I'm getting sporatic errrors when I'm using the instance primitive...
"Modifier in tree; octree stale?"
Apperantly, it says I'm using a modifier in the instance that was already
defined in a .rad?
My rif file list these scenes
scene= overcast.rad - # lighting
material= Officemat.rad #materials only
scene= OfficestallNC.rad #geometry only
scene= memberservices.rad - # contains 2 instances deskdude.oct and
helpcube.oct
# deskdude.oct contains deskdude.rad
# helpcube.oct contains basicdude0.rad and cube.rad
I have looked through to make sure I'm not duplicating (or relying on)
previous modifiers
I'm not getting what I might be missing - the render will work for a while -
then I change a viewpoint and restart the rif - that's when I get the error.
Any Ideas or tips?
Thanks Rob F
Hi Rob,
The error message is probably due to using an unfrozen octree for one (or more) of your instance(s). This particular error is rare, because it requires geometry and materials somehow be reordered in the input, and usually one takes stuff out or adds it, which results in another message, "bad object count." The solution in either case is to apply the -f option when you create the octree(s) you use for your instance(s). This avoids problems when you move the instance octree(s) to a library location, and speeds their loading.
-Greg
From: "Fitzsimmons, Rob" <[email protected]>
Date: March 23, 2004 4:34:14 PM PST
I'm getting sporatic errrors when I'm using the instance primitive...
"Modifier in tree; octree stale?"
Apperantly, it says I'm using a modifier in the instance that was already defined in a .rad?
My rif file list these scenes
scene= overcast.rad - # lighting
material= Officemat.rad #materials only
scene= OfficestallNC.rad #geometry only
scene= memberservices.rad - # contains 2 instances deskdude.oct and helpcube.oct
# deskdude.oct contains deskdude.rad
# helpcube.oct contains basicdude0.rad and cube.rad
I have looked through to make sure I'm not duplicating (or relying on) previous modifiers
I'm not getting what I might be missing - the render will work for a while - then I change a viewpoint and restart the rif - that's when I get the error.
Any Ideas or tips?
Thanks Rob F
Hi Rob,
It is possible that some of the octrees that define your instances need to be recompiled (w/ oconv) because of changes or updates to materials or geometry relating to the instance object. This could occur if you are using one master material file or sharing multiple material files within your scene. A reason that this may only be occuring sporadically as an error is that for views where the instances are not in the view then rays are not going to be traced to the instance object.
If what I have indicated is in fact the basis for the error, then one way to avoid this is to comile your instance objects with the -f option (this in effect binds everything into the octree ) making the object self sustaining. The potential downside to this is if these instance are going to be dependant on changes and upadates to materials then you need to remember to recompile them if you make changes to the material files (that is additions or deletions of material information, changes to current material parameters are fine).
Hope this helps,
-Jack
Fitzsimmons, Rob wrote:
···
I'm getting sporatic errrors when I'm using the instance primitive...
"Modifier in tree; octree stale?"
Apperantly, it says I'm using a modifier in the instance that was already defined in a .rad?
My rif file list these scenes
scene= overcast.rad - # lighting
material= Officemat.rad #materials only
scene= OfficestallNC.rad #geometry only
scene= memberservices.rad - # contains 2 instances deskdude.oct and helpcube.oct
# deskdude.oct contains deskdude.rad
# helpcube.oct contains basicdude0.rad and cube.rad
I have looked through to make sure I'm not duplicating (or relying on) previous modifiers
I'm not getting what I might be missing - the render will work for a while - then I change a viewpoint and restart the rif - that's when I get the error.
Any Ideas or tips?
Thanks Rob F
------------------------------------------------------------------------
_______________________________________________
Radiance-general mailing list
[email protected]
http://www.radiance-online.org/mailman/listinfo/radiance-general
--
# John E. de Valpine
# president
#
# visarc incorporated
# http://www.visarc.com
#
# channeling technology for superior design and construction
Thanks Greg and Jack
I'm using -f and all is fine
···
-----Original Message-----
From: [email protected]
[mailto:[email protected]] On Behalf Of Jack de
Valpine
Sent: Wednesday, March 24, 2004 6:19 AM
To: Radiance general discussion
Subject: Re: [Radiance-general] instances
Hi Rob,
It is possible that some of the octrees that define your instances need to
be recompiled (w/ oconv) because of changes or updates to materials or
geometry relating to the instance object. This could occur if you are using
one master material file or sharing multiple material files within your
scene. A reason that this may only be occuring sporadically as an error is
that for views where the instances are not in the view then rays are not
going to be traced to the instance object.
If what I have indicated is in fact the basis for the error, then one way to
avoid this is to comile your instance objects with the -f option (this in
effect binds everything into the octree ) making the object self sustaining.
The potential downside to this is if these instance are going to be
dependant on changes and upadates to materials then you need to remember to
recompile them if you make changes to the material files (that is additions
or deletions of material information, changes to current material parameters
are fine).
Hope this helps,
-Jack
Fitzsimmons, Rob wrote:
I'm getting sporatic errrors when I'm using the instance primitive...
"Modifier in tree; octree stale?"
Apperantly, it says I'm using a modifier in the instance that was already
defined in a .rad?
My rif file list these scenes
scene= overcast.rad - # lighting
material= Officemat.rad #materials only
scene= OfficestallNC.rad #geometry only
scene= memberservices.rad - # contains 2 instances deskdude.oct and
helpcube.oct
# deskdude.oct contains deskdude.rad
# helpcube.oct contains basicdude0.rad and cube.rad
I have looked through to make sure I'm not duplicating (or relying on)
previous modifiers
I'm not getting what I might be missing - the render will work for a while -
then I change a viewpoint and restart the rif - that's when I get the error.
Any Ideas or tips?
Thanks Rob F
_____
_______________________________________________
Radiance-general mailing list
[email protected]
<mailto:[email protected]>
http://www.radiance-online.org/mailman/listinfo/radiance-general
<http://www.radiance-online.org/mailman/listinfo/radiance-general>
--
# John E. de Valpine
# president
#
# visarc incorporated
# http://www.visarc.com
#
# channeling technology for superior design and construction
Hi Giulio,
Many thanks for your reply, though I haven't been able to use:
void instance 001
9 ./Oct/s_enn.oct -a 1 -t 0 0 0.0064 -i 200
0
0
Also, still a bit in the dark concering your meta suggestion.
Meta isn't a primitive in Radiance, or is it?
Actually I know it only as a graphics format in Radiance?
Hopefully I'm seriously overlooking something here?
Cheers,
-jelle
I guess this is the trouble,
-a 1 -t..... -i 200
a missing 1
ciao
G.
apologies for my 1st answer, it was wrong indeed 
see the reply from greg and jack... as a guidance on how to do it
properly
when I said 'meta' I was trying to say that you may group few louvers
together, may be a full high module and replicate this instead of
replicating the single louvers... nothing rocket science...
let me know if you manage.
ciao
G.
PS alternatively when the mesh is really big I use a mesh primitive...
but I guess you know already about.
···
-----Original Message-----
From: [email protected]
[mailto:[email protected]] On Behalf Of Jelle
Feringa / EZCT Architecture & Design Research
Sent: 26 September 2006 09:16
To: [email protected]
Subject: [Radiance-general] instances
Hi Giulio,
Many thanks for your reply, though I haven't been able to use:
void instance 001
9 ./Oct/s_enn.oct -a 1 -t 0 0 0.0064 -i 200
0
0
Also, still a bit in the dark concering your meta suggestion.
Meta isn't a primitive in Radiance, or is it?
Actually I know it only as a graphics format in Radiance?
Hopefully I'm seriously overlooking something here?
Cheers,
-jelle
I guess this is the trouble,
-a 1 -t..... -i 200
a missing 1
ciao
G.
_______________________________________________
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
Dear Greg, Jack, Giulio,
Many thanks for all your responses!
Thanks, Greg for pointing out the error I made.
Thank you Jack for showing the right implementation!
This list should be added to Radiance' feature list!
Cheers,
jelle
Actually I've been using the 'mesh' primitive for not a long time, since
Radiance on win32 was quite crippled until Francesco's Mingw compiled
version. Since rview wasn't part of this version of Radiance its cool to see
that NRV supports it nicely.
-jelle
···
----
PS alternatively when the mesh is really big I use a mesh
primitive...
but I guess you know already about.
----
Just if someone is not aware of it yet, there is a nice tool called replmarks in radiance distributions, that replaces triangles by meshes, instances or xform'ed geometry.
Very useful for controlling instances from CAD. The only limitation is that you cannot specify mirror transformations using it. And I do not know a converter that writes out instances/references/blocks/symbols in CAD files as their corresponding triangle markers, so you have to care about it yourself. In formZ, use the symbol detail level (e.g. detail level containing geometry, level 2 triangle) and you get the whole thing nicely integrated into your CAD/modeler.
CU Lars.
