[Radiance-general] pkgBSDF weird behavior

[Jack de Valpine]

Hi German,

When you run makeall install, after you are asked for the location for 
the location of the executables, you will get something like the following:
> Current rmake command is:
> #!/bin/sh
> exec make "SPECIAL=" \
>     "OPT=-O2" \
>     "MACH=-Dlinux -D_FILE_OFFSET_BITS=64 -L/usr/X11R6/lib 
> -I/usr/include/X11 -DNOSTEREO" \
>     ARCH=IBMPC "COMPAT=erf.o" \
>     INSTDIR=/Rlib/bin \
>     LIBDIR=/Rlib/lib \
>     ESUFFIX= \
>     CC=gcc "$@" -f Rmakefile
> Do you want to change it?
If you type "Y," this will be opened up in an editor (vi is default 
unless you specify otherwise when you first run makeall). The "OPT=" 
line is the one that you should edit according to Greg's suggestions. So 
I think that you could first try:

    "OPT=-01"

    and then if that does not help, you could try:

    "OPT="

    This will turn of optimizations.

I think that -g gives debugging output. I think that you could do the 
following to compile with this:

    change

    CC=gcc "$@" -f Rmakefile

    to

    CC=gcc -g "$@" -f Rmakefile

Hope this helps,

Best,

-Jack

-- 
Jack de Valpine
President

Visarc Incorporated
www.visarc.com

channeling technology for superior design and construction

On 9/13/2013 8:30 AM, Germán Molina Larrain wrote:
> mmm, that is weird.
>
> I actually did not understand anything you said, haha... I just use 
> the "sudo ./makeall install" command. How do I choose different -O and 
> -g options?
>
> THANKS AGAIN
>
> German
>
>
> 2013/9/12 Greg Ward <gregoryjward at gmail.com 
> <mailto:gregoryjward at gmail.com>>
>
>     Hi Germán,
>
>     I tried it, and I don't get an error.  What -O setting are you
>     using when compiling Radiance?  Anything above -O2 has been known
>     to generate incorrect code with gcc.  Even -O2 sometimes causes
>     runtime errors.
>
>     If you recompile and get an error using -O1, try without -O, just
>     using -g, and run it in the debugger.  This is the only way to
>     find out where it is going wrong.
>
>     Best,
>     -Greg
>
>>     *From: *Germán Molina Larrain <gmolina1 at uc.cl
>>     <mailto:gmolina1 at uc.cl>>
>>
>>     *Date: *September 12, 2013 3:04:29 PM PDT
>>
>>     *
>>     *
>>
>>     Thanks Greg,
>>
>>     I uploaded a file called "SunliteExport.zip" to the same folder.
>>
>>     Bye!
>>
>>     German
>>
>>
>>     2013/9/12 Greg Ward <gregoryjward at gmail.com
>>     <mailto:gregoryjward at gmail.com>>
>>
>>         Hi Germán,
>>
>>         This is an unusual error, and not very informative.  I will
>>         attempt to reproduce it if you can send me all of your files
>>         off-list in a compressed tar file attachment (if it's not too
>>         big to mail).
>>
>>         Best,
>>         -Greg
>>
>>>         *From: *Germán Molina Larrain <gmolina1 at uc.cl
>>>         <mailto:gmolina1 at uc.cl>>
>>>
>>>         *Date: *September 12, 2013 1:03:47 PM PDT
>>>
>>>         *
>>>         *
>>>
>>>         Dear Everyone,
>>>
>>>         I have been playing with pkgBSDF and that kind of stuff, and
>>>         something weird is happening (I think).
>>>
>>>         This is my scene (of course the references are present in
>>>         the respective directories).
>>>
>>>         ##### CASE
>>>
>>>         !xform Geometry/Layer0.rad
>>>
>>>         #East Shading
>>>         !pkgBSDF -s system.xml | xform -t 0 0 .02 -rx 90 -rz 90 -t
>>>         5.02 1.02 2.27 -a 5 -t 0 1.5 0
>>>
>>>         #South Shading
>>>         !pkgBSDF -s blind0.xml | xform -rx -90 -t 2.54 0 1.869
>>>
>>>         !xform Skies/sky.rad
>>>
>>>         Anyway, when I try to render, it only works without ambient
>>>         calculations (-ab 0), and only with one pkgBSDF declaration
>>>         (the BSDFs are in my RAYPATH).
>>>
>>>         When I try to render with -ab 1, it returns:
>>>
>>>          4018 Abort trap: 6  rpict -vf Views/view.vf -ab 1 octree >
>>>         Images/south_only_AB1.hdr
>>>
>>>         When I try both together with -ab 0:
>>>
>>>         043 Abort trap: 6 rpict -vf Views/view.vf -ab 0 octree >
>>>         Images/both_AB0.hdr
>>>
>>>         And, when I try both together with -ab 1:
>>>
>>>         4075 Abort trap: 6  rpict -vf Views/view.vf -ab 1 octree >
>>>         Images/both_AB1.hdr
>>>
>>>         Here you can check the images "just generated".
>>>         https://www.dropbox.com/sh/fqedoyd7jdbn0uv/rNniz9rqoI
>>>
>>>         Might this be a bug? I am using the HEAD version from about
>>>         yesterday...
>>>
>>>         THANKS
>>>
>>>         Germán
>>
>>
>
>     _______________________________________________
>     Radiance-general mailing list
>     Radiance-general at radiance-online.org
>     <mailto:Radiance-general at radiance-online.org>
>     http://www.radiance-online.org/mailman/listinfo/radiance-general
>
>
>
>
> _______________________________________________
> Radiance-general mailing list
> Radiance-general at radiance-online.org
> http://www.radiance-online.org/mailman/listinfo/radiance-general

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.radiance-online.org/pipermail/radiance-general/attachments/20130913/cdee5bce/attachment-0001.html>