listproc.autodesk.com
> *Subject:* Re: Persistent objects ids
>
> ** **
>
> Yes, you're right...I can reproduce a change of ID when deleting an object
> and re-opening a scene. However, I would not dismiss the object GUID, which
> appears to persists between sessions despite the change in the object ID
> value. In the below snippet the sphere object changed ID but not GUID.
>
> ** **
>
> # Name: cylinder
>
> # ID: 114
>
> # GUID: {090F2A9A-D1FC-4112-ACC5-9345BEB48954}
>
> #
>
> # Name: disc
>
> # ID: 92
>
> # GUID: {D03EF191-3962-4540-929D-F5438CE59D79}
>
> #
>
> # Name: torus1
>
> # ID: 68
>
> # GUID: {1AC14CC7-1175-4E30-9CDD-BC0C0C1DE6EA}
>
> #
>
> # Name: sphere
>
> # ID: 80
>
> # GUID: {CC63558A-D5A3-4F55-8621-6D428FE2743A}
>
> # 
>
> ** **
>
> # Delete cylinder, save and re-open scene 
>
> ** **
>
> # Name: disc
>
> # ID: 92
>
> # GUID: {D03EF191-3962-4540-929D-F5438CE59D79}
>
> #
>
> # Name: torus1
>
> # ID: 69
>
> # GUID: {1AC14CC7-1175-4E30-9CDD-BC0C0C1DE6EA}
>
> #
>
> # Name: sphere
>
> # ID: 99
>
> # GUID: {CC63558A-D5A3-4F55-8621-6D428FE2743A}
>
> #
>
> # 
>
> ** **
>
> ** **
>
>  
>
> On Tue, Jun 4, 2013 at 6:38 PM, Matt Lind 
> wrote:
>
> This is not safe between sessions as IDs are dynamically generated and
> assigned in the order objects were created in the scene.  If an older
> object is deleted and the scene reopened, everything after it will have a
> new ID.
>
>  
>
> Matt
>
>  
>
>  
>
>  
>
> *From:* softimage-boun...@listproc.autodesk.com [mailto:
> softimage-boun...@listproc.autodesk.com] *On Behalf Of *Ciaran Moloney
> *Sent:* Tuesday, June 04, 2013 8:52 AM
> *To:* softimage@listproc.autodesk.com
> *Subject:* Re: Persistent objects ids
>
>  
>
> Hi,
>
> did you try XSIUtils.DataRepository.GetIdentifier() passing the
> siObjectGUID argument?
>
> I've used this to keep track of ICE nodes without relying on name strings.
> It's persistent between sessions and is preserved when models are exported.
> 
>
> Works like a charm.
>
> The only problem is that it's possibly a bit tricky to get a pointer back
> to the object, since you need to know what you're looking for. I use 
> Application.FindObjects()
> to get a collection of objects of this class, then check for the ObjectGUID
> I'm looking for.
>
>  
>
> Ciaran
>
>  
>
>  
>
> On Tue, Jun 4, 2013 at 12:49 PM, Michal Doniec  wrote:*
> ***
>
> Hi,
>
>  
>
> Is there any obvious way I am missing which would allow me to identify
> objects in the scene, but not by name?
>
> Something like ObjectID, but persistent between scene saves and loads. I
> can generate UUID or something similar using "ObjectAdded" event and store
> custom property but I'd rather avoid it.
>
>  
>
> I want to be able to track objects regardless if they were renamed or not.
> 
>
>  
>
> --
> --
> Michal
> http://uk.linkedin.com/in/mdoniec 
>
>  
>
> ** **
>

Re: Persistent objects ids

[Raffaele Fragapane]

I doubt it's an issue with GUID at an algorithmic level, if two identical
GUID could be produced by the same person on the same project, and
therefore the conflict noticed, the windows world would be in very deep
S*** :)
Identical GUIDs on one platform should be centuries apart.

It's more likely a case of generation/re-use on Soft's side, either
intentional for non-obvious reasons or a bug.


On Wed, Jun 5, 2013 at 9:00 AM, Ciaran Moloney wrote:

> So, if the GUID isn't unique, that's a pretty lousy algorithm
> Luckily, I haven't observed that so far.
>
> Ciaran
>
>
> On Tue, Jun 4, 2013 at 11:57 PM, Matt Lind wrote:
>
>> Object GUID is persistent, but not granular enough.  I’ve run into cases
>> where multiple items get the same GUID.   Same goes for CLSID.
>>
>> ** **
>>
>> Matt
>>
>> ** **
>>
>> ** **
>>
>> ** **
>>
>> ** **
>>
>> ** **
>>
>> *From:* softimage-boun...@listproc.autodesk.com [mailto:
>> softimage-boun...@listproc.autodesk.com] *On Behalf Of *Ciaran Moloney
>> *Sent:* Tuesday, June 04, 2013 3:55 PM
>>
>> *To:* softimage@listproc.autodesk.com
>> *Subject:* Re: Persistent objects ids
>>
>> ** **
>>
>> Yes, you're right...I can reproduce a change of ID when deleting an
>> object and re-opening a scene. However, I would not dismiss the object
>> GUID, which appears to persists between sessions despite the change in the
>> object ID value. In the below snippet the sphere object changed ID but not
>> GUID.
>>
>> ** **
>>
>> # Name: cylinder
>>
>> # ID: 114
>>
>> # GUID: {090F2A9A-D1FC-4112-ACC5-9345BEB48954}
>>
>> #
>>
>> # Name: disc
>>
>> # ID: 92
>>
>> # GUID: {D03EF191-3962-4540-929D-F5438CE59D79}
>>
>> #
>>
>> # Name: torus1
>>
>> # ID: 68
>>
>> # GUID: {1AC14CC7-1175-4E30-9CDD-BC0C0C1DE6EA}
>>
>> #
>>
>> # Name: sphere
>>
>> # ID: 80
>>
>> # GUID: {CC63558A-D5A3-4F55-8621-6D428FE2743A}
>>
>> # 
>>
>> ** **
>>
>> # Delete cylinder, save and re-open scene 
>>
>> ** **
>>
>> # Name: disc
>>
>> # ID: 92
>>
>> # GUID: {D03EF191-3962-4540-929D-F5438CE59D79}
>>
>> #
>>
>> # Name: torus1
>>
>> # ID: 69
>>
>> # GUID: {1AC14CC7-1175-4E30-9CDD-BC0C0C1DE6EA}
>>
>> #
>>
>> # Name: sphere
>>
>> # ID: 99
>>
>> # GUID: {CC63558A-D5A3-4F55-8621-6D428FE2743A}
>>
>> #
>>
>> # 
>>
>> ** **
>>
>> ** **
>>
>>  
>>
>> On Tue, Jun 4, 2013 at 6:38 PM, Matt Lind 
>> wrote:
>>
>> This is not safe between sessions as IDs are dynamically generated and
>> assigned in the order objects were created in the scene.  If an older
>> object is deleted and the scene reopened, everything after it will have a
>> new ID.
>>
>>  
>>
>> Matt
>>
>>  
>>
>>  
>>
>>  
>>
>> *From:* softimage-boun...@listproc.autodesk.com [mailto:
>> softimage-boun...@listproc.autodesk.com] *On Behalf Of *Ciaran Moloney
>> *Sent:* Tuesday, June 04, 2013 8:52 AM
>> *To:* softimage@listproc.autodesk.com
>> *Subject:* Re: Persistent objects ids
>>
>>  
>>
>> Hi,
>>
>> did you try XSIUtils.DataRepository.GetIdentifier() passing the
>> siObjectGUID argument?
>>
>> I've used this to keep track of ICE nodes without relying on name
>> strings. It's persistent between sessions and is preserved when models are
>> exported. 
>>
>> Works like a charm.
>>
>> The only problem is that it's possibly a bit tricky to get a pointer back
>> to the object, since you need to know what you're looking for. I use 
>> Application.FindObjects()
>> to get a collection of objects of this class, then check for the ObjectGUID
>> I'm looking for.
>>
>>  
>>
>> Ciaran
>>
>>  
>>
>>  
>>
>> On Tue, Jun 4, 2013 at 12:49 PM, Michal Doniec  wrote:
>> 
>>
>> Hi,
>>
>>  
>>
>> Is there any obvious way I am missing which would allow me to identify
>> objects in the scene, but not by name?
>>
>> Something like ObjectID, but persistent between scene saves and loads. I
>> can generate UUID or something similar using "ObjectAdded" event and store
>> custom property but I'd rather avoid it.
>>
>>  
>>
>> I want to be able to track objects regardless if they were renamed or not.
>> 
>>
>>  
>>
>> --
>> --
>> Michal
>> http://uk.linkedin.com/in/mdoniec 
>>
>>  
>>
>> ** **
>>
>
>


-- 
Our users will know fear and cower before our software! Ship it! Ship it
and let them flee like the dogs they are!

Re: Flow along curve speed control

[John Clausing]

Thank you ill give it a shot
I had a "modulate by age" plugged into "set particle speed", which didn't work 
either, 
"Speed" vs "velocity" issue?

J

Sent from my iPhone

On Jun 4, 2013, at 7:39 PM, Vincent Ullmann  
wrote:

> Hi,
> 
> you might be right: The CurveTranslation could be the Problem. The 
> Factory-FlowAlongCurve-Compound gets the TargetPosition in LocalSpace
> 
> Here are two quick solutions:
> 
> 1: FlowToCurve:
> Its something like: Get your originalCurve-EmitPosition and try fly there.
> So its not right flowing along, more a soft-constrain like thing.
> 
> 2. FlowAlongCurvewithConstantSpeed
> simply, normalize the velocity, and multiply it by you desired speed.
> 
> Both solutions arent more or less a workaround and look aweful in my 
> testScene, but thats more an Issue of my "SinWave-AnimationSkills" :-)
> 
> 
> Am 04.06.2013 23:47, schrieb john clausing:
>> hello all,..
>> 
>> i have an animated curve (translating) that emits an ICE cloud by position.
>> the cloud travels by speed "from emit" on the curve with "flow on curve".
>> 
>> it all works fine with the exception of speed control..there is none, 
>> even when i change the speed in the "flow on curve" node.
>> i suspect it all has to do with the curve's translation, or with the 
>> "translating" curve doubling as the emitter.
>> 
>> any thoughts?
>> 
>> thank you as always,
>> 
>> john
> 
> 
> 

Re: Persistent objects ids

[Ciaran Moloney]

Yeah, I figured as much.
Not that I doubt Matt's observations, but it would be nice to have some
official word on the reliability of this method - unique ID's are a request
that come up all the time.
In the mean time I'll proceed with caution, since it's the simplest general
purpose solution.

Ciaran

On Wed, Jun 5, 2013 at 12:35 AM, Raffaele Fragapane <
raffsxsil...@googlemail.com> wrote:

> I doubt it's an issue with GUID at an algorithmic level, if two identical
> GUID could be produced by the same person on the same project, and
> therefore the conflict noticed, the windows world would be in very deep
> S*** :)
> Identical GUIDs on one platform should be centuries apart.
>
> It's more likely a case of generation/re-use on Soft's side, either
> intentional for non-obvious reasons or a bug.
>
>
> On Wed, Jun 5, 2013 at 9:00 AM, Ciaran Moloney 
> wrote:
>
>> So, if the GUID isn't unique, that's a pretty lousy algorithm
>> Luckily, I haven't observed that so far.
>>
>> Ciaran
>>
>>
>> On Tue, Jun 4, 2013 at 11:57 PM, Matt Lind wrote:
>>
>>> Object GUID is persistent, but not granular enough.  I’ve run into cases
>>> where multiple items get the same GUID.   Same goes for CLSID.
>>>
>>> ** **
>>>
>>> Matt
>>>
>>> ** **
>>>
>>> ** **
>>>
>>> ** **
>>>
>>> ** **
>>>
>>> ** **
>>>
>>> *From:* softimage-boun...@listproc.autodesk.com [mailto:
>>> softimage-boun...@listproc.autodesk.com] *On Behalf Of *Ciaran Moloney
>>> *Sent:* Tuesday, June 04, 2013 3:55 PM
>>>
>>> *To:* softimage@listproc.autodesk.com
>>> *Subject:* Re: Persistent objects ids
>>>
>>> ** **
>>>
>>> Yes, you're right...I can reproduce a change of ID when deleting an
>>> object and re-opening a scene. However, I would not dismiss the object
>>> GUID, which appears to persists between sessions despite the change in the
>>> object ID value. In the below snippet the sphere object changed ID but not
>>> GUID.
>>>
>>> ** **
>>>
>>> # Name: cylinder
>>>
>>> # ID: 114
>>>
>>> # GUID: {090F2A9A-D1FC-4112-ACC5-9345BEB48954}
>>>
>>> #
>>>
>>> # Name: disc
>>>
>>> # ID: 92
>>>
>>> # GUID: {D03EF191-3962-4540-929D-F5438CE59D79}
>>>
>>> #
>>>
>>> # Name: torus1
>>>
>>> # ID: 68
>>>
>>> # GUID: {1AC14CC7-1175-4E30-9CDD-BC0C0C1DE6EA}
>>>
>>> #
>>>
>>> # Name: sphere
>>>
>>> # ID: 80
>>>
>>> # GUID: {CC63558A-D5A3-4F55-8621-6D428FE2743A}
>>>
>>> # 
>>>
>>> ** **
>>>
>>> # Delete cylinder, save and re-open scene 
>>>
>>> ** **
>>>
>>> # Name: disc
>>>
>>> # ID: 92
>>>
>>> # GUID: {D03EF191-3962-4540-929D-F5438CE59D79}
>>>
>>> #
>>>
>>> # Name: torus1
>>>
>>> # ID: 69
>>>
>>> # GUID: {1AC14CC7-1175-4E30-9CDD-BC0C0C1DE6EA}
>>>
>>> #
>>>
>>> # Name: sphere
>>>
>>> # ID: 99
>>>
>>> # GUID: {CC63558A-D5A3-4F55-8621-6D428FE2743A}
>>>
>>> #
>>>
>>> # 
>>>
>>> ** **
>>>
>>> ** **
>>>
>>>  
>>>
>>> On Tue, Jun 4, 2013 at 6:38 PM, Matt Lind 
>>> wrote:
>>>
>>> This is not safe between sessions as IDs are dynamically generated and
>>> assigned in the order objects were created in the scene.  If an older
>>> object is deleted and the scene reopened, everything after it will have a
>>> new ID.
>>>
>>>  
>>>
>>> Matt
>>>
>>>  
>>>
>>>  
>>>
>>>  
>>>
>>> *From:* softimage-boun...@listproc.autodesk.com [mailto:
>>> softimage-boun...@listproc.autodesk.com] *On Behalf Of *Ciaran Moloney
>>> *Sent:* Tuesday, June 04, 2013 8:52 AM
>>> *To:* softimage@listproc.autodesk.com
>>> *Subject:* Re: Persistent objects ids
>>>
>>>  
>>>
>>> Hi,
>>>
>>> did you try XSIUtils.DataRepository.GetIdentifier() passing the
>>> siObjectGUID argument?
>>>
>>> I've used this to keep track of ICE nodes without relying on name
>>> strings. It's persistent between sessions and is preserved when models are
>>> exported. 
>>>
>>> Works like a charm.
>>>
>>> The only problem is that it's possibly a bit tricky to get a pointer
>>> back to the object, since you need to know what you're looking for. I use 
>>> Application.FindObjects()
>>> to get a collection of objects of this class, then check for the ObjectGUID
>>> I'm looking for.
>>>
>>>  
>>>
>>> Ciaran
>>>
>>>  
>>>
>>>  
>>>
>>> On Tue, Jun 4, 2013 at 12:49 PM, Michal Doniec 
>>> wrote:
>>>
>>> Hi,
>>>
>>>  
>>>
>>> Is there any obvious way I am missing which would allow me to identify
>>> objects in the scene, but not by name?
>>>
>>> Something like ObjectID, but persistent between scene saves and loads. I
>>> can generate UUID or something similar using "ObjectAdded" event and store
>>> custom property but I'd rather avoid it.
>>>
>>>  
>>>
>>> I want to be able to track objects regardless if they were renamed or
>>> not.
>>>
>>>  
>>>
>>> --
>>> --
>>> Michal
>>> http://uk.linkedin.com/in/mdoniec 
>>>
>>>  
>>>
>>> ** **
>>>
>>
>>
>
>
> --
> Our users 

Re: Persistent objects ids

[Raffaele Fragapane]

I tend to agree with Matt, and to be really honest granting UID persistency
in a large software is frequently done on the side, specifically with that
intention, and either user controlled, or if left to heuristics either very
conservative or very changing (IE: at what point do you decide to lock an
ID? On creaiton? On Topo change? What happens if you swap the primitive
like you can do with Curves?).

Your best bet is emulating the above yourself by tagging objects at the
stages you want to ID them, and changing it if and when you need it. It's
actually not hard to do in terms of difficulty, but it can result in a
large and pervasive toolset before it's reliable enough, which is probably
why Michal outlined he'd rather avoid it.
It's what we do in a couple places actually to support name agnostic
operations in our pipe (plenty renaming happening since, in example, the
same rig can be imported several times for different characters).

I'm afraid there's little alternative to controlling this yourself, and to
be honest this is a rather generig CS problem more so than it's a Softimage
specific one.
The Identity of an item is a very ambiguous and volatile concept that every
user and pipeline will have a different idea of the extent of.

What Soft might want to do is offer an open parameter with a creation ID,
that way there's always something fairly static couple with the object, and
leave it to the user to modify it if they require more frequent mutation
than the first time the primitive comes into existence. This will need to
be wrapped around by every command offering creation of sorts. Gets,
Extracts, Duplicates and so on.
Log the request if you feel it's important.


On Wed, Jun 5, 2013 at 9:49 AM, Ciaran Moloney wrote:

> Yeah, I figured as much.
> Not that I doubt Matt's observations, but it would be nice to have some
> official word on the reliability of this method - unique ID's are a request
> that come up all the time.
> In the mean time I'll proceed with caution, since it's the simplest
> general purpose solution.
>
> Ciaran
>
>
> On Wed, Jun 5, 2013 at 12:35 AM, Raffaele Fragapane <
> raffsxsil...@googlemail.com> wrote:
>
>> I doubt it's an issue with GUID at an algorithmic level, if two identical
>> GUID could be produced by the same person on the same project, and
>> therefore the conflict noticed, the windows world would be in very deep
>> S*** :)
>> Identical GUIDs on one platform should be centuries apart.
>>
>> It's more likely a case of generation/re-use on Soft's side, either
>> intentional for non-obvious reasons or a bug.
>>
>>
>> On Wed, Jun 5, 2013 at 9:00 AM, Ciaran Moloney 
>> wrote:
>>
>>> So, if the GUID isn't unique, that's a pretty lousy algorithm
>>> Luckily, I haven't observed that so far.
>>>
>>> Ciaran
>>>
>>>
>>> On Tue, Jun 4, 2013 at 11:57 PM, Matt Lind wrote:
>>>
 Object GUID is persistent, but not granular enough.  I’ve run into
 cases where multiple items get the same GUID.   Same goes for CLSID.***
 *

 ** **

 Matt

 ** **

 ** **

 ** **

 ** **

 ** **

 *From:* softimage-boun...@listproc.autodesk.com [mailto:
 softimage-boun...@listproc.autodesk.com] *On Behalf Of *Ciaran Moloney
 *Sent:* Tuesday, June 04, 2013 3:55 PM

 *To:* softimage@listproc.autodesk.com
 *Subject:* Re: Persistent objects ids

 ** **

 Yes, you're right...I can reproduce a change of ID when deleting an
 object and re-opening a scene. However, I would not dismiss the object
 GUID, which appears to persists between sessions despite the change in the
 object ID value. In the below snippet the sphere object changed ID but not
 GUID.

 ** **

 # Name: cylinder

 # ID: 114

 # GUID: {090F2A9A-D1FC-4112-ACC5-9345BEB48954}

 #

 # Name: disc

 # ID: 92

 # GUID: {D03EF191-3962-4540-929D-F5438CE59D79}

 #

 # Name: torus1

 # ID: 68

 # GUID: {1AC14CC7-1175-4E30-9CDD-BC0C0C1DE6EA}

 #

 # Name: sphere

 # ID: 80

 # GUID: {CC63558A-D5A3-4F55-8621-6D428FE2743A}

 # 

 ** **

 # Delete cylinder, save and re-open scene 

 ** **

 # Name: disc

 # ID: 92

 # GUID: {D03EF191-3962-4540-929D-F5438CE59D79}

 #

 # Name: torus1

 # ID: 69

 # GUID: {1AC14CC7-1175-4E30-9CDD-BC0C0C1DE6EA}

 #

 # Name: sphere

 # ID: 99

 # GUID: {CC63558A-D5A3-4F55-8621-6D428FE2743A}

 #

 # 

 ** **

 ** **

  

 On Tue, Jun 4, 2013 at 6:38 PM, Matt Lind 
 wrote:

 This is not safe between sessions as IDs are dynamically

Re: Compositing Levels

[Christopher]

Thanks for the suggestions and tips, very helpful :) I have composite
installed, I wonder if I should even use After Effects if working in
linear is so much problem, rather go between FXTree and/or composite,
composite for 2013 is probably the most stable, I can't speak for 2014.
(did this message go though, lately having server problems --- delay ?)
::Christopher::

Rob Wuijster wrote:
> You could use AE, but it's still a pain in the behind. Adobe just
> doesn't get linear workflow.
>
> And PS might be better in CS4,5,whatever, it's no compositor.
> You can try the FXTree, or install Composite, it comes with Softimage.
>
> Other than that you can try Blender, it has a pretty decent compositor
> for a free app.
> Nuke, or Fusion are too expensive for most people, despite being the
> better apps for compositing.
> Rob
>
> \/-\/\/
> On 3-6-2013 18:50, pete...@skynet.be wrote:
>> changing the image/look radically.
>>  
>> Adding a ton of technical passes together in a “correct” way –
>> creating the look in compositing and then heavily
>> modifying/manipulating it is a minefield.
>> I would only recommend Nuke for that.
>>  
>> If you just put the layers together – with a few small changes
>> perhaps – than FXtree is just fine.
>>  
>> *From:* Christopher 
>> *Sent:* Monday, June 03, 2013 6:30 PM
>> *To:* softimage@listproc.autodesk.com
>> 
>> *Subject:* Re: Compositing Levels
>>  
>> I have never touched the FXTree, I will start now with this comp :)
>> What do you mean transform the image ?
>>
>> ::Christopher::
>>
>> pete...@skynet.be wrote:
>>> better not ask my opinion on any adobe product – at least I can see
>>> some use for Photoshop – but compositing it aint.
>>>  
>>> I whish I could recommend FXtree without caveats – as you have it
>>> right there in XSI and to a degree it will work - quite well even.
>>> But there are some pitfalls. Clipping is one of them – and it
>>> doesn’t have display gamma afaik – so you’ll have to resort to
>>> adding a 2.2 gamma in the end of the comp (not on the luma but on R
>>> G B each!)
>>> Last I did a multi-channel-comp in FXtree compared to beauty render
>>> – I got it almost perfect. For a simple case that is.
>>> If you don’t intend totally transform the image it’s probably fine –
>>> and by any means, much less of a headache than Photoshop.
>>>  
>>> *From:* Christopher 
>>> *Sent:* Monday, June 03, 2013 6:07 PM
>>> *To:* softimage@listproc.autodesk.com
>>> 
>>> *Subject:* Re: Compositing Levels
>>>  
>>> For a program that cost $1K such as Ps it's quite restricted for
>>> this type of work, I still wonder why I have a zillion color
>>> profiles in Photoshop.  I imported the levels as linear gamma, but
>>> upon import in photoshop I let photoshop apply sRGB (2.2 gamma), I
>>> suppose I shouldn't.  You make it sound as though I will fight will
>>> Ps to get things right, although you probably are right :)
>>>  
>>> I'm going to do the comp all in linear, see what results I get.  I
>>> need to find an alternative to Photoshop, Nuke I don't use here.  Is
>>> AE more better then Ps, or more of the same ?
>>>
>>> ::Christopher::
>>>
>>> pete...@skynet.be wrote:
 > If the above is correct, what is ticking me off is it is so dark,
 when I render in the renderer, it's nice,...
  
 are you looking at it with proper gamma correction?
 oh right it’s photoshop...
  
 well, in XSI you are probably seeing the linear image with a gamma
 correction added (which is the desired way to work) – but most
 other software will show it without gamma - So if you add a gamma
 of 2.2, chances are it will look like what you expected.
 But here’s the twist: you should not be using that gamma – first do
 the comp and add it in the end. Or actually – better not add the
 gamma at all and rather export the image to the correct target
 color space.
  
 > Photoshop 'add' blend mode doesn't work in 32-bit color space,
 what is the correct alternative ?

 Nuke.
  
 Really – you can’t expect to use Photoshop for a comp like that and
 have the same result as in the renderer – it has the maths all
 wrong and doesn’t even know how to handle an alpha.
 What you’re trying to do will work correctly the first time around
 you try it in nuke (ok – perhaps that’s wishful thinking)
  
 Add (or Plus in Nuke terms) is the one and only proper blend mode
 to use - it’s called linear dodge or something senseless like that
 in Photoshop I think.
  
 Screen is not correct - though handy at times, it will never give
 you the same result as in the renderer. (except for the speculars
 in the mr skin shader – but that’s another story)
  
 Seriously – photoshop isn’t worth all the pain for this kind