[ICE] curve manipulation issues

32 views
Skip to first unread message

Martin Chatterjee

unread,
May 7, 2012, 5:11:16 AM5/7/12
to soft...@listproc.autodesk.com
Hello clever people out there,

I'm experiencing some annoying issues when trying to deal with Curve Manipulation in ICE - hopefully somebody can help me out here with some insight?


1.) I am looking for the Curve equivalent to self.ID (for PointClouds)  or self.VertexIndex (for polyMeshes). Bottom line I want to be able to access a Curve Point Location in the same way I can access PointCloud points via "ID to Location" or Polymesh vertices via "PointIndex To Location"


2.) While attempting to workaround this issue I have stumbled over this: (see attached screenshot)

When trying to get a full array of all Point Locations of a curve via "Get Closest Points" I NEVER get the correct point count...

When the curve is closed, it will give me one more, when the curve is open I will get two less than the actual point count.

Has anybody out there succeeded in working around this somehow?

Thanks a lot in advance for any hints,

Cheers, Martin 

--
        Martin Chatterjee

[ Freelance Technical Director ]
[    http://www.chatterjee.de    ]

ICE_Curve_Issues.png

Sebastian Kowalski

unread,
May 7, 2012, 6:32:24 AM5/7/12
to soft...@listproc.autodesk.com
hey martin,
see attached screenshot, is this what you loooking for?
curve_id.JPG

Chris Chia

unread,
May 7, 2012, 6:33:44 AM5/7/12
to soft...@listproc.autodesk.com
Hi Martin,
I think you are looking for this.
You can use "Select in Array" ice node to query the individual vertex's position.

[cid:c02c4ea5-dd64-448b-91cb-a7022225c756]




Regards,

Chris Chia

Softimage QA Analyst

Autodesk Media & Entertainment


________________________________
From: softimag...@listproc.autodesk.com [softimag...@listproc.autodesk.com] on behalf of Martin Chatterjee [martin.chat...@googlemail.com]
Sent: Monday, May 07, 2012 5:11 PM
To: soft...@listproc.autodesk.com
Subject: [ICE] curve manipulation issues
curve_pointPosition.gif

Martin Chatterjee

unread,
May 7, 2012, 6:59:27 AM5/7/12
to soft...@listproc.autodesk.com
Hey Sebastian, hey Chris,

awesome, both solutions work perfectly - cheers for that! :-)

Wasn't aware at all of both "Get Element Index" and "Build Array From Set".  

Thanks again,

Martin
--
        Martin Chatterjee

[ Freelance Technical Director ]
[    http://www.chatterjee.de    ]



Chris Chia

unread,
May 7, 2012, 8:17:03 AM5/7/12
to soft...@listproc.autodesk.com
Hi Martin,
>From what i know, "Get Closest Points" works for point cloud data.
If you need to find the "closest point position to a geometry", you can do something like my screenshot below.
I have used null as the geometry.

I am sure others can do a simpler ice tree. But I believe this one is more comprehensive.

[cid:56b2533f-ae43-45f5-952c-05915526f22d]




Regards,
Chris Chia
Softimage QA Analyst
Autodesk Media & Entertainment

________________________________
From: martin.c...@googlemail.com [martin.c...@googlemail.com] on behalf of Martin Chatterjee [martin.chat...@googlemail.com]
Sent: Monday, May 07, 2012 7:27 PM
To: Chris Chia
Subject: Re: [ICE] curve manipulation issues

Hi Chris,

thanks again for the quick reply on the list.

For the sake of completeness I'd love to see the faulty behaviour of the "Get Closest Points" node logged as a bug.

It's been a while - in fact I guess this might be my first official bug report since the aquisition...

Soo, what is the correct/easiest way of formally logging a bug?

Cheers, Martin
--
Martin Chatterjee

[ Freelance Technical Director ]
[ http://www.chatterjee.de ]



On Mon, May 7, 2012 at 12:33 PM, Chris Chia <chris...@autodesk.com<mailto:chris...@autodesk.com>> wrote:
Hi Martin,
I think you are looking for this.
You can use "Select in Array" ice node to query the individual vertex's position.

[cid:c02c4ea5-dd64-448b-91cb-a7022225c756]




Regards,

Chris Chia

Softimage QA Analyst

Autodesk Media & Entertainment


________________________________
From: softimag...@listproc.autodesk.com<mailto:softimag...@listproc.autodesk.com> [softimag...@listproc.autodesk.com<mailto:softimag...@listproc.autodesk.com>] on behalf of Martin Chatterjee [martin.chat...@googlemail.com<mailto:martin.chat...@googlemail.com>]
Sent: Monday, May 07, 2012 5:11 PM
To: soft...@listproc.autodesk.com<mailto:soft...@listproc.autodesk.com>
find_closest_in_array.gif

Martin Chatterjee

unread,
May 7, 2012, 9:46:08 AM5/7/12
to soft...@listproc.autodesk.com
Hi Chris,

well, I don't want to be a pain - but I've used "Get Closest Points" quite often and very successfully with both PointClouds and PolyMeshes.

I "mis-used" it to get an array of all Point Locations in the per-object context (as I didn't know "Build Array From Set"  ... :-/ )

This node works for Polymeshes and Point Clouds, and sort-of-works for Curves - but with wrong results.

So my reasoning is that this is a bug that should be logged.

I personally don't see a reason why this should work on polymeshes but not on curves. But if that is the case then the node should turn red and not work at all on Curves.

But regardless of this specific issue - what is the best way to formally log a bug nowadays?

Thanks, Martin
--
        Martin Chatterjee

[ Freelance Technical Director ]
[    http://www.chatterjee.de    ]



Martin Chatterjee

unread,
May 7, 2012, 10:29:46 AM5/7/12
to soft...@listproc.autodesk.com
Excellent, thanks Chris.

Cheers, Martin
--
        Martin Chatterjee

[ Freelance Technical Director ]
[    http://www.chatterjee.de    ]



On Mon, May 7, 2012 at 4:07 PM, Chris Chia <chris...@autodesk.com> wrote:
Hi Martin,
I will bring this up to the development team to see how it should support with the curves.

I will handle the filing of defects. If you ever find any defects in the future, you can report it here in this mailing list.

Cheers.


Regards,
Chris Chia



On 7 May, 2012, at 9:46 PM, "Martin Chatterjee" <martin.chat...@googlemail.com<mailto:martin.chat...@googlemail.com>> wrote:

Hi Chris,

well, I don't want to be a pain - but I've used "Get Closest Points" quite often and very successfully with both PointClouds and PolyMeshes.

I "mis-used" it to get an array of all Point Locations in the per-object context (as I didn't know "Build Array From Set"  ... :-/ )

This node works for Polymeshes and Point Clouds, and sort-of-works for Curves - but with wrong results.

So my reasoning is that this is a bug that should be logged.

I personally don't see a reason why this should work on polymeshes but not on curves. But if that is the case then the node should turn red and not work at all on Curves.

But regardless of this specific issue - what is the best way to formally log a bug nowadays?

Thanks, Martin
--
       Martin Chatterjee

[ Freelance Technical Director ]
[    http://www.chatterjee.de    ]



On Mon, May 7, 2012 at 2:17 PM, Chris Chia <chris...@autodesk.com<mailto:chris...@autodesk.com>> wrote:
Hi Martin,
>From what i know, "Get Closest Points" works for point cloud data.
If you need to find the "closest point position to a geometry", you can do something like my screenshot below.
I have used null as the geometry.

I am sure others can do a simpler ice tree. But I believe this one is more comprehensive.

[cid:56b2533f-ae43-45f5-952c-05915526f22d]




Regards,
Chris Chia
Softimage QA Analyst
Autodesk Media & Entertainment

________________________________

Sent: Monday, May 07, 2012 7:27 PM
To: Chris Chia
Subject: Re: [ICE] curve manipulation issues

Hi Chris,

thanks again for the quick reply on the list.

For the sake of completeness I'd love to see the faulty behaviour of the "Get Closest Points" node logged as a bug.

It's been a while - in fact I guess this might be my first official bug report since the aquisition...

Soo, what is the correct/easiest way of formally logging a bug?

Cheers, Martin
--
      Martin Chatterjee

[ Freelance Technical Director ]
[    http://www.chatterjee.de    ]



On Mon, May 7, 2012 at 12:33 PM, Chris Chia <chris...@autodesk.com<mailto:chris...@autodesk.com><mailto:chris...@autodesk.com<mailto:chris...@autodesk.com>>> wrote:
Hi Martin,
I think you are looking for this.
You can use "Select in Array" ice node to query the individual vertex's position.

[cid:c02c4ea5-dd64-448b-91cb-a7022225c756]




Regards,

Chris Chia

Softimage QA Analyst

Autodesk Media & Entertainment


________________________________

Sent: Monday, May 07, 2012 5:11 PM

Chris Chia

unread,
May 7, 2012, 10:07:38 AM5/7/12
to <softimage@listproc.autodesk.com>
Hi Martin,
I will bring this up to the development team to see how it should support with the curves.

I will handle the filing of defects. If you ever find any defects in the future, you can report it here in this mailing list.

Cheers.


Regards,
Chris Chia


On 7 May, 2012, at 9:46 PM, "Martin Chatterjee" <martin.chat...@googlemail.com<mailto:martin.chat...@googlemail.com>> wrote:

Hi Chris,

well, I don't want to be a pain - but I've used "Get Closest Points" quite often and very successfully with both PointClouds and PolyMeshes.

I "mis-used" it to get an array of all Point Locations in the per-object context (as I didn't know "Build Array From Set" ... :-/ )

This node works for Polymeshes and Point Clouds, and sort-of-works for Curves - but with wrong results.

So my reasoning is that this is a bug that should be logged.

I personally don't see a reason why this should work on polymeshes but not on curves. But if that is the case then the node should turn red and not work at all on Curves.

But regardless of this specific issue - what is the best way to formally log a bug nowadays?

Thanks, Martin
--
Martin Chatterjee

[ Freelance Technical Director ]
[ http://www.chatterjee.de ]



On Mon, May 7, 2012 at 2:17 PM, Chris Chia <chris...@autodesk.com<mailto:chris...@autodesk.com>> wrote:
Hi Martin,
>From what i know, "Get Closest Points" works for point cloud data.
If you need to find the "closest point position to a geometry", you can do something like my screenshot below.
I have used null as the geometry.

I am sure others can do a simpler ice tree. But I believe this one is more comprehensive.

[cid:56b2533f-ae43-45f5-952c-05915526f22d]




Regards,
Chris Chia
Softimage QA Analyst
Autodesk Media & Entertainment

________________________________
From: martin.c...@googlemail.com<mailto:martin.c...@googlemail.com> [martin.c...@googlemail.com<mailto:martin.c...@googlemail.com>] on behalf of Martin Chatterjee [martin.chat...@googlemail.com<mailto:martin.chat...@googlemail.com>]
Sent: Monday, May 07, 2012 7:27 PM
To: Chris Chia
Subject: Re: [ICE] curve manipulation issues

Hi Chris,

thanks again for the quick reply on the list.

For the sake of completeness I'd love to see the faulty behaviour of the "Get Closest Points" node logged as a bug.

It's been a while - in fact I guess this might be my first official bug report since the aquisition...

Soo, what is the correct/easiest way of formally logging a bug?

Cheers, Martin
--
Martin Chatterjee

[ Freelance Technical Director ]
[ http://www.chatterjee.de ]



On Mon, May 7, 2012 at 12:33 PM, Chris Chia <chris...@autodesk.com<mailto:chris...@autodesk.com><mailto:chris...@autodesk.com<mailto:chris...@autodesk.com>>> wrote:
Hi Martin,
I think you are looking for this.
You can use "Select in Array" ice node to query the individual vertex's position.

[cid:c02c4ea5-dd64-448b-91cb-a7022225c756]




Regards,

Chris Chia

Softimage QA Analyst

Autodesk Media & Entertainment


________________________________
From: softimag...@listproc.autodesk.com<mailto:softimag...@listproc.autodesk.com><mailto:softimag...@listproc.autodesk.com<mailto:softimag...@listproc.autodesk.com>> [softimag...@listproc.autodesk.com<mailto:softimag...@listproc.autodesk.com><mailto:softimag...@listproc.autodesk.com<mailto:softimag...@listproc.autodesk.com>>] on behalf of Martin Chatterjee [martin.chat...@googlemail.com<mailto:martin.chat...@googlemail.com><mailto:martin.chat...@googlemail.com<mailto:martin.chat...@googlemail.com>>]
Sent: Monday, May 07, 2012 5:11 PM
To: soft...@listproc.autodesk.com<mailto:soft...@listproc.autodesk.com><mailto:soft...@listproc.autodesk.com<mailto:soft...@listproc.autodesk.com>>
winmail.dat
Reply all
Reply to author
Forward
0 new messages