Hi,
I have downloaded some nice (and very complex) furniture-geometry from the internet that I like to render in Radiance, but when I create an octree I get the "set overflow in addobject" error. The -r -and -n option do not help.
Is there a way to just skip the 'wrong' geometry (that is causing the error) in oconv ?
Any other options (instead of modelling the furniture myself) ?
-Iebele
have you tried with mesh?
(obj2mesh)
-----Original Message-----
From: [email protected]
[mailto:[email protected]] On Behalf Of
atelier iebele abel
Sent: 21 November 2005 16:18
To: Radiance general discussion
Subject: [Radiance-general] set overflow
Hi,
I have downloaded some nice (and very complex) furniture-geometry from
the internet that I like to render in Radiance, but when I create an
octree I get the "set overflow in addobject" error. The -r -and -n
option do not help.
Is there a way to just skip the 'wrong' geometry (that is causing the
error) in oconv ?
Any other options (instead of modelling the furniture myself) ?
-Iebele
_______________________________________________
Radiance-general mailing list
[email protected]
http://www.radiance-online.org/mailman/listinfo/radiance-general
____________________________________________________________
Electronic mail messages entering and leaving Arup business
systems are scanned for acceptability of content and viruses
atelier iebele abel wrote:
Hi,
I have downloaded some nice (and very complex) furniture-geometry from the internet that I like to render in Radiance, but when I create an octree I get the "set overflow in addobject" error. The -r -and -n option do not help.
What do you mean by "the -r option does not work"? I have had to set -r pretty darned high sometimes, to get big models with fine detail to compile. You get a large octree, and it takes a while to generate, but it works. By "pretty darned high", I am talking about -r as high as one million.
Hey all,
There are a couple of things here. You should probably try to take a look at the geometry and see if there is any extra/duplicate geomery. Doubled up geomery can oftern be a cause for a set overflow like this. You should be able to control a lot with -r.
Guilio's suggestion is also an excellent one. I have found that I can deal with much more complex geometry sets with obj2mesh. And it looks like the file size of the rtm is typically smaller than a corresponding oct file.
-Jack
Rob Guglielmetti wrote:
atelier iebele abel wrote:
Hi,
I have downloaded some nice (and very complex) furniture-geometry from the internet that I like to render in Radiance, but when I create an octree I get the "set overflow in addobject" error. The -r -and -n option do not help.
What do you mean by "the -r option does not work"? I have had to set -r pretty darned high sometimes, to get big models with fine detail to compile. You get a large octree, and it takes a while to generate, but it works. By "pretty darned high", I am talking about -r as high as one million.
_______________________________________________
Radiance-general mailing list
[email protected]
http://www.radiance-online.org/mailman/listinfo/radiance-general
--
# Jack de Valpine
# president
#
# visarc incorporated
# http://www.visarc.com
#
# channeling technology for superior design and construction
Hi Jack and Rob,
I've tried both your suggestions. Helas, it did not work. I can't spot the error, so I have to re-model the geometry.
Thanks for your support!
Iebele
Jack de Valpine wrote:
Hey all,
There are a couple of things here. You should probably try to take a look at the geometry and see if there is any extra/duplicate geomery. Doubled up geomery can oftern be a cause for a set overflow like this. You should be able to control a lot with -r.
Guilio's suggestion is also an excellent one. I have found that I can deal with much more complex geometry sets with obj2mesh. And it looks like the file size of the rtm is typically smaller than a corresponding oct file.
-Jack
Rob Guglielmetti wrote:
atelier iebele abel wrote:
Hi,
I have downloaded some nice (and very complex) furniture-geometry from the internet that I like to render in Radiance, but when I create an octree I get the "set overflow in addobject" error. The -r -and -n option do not help.
What do you mean by "the -r option does not work"? I have had to set -r pretty darned high sometimes, to get big models with fine detail to compile. You get a large octree, and it takes a while to generate, but it works. By "pretty darned high", I am talking about -r as high as one million.
_______________________________________________
Radiance-general mailing list
[email protected]
http://www.radiance-online.org/mailman/listinfo/radiance-general
Hi Iebele,
The set overflow error includes a hint as to the problem in the form of a surface id. Most likely, the trouble in your geometry is near this surface. If it's not overlapping or collapsed geometry, then it might be a disk that's been converted to hundreds of triangles, all culminating at a single point, which cannot be resolved by oconv because there are more than 511 surfaces in an infinitesimal volume.
If you can identify all the triangles associated with this disk, you should be able to replace it with a single "ring" primitive, or if the edges need to match exactly, with a many-sided face primitive.
-Greg
From: atelier iebele abel <[email protected]>
Date: November 28, 2005 4:10:44 AM PSTHi Jack and Rob,
I've tried both your suggestions. Helas, it did not work. I can't spot the error, so I have to re-model the geometry.
Thanks for your support!
Iebele
Hi Greg,
Indeed, there are triangulated rings in the geometry. These triangulated rings are easy to identify.
Great! Thanks!
-Iebele
Greg Ward wrote:
Hi Iebele,
The set overflow error includes a hint as to the problem in the form of a surface id. Most likely, the trouble in your geometry is near this surface. If it's not overlapping or collapsed geometry, then it might be a disk that's been converted to hundreds of triangles, all culminating at a single point, which cannot be resolved by oconv because there are more than 511 surfaces in an infinitesimal volume.
If you can identify all the triangles associated with this disk, you should be able to replace it with a single "ring" primitive, or if the edges need to match exactly, with a many-sided face primitive.
-Greg
From: atelier iebele abel <[email protected]>
Date: November 28, 2005 4:10:44 AM PSTHi Jack and Rob,
I've tried both your suggestions. Helas, it did not work. I can't spot the error, so I have to re-model the geometry.
Thanks for your support!
Iebele
_______________________________________________
Radiance-general mailing list
[email protected]
http://www.radiance-online.org/mailman/listinfo/radiance-general