Hi:
Does anybody knows why the different number of "INDIRECT" in .rif file could
influence the result of illuminance in point calculation?
I found that if i changed the "indirect" in .rif file, the illuminances in
the defined grid are different even though the "ab" is the same in rtrace
command. is anybody knows the reason?
Best Regards!
Victor.
*.rif files are not used for rtrace comands. Whatever differences
you find they are not caused by settings in the *.rif file.
It has been a while since I looked at the rad code but I'm pretty sure
that INDIRECT only affects the "-ab" value of the rpict comand.
Depending on your QUALITY setting is either INDIRECT = -ab
or INDIRECT=-ab-1. To keep the -ab value constant you have to
override it in with the "render=" option.
If you get different values for rtrace calculations you either just
see the typical error of the stochastic approach of Radiance or
your other ambient options are not high enough and your result
have a large variation.
How do you use the rif file with rtrace? Can you send details of
the rtrace commands and some info on the scene?
Regards,
Thomas
On Tue, Jan 26, 2010 at 7:40 AM, victor li <[email protected]> wrote:
Hi:
Does anybody knows why the different number of "INDIRECT" in .rif file could
influence the result of illuminance in point calculation?I found that if i changed the "indirect" in .rif file, the illuminances in
the defined grid are different even though the "ab" is the same in rtrace
command. is anybody knows the reason?Best Regards!
_______________________________________________
Radiance-general mailing list
[email protected]
http://www.radiance-online.org/mailman/listinfo/radiance-general
Thank you for your reply!
It has been a while since I looked at the rad code but I'm pretty sure
that INDIRECT only affects the "-ab" value of the rpict comand.
Depending on your QUALITY setting is either INDIRECT = -ab
or INDIRECT=-ab-1. To keep the -ab value constant you have to
override it in with the "render=" option.
I am confused by the " INDIRECT = -abor INDIRECT=-ab-1. To keep the -ab
value constant you have to override it in with the "render=" option." In my
setting, "INDIRECT would be a number, like INDIRECT = 5.
For my setting in rtrace, other ambient options are high enough to avoid the
mistake. And the INDIRECT is est in .rif file and rtrace is set in another
file. My setting is as following:
rtrace -I -h -dp 2046 -ar 32 -ms 0.063 -ds .2 -dt .05 -dc .75 -dr 3 -sj 1
-st .01 -ab 8 -aa .1 -ad 2048 -as 1024 -av 0.01 0.01 0.01 -lr 12 -lw .0005
-af 5.amb 5.oct < 5.pts > 5.dat
RESOLUTION= 640 480
DETAIL= MEDIUM
VARIABILITY= MEDIUM
QUALITY= MEDIUM
INDIRECT= 5
REPORT= 2
render= -i
Hi!
rtrace -I -h -dp 2046 -ar 32 -ms 0.063 -ds .2 -dt .05 -dc .75 -dr 3 -sj 1 -st .01 -ab 8 -aa .1 -ad 2048 -as 1024 -av 0.01 0.01 0.01 -lr 12 -lw .0005 -af 5.amb 5.oct < 5.pts > 5.dat
RESOLUTION= 640 480
DETAIL= MEDIUM
VARIABILITY= MEDIUM
QUALITY= MEDIUM
INDIRECT= 5
REPORT= 2render= -i
As Thomas wrote, rtrace does not care about this. You can use a RIF-file when you pass its name to rad as a command line argument, e.g.
bash#rad myRenderings.rif
Now rad would read in the setting, and call the commands necessary to do all calculations as defined in the RIF-file.
If you call rtrace, the rtrace-options apply. rtrace does not know anything about the existance of you RIF-file. The file name of the RIF-file does not appear on the command line you posted - how should rtrace know to use it?
How did you arrive at the ambient parameters you are using? Especially ar 32 combined with very high ad amd as? Did you consider scene size and level of detail / irradiance variance?
Cheers Lars.
Thank you for your reply!
I am confused by the " INDIRECT = -abor INDIRECT=-ab-1. To keep the -ab
value constant you have to override it in with the "render=" option." In my
setting, "INDIRECT would be a number, like INDIRECT = 5.
Sorry, I wanted to say that the value of "-ab" is calculated from
the value for INDIRECT depending on your QUALITY setting. If
you have QUALITY=low then "-ab" will be your INDIRECT value,
if you have QUALITY=medium or high then "-ab" will be the
INDIRECT value + 1. It was written in a rush.
If you want to force a particular value for an rpict (or rvu) option
you can add it to the "render=" setting (like the "-i" in your case).
Rad will add these options at the end of the generated rpict
command and so override all options that have been calculated
from the QUALITY etc. settings.
For my setting in rtrace, other ambient options are high enough to avoid the
mistake. [...] My setting is as following:rtrace -I -h -dp 2046 -ar 32 -ms 0.063 -ds .2 -dt .05 -dc .75 -dr 3 -sj 1
-st .01 -ab 8 -aa .1 -ad 2048 -as 1024 -av 0.01 0.01 0.01 -lr 12 -lw .0005
-af 5.amb 5.oct < 5.pts > 5.dat
I wouldn't call "-aa 0.1" and "-ar 32" particularly high. Can you
produce similar results for two calculations if you don't change
anything?
As Lars already wrote, the accuracy of rtrace is definded only
by the parameters above. The settings below have nothing to
do with it.
RESOLUTION= 640 480
DETAIL= MEDIUM
VARIABILITY= MEDIUM
QUALITY= MEDIUM
INDIRECT= 5
REPORT= 2render= -i
Regards,
Thomas
On Wed, Jan 27, 2010 at 6:34 AM, victor li <[email protected]> wrote:
Hi Lars:
Actually, i know what you mean, but the result is strange, it is not what we
expect. I donot know what is wrong in my setting. I used rad and rtrace
command as follwoing:
RAD 5.rif
rtrace -I -h -dp 2046 -ar 32 -ms 0.063 -ds .2 -dt .05 -dc .75 -dr 3 -sj 1
-st .01 -ab 8 -aa .1 -ad 512 -as 256 -av 0.01 0.01 0.01 -lr 12 -lw .0005 -af
5.amb 5.oct < 5.pts > 5.dat
In the .rif file:
RESOLUTION= 640 480
DETAIL= MEDIUM
VARIABILITY= MEDIUM
QUALITY= MEDIUM
INDIRECT= 5
REPORT= 2
render= -i
So the result of rtrace command would donot matter with the seting in .rif.
However, if i changed the "INDIRECT" and keep any other settings the same,
the result of rtrace command is different. Do you know the reason?
Hi Victor!
Actually, i know what you mean, but the result is strange, it is not what we expect. I donot know what is wrong in my setting. I used rad and rtrace command as follwoing:
Hm, I still do not get why these two programs could be related.
RAD 5.rif
I guess this is a typo, and what you actually type in is in small letters (rad 5.rif?). What is the result of this, I think there is no view given in your RIF. Is there any result from the run of rad? I would expect nothing to happen, as rad has nothing to do?
rtrace -I -h -dp 2046 -ar 32 -ms 0.063 -ds .2 -dt .05 -dc .75 -dr 3 -sj 1 -st .01 -ab 8 -aa .1 -ad 512 -as 256 -av 0.01 0.01 0.01 -lr 12 -lw .0005 -af 5.amb 5.oct < 5.pts > 5.dat
How is this related to your rad-call? Why do you have this low ar-setting? Where are all these values derived from?
In the .rif file: RESOLUTION= 640 480
DETAIL= MEDIUM
VARIABILITY= MEDIUM
QUALITY= MEDIUM
INDIRECT= 5
REPORT= 2render= -i
So the result of rtrace command would donot matter with the seting in .rif. However, if i changed the "INDIRECT" and keep any other settings the same, the result of rtrace command is different. Do you know the reason?
Delete your RIF-file. Run the rtrace-command with the same parameters (maybe increase the ar first) several times, and compare the output in 5.dat.
Cheers, Lars.
If i delete the rif file and change the command " rad 5.rif" into "oconv
5.rad 5_sky.rad > 5.oct". And run again, the results looks the same.
2010/1/29 Lars O. Grobe <[email protected]>
Hi Victor!
Actually, i know what you mean, but the result is strange, it is not what
we expect. I donot know what is wrong in my setting. I used rad and rtrace
command as follwoing:Hm, I still do not get why these two programs could be related.
RAD 5.rif
I guess this is a typo, and what you actually type in is in small letters
(rad 5.rif?). What is the result of this, I think there is no view given in
your RIF. Is there any result from the run of rad? I would expect nothing to
happen, as rad has nothing to do?rtrace -I -h -dp 2046 -ar 32 -ms 0.063 -ds .2 -dt .05 -dc .75 -dr 3 -sj 1
-st .01 -ab 8 -aa .1 -ad 512 -as 256 -av 0.01 0.01 0.01 -lr 12 -lw .0005 -af
5.amb 5.oct < 5.pts > 5.datHow is this related to your rad-call? Why do you have this low ar-setting?
Where are all these values derived from?In the .rif file: RESOLUTION= 640 480
DETAIL= MEDIUM
VARIABILITY= MEDIUM
QUALITY= MEDIUM
INDIRECT= 5
REPORT= 2render= -i
So the result of rtrace command would donot matter with the seting in
.rif. However, if i changed the "INDIRECT" and keep any other settings the
same, the result of rtrace command is different. Do you know the reason?Delete your RIF-file. Run the rtrace-command with the same parameters
(maybe increase the ar first) several times, and compare the output in
5.dat.Cheers, Lars.
_______________________________________________
Radiance-general mailing list
[email protected]
http://www.radiance-online.org/mailman/listinfo/radiance-general
victor li wrote:
If i delete the rif file and change the command " rad 5.rif" into "oconv 5.rad 5_sky.rad > 5.oct". And run again, the results looks the same.
Hi Victor,
sorry it is hard to help out. Can you please try to give clear information that allows us to reproduce what happened? If I take your words above literally, the "results" should be just one file in the current directory called "5.oct". This will always look the same as long as you do not change the file name and scene description. So, if you experience different results for different command sequences, exactly list the commands in the order you executed them and specify what you got as result. Usually, result will be a error message, one or more values or in some cases image files, that you could upload somewhere and refer to by giving a link in your email. So far we have no reproduceable problem description and cannot see what you really did or got back.
Cheers, Lars.
I am sorry about my explianation!
If i delete the rif file and change the command " rad 5.rif" into "oconv
5.rad 5_sky.rad > 5.oct". And run again, the result looks right and
reasonable.
If i donot delete rif file. And simulate with differrent "INDIRECT" number
in rif file and keep other settings the same. The results of illumiance in
the grid are differreent with the changing of "INDIRECT"
I thought the results of illumiance would not change with the different
"INDIRECT" However, it is not. I am confulsed by this. Do you know why?
2010/2/4 Lars O. Grobe <[email protected]>
victor li wrote:
If i delete the rif file and change the command " rad 5.rif" into "oconv
5.rad 5_sky.rad > 5.oct". And run again, the results looks the same.Hi Victor,
sorry it is hard to help out. Can you please try to give clear information
that allows us to reproduce what happened? If I take your words above
literally, the "results" should be just one file in the current directory
called "5.oct". This will always look the same as long as you do not change
the file name and scene description. So, if you experience different results
for different command sequences, exactly list the commands in the order you
executed them and specify what you got as result. Usually, result will be a
error message, one or more values or in some cases image files, that you
could upload somewhere and refer to by giving a link in your email. So far
we have no reproduceable problem description and cannot see what you really
did or got back.Cheers, Lars.
_______________________________________________
Radiance-general mailing list
[email protected]
http://www.radiance-online.org/mailman/listinfo/radiance-general