Ximage illuminance readings

radiance-general
#1

I know this is supposed to be fairly basic but needs clarification: I want to get lux readings from my pic files
When generating *.pic files (with -i option for irradiance) , viewing them with ximage and
selecting a point with the ximage 'l' command gives me values that are clearly NOT lux values (values appear to be 1/10th of my expected lux measurement).
Correct me if I'm wrong, but isn't the value in W/m2 at an efficacy of 179lm/watt? For daylight only or all irradiances? If so, what conversion to get lux values?

regards
Steve

···

_________________________________________________________________
Create a handy button so your friends can add U to their buddy list. Try it now!
http://g.msn.ca/ca55/211

#2

Dear Steve,

The 'l' command in ximage applies the 179 lumens/watt conversion factor, so if the input is a picture created by rpict with the -i option, you should be seeing lux values on your opaque surfaces. If you are off by a factor of ten, then I would suspect the input values or the calculation are at fault.

Best,
-Greg

···

From: steve michel <smichel_designer@hotmail.com>
Date: March 10, 2008 10:48:33 AM PDT

I know this is supposed to be fairly basic but needs clarification: I want to get lux readings from my pic files
When generating *.pic files (with -i option for irradiance) , viewing them with ximage and
selecting a point with the ximage 'l' command gives me values that are clearly NOT lux values (values appear to be 1/10th of my expected lux measurement).
Correct me if I'm wrong, but isn't the value in W/m2 at an efficacy of 179lm/watt? For daylight only or all irradiances? If so, what conversion to get lux values?

regards
Steve

#3

Greg,

Thanks for confirming the conversion factor..that was a the gap in my knowledge of radiance. Now, without getting into into sordid details, the scene pic in question has a skylight and some T5 direct/indirect lighting. From the rendered night and day pics all seems well with ximage's human exposure setting corroborating with what I would expect. The daytime pic shows strong illumination on the work plane but yields a 'lux' value lower than I expected (250 to 300). I did not use a secondary lightsource for the skylight. With a night, with only the T5 luminaires (ies2rad in meters), which, from manufacturer's data and experience, should give a 300 to 500lux illuminance, gives me a reading approx. 40lux(?).

I tried rtrace (courtesy of man pages) with ximage -t command on the daytime pic
$ ximage office-i.pic | rtrace -h -x 1 -i office2xx.oct | rcalc -e '$1=47.4*$1+120*$2+11.6*$3'
18.3891712
21.1014687
20.6663913
???

I like ximage's interactive features, but something may be missing in that pipe to rcalc.

Im using radiance to better (re)present my interior designs but
feel out place here using scientific instruments for my art supplies

thnks always
Steve

···

----------------------------------------

From: gward@lmi.net
Subject: Re: [Radiance-general] Ximage illuminance readings
Date: Mon, 10 Mar 2008 14:04:06 -0700
To: radiance-general@radiance-online.org

Dear Steve,

The 'l' command in ximage applies the 179 lumens/watt conversion
factor, so if the input is a picture created by rpict with the -i
option, you should be seeing lux values on your opaque surfaces. If
you are off by a factor of ten, then I would suspect the input values
or the calculation are at fault.

Best,
-Greg

From: steve michel
Date: March 10, 2008 10:48:33 AM PDT

I know this is supposed to be fairly basic but needs clarification:
I want to get lux readings from my pic files
When generating *.pic files (with -i option for irradiance) ,
viewing them with ximage and
selecting a point with the ximage 'l' command gives me values that
are clearly NOT lux values (values appear to be 1/10th of my
expected lux measurement).
Correct me if I'm wrong, but isn't the value in W/m2 at an efficacy
of 179lm/watt? For daylight only or all irradiances? If so, what
conversion to get lux values?

regards
Steve

_______________________________________________
Radiance-general mailing list
Radiance-general@radiance-online.org
http://www.radiance-online.org/mailman/listinfo/radiance-general

_________________________________________________________________
Like solving puzzles? Then you'll love Flexicon! Play now!
http://g.msn.ca/ca55/213

#4

Hi Steve,

A direct/indirect fixture needs an interreflection calculation for any level of accuracy. I recommend setting -ab 1 on the rtrace command line, maybe even -ab 2. This will take considerably longer to produce a result, but should be closer to what you expect if there is nothing amiss with your model. The command you are using otherwise looks OK.

I'm a little surprised by the factor of 10, unless the fixture throws over 90% of its light towards the ceiling.

-Greg

···

From: steve michel <smichel_designer@hotmail.com>
Date: March 10, 2008 6:56:26 PM PDT

Greg,

Thanks for confirming the conversion factor..that was a the gap in my knowledge of radiance. Now, without getting into into sordid details, the scene pic in question has a skylight and some T5 direct/indirect lighting. From the rendered night and day pics all seems well with ximage's human exposure setting corroborating with what I would expect. The daytime pic shows strong illumination on the work plane but yields a 'lux' value lower than I expected (250 to 300). I did not use a secondary lightsource for the skylight. With a night, with only the T5 luminaires (ies2rad in meters), which, from manufacturer's data and experience, should give a 300 to 500lux illuminance, gives me a reading approx. 40lux(?).

I tried rtrace (courtesy of man pages) with ximage -t command on the daytime pic
$ ximage office-i.pic | rtrace -h -x 1 -i office2xx.oct | rcalc -e '$1=47.4*$1+120*$2+11.6*$3'
18.3891712
21.1014687
20.6663913
???

I like ximage's interactive features, but something may be missing in that pipe to rcalc.

Im using radiance to better (re)present my interior designs but
feel out place here using scientific instruments for my art supplies

thnks always
Steve

#5

I have a similar question. I am running the example in the “Radiance_tutorial”, and the dimensions of the room are 4m5m3m. Then, I placed two bulbs at a height of 2.6m, and the total lumens of each bulb is 1,500 lumens.

In the tutorial, these two bulbs will give approx. 500lux illuminance at a height of 0.85m, but my simulation just gives me approx. 100 lux?

This is the commands I used to calculate the Illuminance values.
"cat data/line.pts | rtrace -I -ab 3 -h -oov scene.oct | rcalc -e ’$1=179*(.265*$1+.670*$2+.065*$3)’ "

I don’t know what is missing in my calculation

Thanks
Fan

Rtrace illuminance reading in Radiance_tutorial
#6

Your “rtrace -oov” option should just be “rtrace -ov” – the extra ‘o’ outputs the origin, so that is what is being used in the subsequent rcalc command rather than irradiance.

#7

My bad.The command I used is actually this one.

"cat data/line.pts | rtrace -I -ab 3 -h -oov scene.oct | rcalc -e ’$1=179*(.265*$4+.670*$5+.065*$6)’ "

#8

OK, then I don’t have enough information to help. Which tutorial are you using? Can you share your files? What is in “data/line.pts?” Usual errors are having the points in the wrong place, or the normals pointing the wrong direction, or some other problem with the geometry. No one can debug it without knowing where to look for issues. It’s even possible that the tutorial values are incorrect.

#9

Here are the files I used. That is basically the examples from Radiance’s installation directory. And I also put the tutorial in this folder. This tutorial is the official Radiance tutorial by Axel Jacobs.

https://drive.google.com/drive/folders/1mqbjLVl5yhXlTt0EF57xIY7zq4R8bXdq?usp=sharing

#10

OK, but you still haven’t shared your “data/line.pts” file or explained how you computed the "light’ values of your “objects/bulb.rad” file, which had an undefined modifier error in it. Axel’s tutorial is excellent, but I don’t know what part of it you have taken or if the files are from Axel or ones you created yourself.

Please bear in mind that in order for anyone to help you, you need to have a reproducible error that is clearly unexpected.

#11

Sorry for the late reply. I just uploaded my “data/line.pts” file and corrected that modifier error. For the “light” values, I am using “lampcolor” program to calculate them, and the settings for my bulbs are listed below.

    lamp type: white
    length unit: meter
    lamp geometry: sphere
    Radius: 0.03
    lamp lumens: 1,500.

Here is the new link: SharedDrive - Google Drive
Besides, my calculate is taken from Part 5.2.2 from Axel’s tutorial. His results are shown below
image

Hope this time you are able to reproduce my results.

Sorry for the late reply again, and thanks always.

Fan

#12

Hi Fan,

I am not sure what is going wrong with your rcalc run. Your model and input values all look OK. However, you get very different numbers than I get when I run it. Here is my output with the latest version of Radiance:

rtrace -I -ab 3 -h -oov scene.oct < data/line.pts | rcalc -e '$1=$2;$2=179*(.265*$4+.670*$5+.065*$6)'
0.5	79.8487986
1	101.613837
1.5	115.581261
2	127.947209
2.5	129.046369
3	129.947958
3.5	128.105882
4	116.308692
4.5	98.2985327

This seems quite reasonable. What do you see when you run the following:

rtrace -version
rtrace -defaults

The current version is 5.4a. It could be that you have such an old version of Radiance that the default settings are not serving you well in this calculation. You can try copying the -a* settings from the current defaults. I am copying the relevant settings below:

-aa 0.100000			# ambient accuracy
-ar 256      			# ambient resolution
-ad 1024     			# ambient divisions
-as 512      			# ambient super-samples
-lr -10      			# limit reflection (Russian roulette)
-lw 2.00e-03			# limit weight
-dt 0.030000			# direct threshold
-dc 0.750000			# direct certainty

The remaining settings should not have much influence. If applying the above parameters does not solve your problem, then I have some concern that your copy of Radiance is somehow producing inaccurate output. In any case, you may wish to install the latest version if you have not already from our github server.

Cheers,
-Greg

#13

Hi, Greg,
Actually, my output is the same as yours. The following results are directly taken from Axel’s tutorial, and it is different. That is why I thought my output is weird.

I thought the simulation model in Axel’s tutorials is exactly the same as the example files. While it seems not. So maybe the model setting in his tutorials is different.

Anyway, I really appreciate your help.

image

#14

Sorry – I misunderstood the results you quoted as being yours, not the tutorial’s. I don’t know what to say about those results. Even going back to the settings from version 3.7 (which Axel seems to have used to create the tutorial), my values are not as large nor as random. In fact, the settings have not really changed so much.

At this point, I would just assume you created the tutorial files correctly and the simulation is working as it should be.

Cheers,
-Greg