Ecotect Analysis Forum (Read Only)
cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 

Add custom render parameters to Radiance export

14 REPLIES 14
SOLVED
Reply
Message 1 of 15
m.soderlund
3278 Views, 14 Replies

Add custom render parameters to Radiance export

Good afternooon,

 

I'm having trouble adding my own custom parameters whilst exporting daylight factor calculations to Radiance. Radiance/Ecotect ignores the custom parameters you've specified in the "Radiance analysis" wizard.

 

According to this post (http://forums.autodesk.com/t5/Autodesk-Ecotect-Analysis/Ecotect-does-not-keep-my-Radiance-parameters... ) the workaround would be to go through the Radiance Control Panel and save the "render="-parameters to the RIF file manually, which I've done. The problem tho is that when I now try to make a new export thro the wizard, it just overwrites my modified RIF file with the standard file and the radiance command prompt just shows the standard parameters. 

 

Is there any way to export to Radiance without writing over the RIF file?

 

I know I can use Radiance Control Panel and make a calculation from there but, as far as I know, you can only render a image from RCP and not import any grid/point values back to Ecotect, or have I missed something?

 

I'm running Ecotect Analysis 2011 with Desktop Radiance v.2.00 Beta or MinGW Radiance 2008-08-14 (tried both versions).

 

Please get back to me, it would be appreciated.

 

Regards

Mikael

 

14 REPLIES 14
Message 2 of 15
m.soderlund
in reply to: m.soderlund

I've done some more investigating and found out that it seems like the actual parameters that Radiance are using is inside the BAT file and not the RIF file. The BAT file includes the rtrace command with all parameters, and it looks like the standard -ab is set to 8 bounces. Is that correct? Doesn't radiance take the parameters set in the RIF file into consideration? 

 

I do get different results if I, for example, set the "Indirect Reflections" to different values in the Radiance Analysis wizard so the RIF file must be of some importance but it's quite unclear for me how right now.

 

Please share me your thoughts on the subject

 

Regards

Mikael

Message 3 of 15
m.soderlund
in reply to: m.soderlund

Nevermind, I figured it out eventually.

 

Solution:

Change the .RIF file to your liking (the render= parameter) and then manually run the .BAT file Ecotect created for the project. When done calculating import the results in the newly created .OK file to Ecotect thru Grid Management -> Manage Grid Data -> Import Data. 

 

Voilá

 

 

Message 4 of 15
javeds
in reply to: m.soderlund

I am trying to import data that I have generated using the rediance control panel into an ecotect analysis grid without success.  You seem to have solved the problem but I could not replicate your solution.  I would appreciate if you can explain your steps a bit further.  Thanks in advance!

Message 5 of 15

my question seems related to the other questions:

 

What is the diference between the rtrace settings in the batch-file and the settings in the rif-file?

which should be altered for diferent settings in rtrace for grid calculations?

 

I would asume that the rif-file would only effect the radiance calculation and therefor the oct-file that is made and then used by rtrace, i would therefor assume the amount of reflections (-ab) set in the rif-file should have no effect on results in rtrace. However the amount set for -ab in the rif-file seems to have more effect then the amount for -ab in the batch-file on grid-calculation results.

Message 6 of 15
Pennetier1
in reply to: mercyfull_fate

Hello mercyfull_fate, 

 

The rtrace settings in the batch file are originally created based on the settings set within the Ecotect wizard; this batch file is created from the settings of the RIF file.  Now, you can edit the settings of the RIF file with the RCP, save back as the same .rif file and re-run the file.  You will see that the DOS windows will first list the original settings - as per the first run based on the batch file - but then your modified .rif file will overwrite these original values.  You can see the new values in the DOS window if you keep reading the settings. So the settings in the batch file are the original settings from the first run, and the .rif settings are what will be used, as they are overwriting the original batch parameters.

 

The parameters of the batch file are derived from the original .rif file, so there should not be any differences in results.  Are you saying that when you edit your .rif file to have the same -ab value as your original batch file, the results come out different?

 

Cheers,

Olivier A. PENNETIER

SYMPHYSIS

www.symphysis.net

Message 7 of 15
mercyfull_fate
in reply to: Pennetier1

Hello Olivier A. Pennetier,

The general settings at the end of the export wizard go into the rif file. When calculating a grid there is an advanced parameters page in the wizard before the last page of the export wizard. The changes in those are put in the batchfile as settings for rtrace.
Both parameters, the rif file parameters and the rtrace parameters, are changed indipendently from each other.
I do notice the rif file settings being used as parameters in the first calculation which is shown in the dos window during calculations.

And the parameters in the batchfile for rtrace are being used as they do have effect but are not shown in the doswindow that apears during calculations.

Changes in both/either of the settings -ab value have effect on the rtrace calculation for the grid points. While i would asume that in the grid calculation the internal reflection (-ab) of only the rtrace settings should have effect. Or am i wrong in that asumption?

 

To ilustrate what i mean i added a document with the contents of the batch-file and the dos-window with in yellow the rtrace calculation/parameters and in green the radiance calculation for the generation of, among other things, the oct.file.

 

Kind regards,

Wouter

Message 8 of 15
Pennetier1
in reply to: mercyfull_fate

Hello Wouter, 

 

I think I understand better your question.  It appears that the .bat file and settings set in the step 10 of 11 of the wizard are applied when using point/object lighting analysis, such as on the analysis grid; the setting set in the last summary page for the RIF file will be ignored; i.e. for a specific -ab setting in step 10 of 11, the indirect reflection setting for the RIF file in step 11 will not affect the grid/point results.

However, if what you are trying to do is create a Radiance rendering, then step 10 of 11 is by-passed and the setting for the RIF file in the last step of the wizard will be used.

This makes some sense as it is difficult to run a point/object analysis from a RIF file in the RCP and re-import the result in the Ecotect model; however, running RIF file in the RCP is straighforward when doing Radiance renderings.

 

Let me know if that makes sense; if so, please accept as a solution so that others can benefit from this information.


Cheers,

Olivier A. PENNETIER

SYMPHYSIS

www.symphysis.net

Message 9 of 15
mercyfull_fate
in reply to: Pennetier1

Hello Olivier,

 

I don't think you completely understand me however were getting there, if i understand what you mean then my problem is that that is not actualy happening in the calculation.

 

I did a test with the settings where i only adjusted the -ab settings in step 8/9 and the indirect reflections in step 9/9 and then i checked the daylightfactor results in a grid and put it in a excelsheet. The blank spaces indicate there are no changes compared to the calculation above. I have detail, variability and quality on medium:

 

Naamloos.png

 

So the settings in step 8/9 are the same as the settings for rtace in the .bat file, the settings from step 9/9 are the same as the settings in the .rif-file and in the .rif file render= settings supersede the basic settings. I understand all that.

 

However my problem is that both the -ab setting from step 8 and 9 have an (simultanious) effect on the calculationresults of the grid.

 

I'm aware the diferences in results for this case are negligible however i am trying to validate the programs for my thesis so i do need to know where the diferences come from.

 

Regards,

Wouter

Message 10 of 15
Pennetier1
in reply to: mercyfull_fate

Hi Wouter, 

 

I agree there should not be any differences, i.e., the results should be the same when the -ab of 8/9 is set to a particular value and the indirect reflection is whatever value; I could not reproduce your results.  I have ran quite a few tests on this as well, and I always get the same results for a particular -ab as set in 8/9 (.bat), regardless of the indirect bouncing in 9/9 (.rif).  I have tried this for illuminance calculations as well as Daylight Factor, on both a grid and a single point object.

 

I am curious to know what is causing the shift on your side. Can you print screen and show me your final settings (9/9) so I can try to reproduce.  Maybe another setting such as interior/exterior view is causing the shift?

 

Cheers,

Olivier A. PENNETIER

SYMPHYSIS

www.symphysis.net

Message 11 of 15
mercyfull_fate
in reply to: Pennetier1

Hey Olivier,

 

I added a pdf with my settings for step 8/9 and step 9/9.

in step9/9 sometimes i have current model view on and sometimes of but that doesn't seem to make a diference.

 

Maybe it is a problem in the old version i am using (2011)?

 

Regards,

Wouter

Message 12 of 15
Pennetier1
in reply to: mercyfull_fate

Hi Wouter, 

 

I do not think it has to do with your version of the software as you have the latest one (2011).

 

I have ran again some tests and I still get very similar values, i.e. Radiance prioritizes the settings of the batch file rather than the RIF settings.

Only a couple of runs shows slightly different results when the Indirect bounce of the RIF was set to 0.

 

Batch vs RIF.jpg

 

My settings are very similar to yours:

 

RIF settings.jpg

 

I do not know why you are getting different results.  A thought is that because Radiance uses pseudo-random ray tracing, it cannot always be perfectly similar, but my results would disprove that theory as the results are quite consistent.  I have used a single point object instead of a grid.  Have you tried single point object?

 

Cheers,

Olivier A. PENNETIER

SYMPHYSIS

www.symphysis.net

Message 13 of 15
mercyfull_fate
in reply to: Pennetier1

Hey Olivier,

 

You might be on to something, first i ran 4 test with a (multiple points) object, it had similar problems to a grid, but when i tried a singlepoint object it did apear to prioritize the advanced parameters after my first adjustment to the advanced parameters.

I put the tests in the folowing table, the multiple point object and singlepoint object are seperate tests, the first setting was the default setting.

test 2 ecotectexport.png

After this test i added another point and calculated both points, again advanced parameters(step8/9) were corectly prioritized, because i continued with previous test step 8/9 was prioritized immediately.

I got the following results, point 1 changed slightly compared with the first test (also in this table) but that is to be expected because the model changed by adding a sensor point:

test 3 ecotectexport.png

 

to sumarize:

It does something with both settings in case of a grid or (multiple point) object.

When calculating single point objects it apears to prioritize step 9/9 untill i adjust the advanced parameters(step 8/9) after that the program prioritizes step 8/9.

So single point calculations work like they are suposed to.

 

So a posible solution of my problem would be to only use single point sensor and no grid however the grid is actualy very usefull for visualization of the results.

 

Regards,

Wouter

Message 14 of 15

hey Olivier,

 

I have found a solution to my problem, i copy the setting used for rendering the views (in RadianceCP i click on advanced render settings, then click on load current default values) to the advanced rendering parameters (step 8/9).

When using the same settings in both places the grid result are simular to the singlepoint object results so i asume the calculation goes correct with those settings.

 

This solution is not ideal because of the need to fill in all the settings twice but it does solve the problem.

 

Thanks for the help,

Wouter

 

ps:

I'm not sure how to mark my question as solved or if im able to since i did not post the original topic.

Message 15 of 15
Pennetier1
in reply to: mercyfull_fate

Great Wouter, 

 

Indeed, the RIF file was using the other variables from the RIF file, not from the advanced settings from 8/9.

It is also easy to copy the values in the advanced parameters of the RIF file, and "paste all" in the wizard.

 

Glad you got it figured out.

 

Cheers,

Olivier A. PENNETIER

SYMPHYSIS

www.symphysis.net

Can't find what you're looking for? Ask the community or share your knowledge.

Post to forums  

Autodesk Design & Make Report