gensky_-B option

Dear experts,

I am testing glare at restaurants of Barcelona for my thesis. I am using
Evalglare v 1.11 with real illuminance (-I Ev) data for photographs
converted to hdr by WebHDR and for visualizations got by Radiance.
Furthermore, I am testing glare by DIVA. I have tested with both, Evalglare
and DIVA’s Evalglare with Clear Sky without Illuminance data and with
illuminance data.

# gensky 7 22 11 +s -a 41 -o -2 -m -15

# gensky 7 22 11 +s -a 41 -o -2 -m -15 -B 458.100

I have taken photograph’s Daylight Glare Probability index with real
illuminance data as reference. Therefore, without Ev data, DGP’s with
Clear Sky without illuminance data is more similar than Sky with
illuminance data (-B):

dgp photographs= 71%
dgp DIVA Sky without illuminance data= 63%
dgp DIVA Sky with illuminance data (-B)= 100%
dgp (just Radiance-Evalglare) Clear Sky without illuminance data= 62%
dgp (just Radiance-Evalglare) Clear Sky with illuminance data (-B)= 100%

My illuminance data has both, direct and diffuse irradiance. Could someone
confirm if –B parameter is only for diffuse radiance? I think, that is
explained in the program specifications.
Thank you in advance.

Kind regards,
Urtza.

Dear Urtza,

it is not 100% clear to me what you have measured besides your image. Do you have an illuminance value also at camera level? This is very much needed to check the validity of the image, especially when you have spots of high luminances (e.g. the sun or reflections of it) in the field of view. Do you have an fish-eye lens?
So the first step is to check compare the calculated illuminance from you fish-eye image with this value to check if you have problems with pixel overflow/saturation. This would explain big differences between image calculated illuminance and measured. If you don't have a fish-eye lens you cannot make this simple check. Then you need to measure with a spot luminance meter the highest luminance in the field of view and compare it with the value in the image.

If you have measurements of the Direct and Diffuse Irradiation, then it is better to use gendaylit, where you can provide both values.

For your simulated data, there must be an mistake in your workflow. If you calculate the illuminance by "rtrace -I" you should get the same value for the illuminance than calculated from the image (you can calculate the illuminance with evalglare -V . If you use the -I option in evalglare, it is just replacing the internal calculated illuminance value by the external one, all other algorithms are the same. So if you get different values, then you have a problem either with your image or with the illumiance calculation. So check first the difference in the illuminance. In case you are simulating a trans-material (e.g. for a fabric), pixel sampling, limit weight ect. might play a big role and could cause huge differences between image and rtrace calculations. But there could be many other reasons, depending on you scene you are simulating. Then parameter settings play a big role.

And another important comment is, that I don't believe that the DGP is a metrics to describe glare in a restaurant. The needs are very much different between an office and a restaurant. In a restaurant you are dealing mainly with disability glare, I'm not sure if discomfort glare is perceived so much for this kind of usage. Of course with the DGP you can compare different variants and judge which variant has more often glare than other ones. But I would say you cannot make a distinct glare evaluation in a restaurant using DGP.

Jan

···

Am 11/26/15 um 9:30 AM schrieb [email protected]:

Dear experts,

I am testing glare at restaurants of Barcelona for my thesis. I am using Evalglare v 1.11 with real illuminance (-I Ev) data for photographs converted to hdr by WebHDR and for visualizations got by Radiance. Furthermore, I am testing glare by DIVA. I have tested with both, Evalglare and DIVA�s Evalglare with Clear Sky without Illuminance data and with illuminance data.

# gensky 7 22 11 +s -a 41 -o -2 -m -15

# gensky 7 22 11 +s -a 41 -o -2 -m -15 -B 458.100

I have taken photograph�s Daylight Glare Probability index with real illuminance data as reference. Therefore, without Ev data, DGP�s with Clear Sky without illuminance data is more similar than Sky with illuminance data (-B):

dgp photographs= 71%
dgp DIVA Sky without illuminance data= 63%
dgp DIVA Sky with illuminance data (-B)= 100%
dgp (just Radiance-Evalglare) Clear Sky without illuminance data= 62%
dgp (just Radiance-Evalglare) Clear Sky with illuminance data (-B)= 100%

My illuminance data has both, direct and diffuse irradiance. Could someone confirm if �B parameter is only for diffuse radiance? I think, that is explained in the program specifications.
Thank you in advance.

Kind regards,
Urtza.

_______________________________________________
Radiance-general mailing list
[email protected]
http://www.radiance-online.org/mailman/listinfo/radiance-general

--
Dr.-Ing. Jan Wienold
Ecole Polytechnique F�d�rale de Lausanne (EPFL)
EPFL ENAC IA LIPID

http://people.epfl.ch/jan.wienold
LE 1 111 (Office)
Phone +41 21 69 30849

Dear Jan,

Thank you very much for your answer and suggestions.

I had illuminance value also at camera level, because I had spots of high
luminance. I have an fish-eye lens, the specifications said that it gets
180º but it catches 160º. I hope that it can work. Ok, first I will check
if calculated illuminance is the same as measured.

I understand that perhaps DGP is not the best option to describe glare in a
restaurant, excuse me. However, we have problems with highly glazed façade
that provides very high luminance values to achieve good concentration on
the taste and calm quality atmosphere, as well as, privacy. We have done
surveys and usually people tolerate more visual discomfort in this
activity, due to the habit probably, but when you show atmospheres with
less out luminance values, they prefer. I am not sure if I am ok, but, the
perception results are quite similar to the DGP results. Anyway, thank you
again and I will listen to you and check all variations to demonstrate this
problem.

Yours sincerely,
Urtza.

Jan Wienold <[email protected]> escribió:

Dear Urtza,

it is not 100% clear to me what you have measured besides your image. Do
you have an illuminance value also at camera level? This is very much
needed to check the validity of the image, especially when you have
spots of high luminances (e.g. the sun or reflections of it) in the
field of view. Do you have an fish-eye lens?
So the first step is to check compare the calculated illuminance from
you fish-eye image with this value to check if you have problems with
pixel overflow/saturation. This would explain big differences between
image calculated illuminance and measured. If you don't have a fish-eye
lens you cannot make this simple check. Then you need to measure with a
spot luminance meter the highest luminance in the field of view and
compare it with the value in the image.

If you have measurements of the Direct and Diffuse Irradiation, then it
is better to use gendaylit, where you can provide both values.

For your simulated data, there must be an mistake in your workflow. If
you calculate the illuminance by "rtrace -I" you should get the same
value for the illuminance than calculated from the image (you can
calculate the illuminance with evalglare -V . If you use the -I option
in evalglare, it is just replacing the internal calculated illuminance
value by the external one, all other algorithms are the same. So if you
get different values, then you have a problem either with your image or
with the illumiance calculation. So check first the difference in the
illuminance. In case you are simulating a trans-material (e.g. for a
fabric), pixel sampling, limit weight ect. might play a big role and
could cause huge differences between image and rtrace calculations. But
there could be many other reasons, depending on you scene you are
simulating. Then parameter settings play a big role.

And another important comment is, that I don't believe that the DGP is a
metrics to describe glare in a restaurant. The needs are very much
different between an office and a restaurant. In a restaurant you are
dealing mainly with disability glare, I'm not sure if discomfort glare
is perceived so much for this kind of usage. Of course with the DGP you
can compare different variants and judge which variant has more often
glare than other ones. But I would say you cannot make a distinct glare
evaluation in a restaurant using DGP.

Jan

  Dear experts,

I am testing glare at restaurants of Barcelona for my thesis. I am using
Evalglare v 1.11 with real illuminance (-I Ev) data for photographs
converted to hdr by WebHDR and for visualizations got by Radiance.
Furthermore, I am testing glare by DIVA. I have tested with both, Evalglare
and DIVA’s Evalglare with Clear Sky without Illuminance data and with
illuminance data.

# gensky 7 22 11 +s -a 41 -o -2 -m -15

# gensky 7 22 11 +s -a 41 -o -2 -m -15 -B 458.100

I have taken photograph’s Daylight Glare Probability index with real
illuminance data as reference. Therefore, without Ev data, DGP’s with
Clear Sky without illuminance data is more similar than Sky with
illuminance data (-B):

dgp photographs= 71%
dgp DIVA Sky without illuminance data= 63%
dgp DIVA Sky with illuminance data (-B)= 100%
dgp (just Radiance-Evalglare) Clear Sky without illuminance data= 62%
dgp (just Radiance-Evalglare) Clear Sky with illuminance data (-B)= 100%

My illuminance data has both, direct and diffuse irradiance. Could someone
confirm if –B parameter is only for diffuse radiance? I think, that is
explained in the program specifications.
Thank you in advance.

Kind regards,
Urtza.

_______________________________________________ Radiance-general mailing
list
[email protected]://www.radiance-online.org/mailman/listinfo/radiance-general
-- Dr.-Ing. Jan Wienold Ecole Polytechnique Fédérale de Lausanne (EPFL)
EPFL ENAC IA LIPID http://people.epfl.ch/jan.wienoldLE 1 111 (Office)
Phone +41 21 69 30849

···

  Am 11/26/15 um 9:30 AM schrieb [email protected]:

Dear Urtza,

the 160 instead of 180 degree should not be a big problem, of course you underestimate the illuminance a bit, but you can provide the external measured illuminance value.

In case you have a problem in matching the calculated illuminance of the captured image (evalglare -V) with the measured illuminance, caused by pixel overflow, there exist since long time an undocumented option in evalglare to adjust the overflow pixels to match the measured illuminance. I could help you with that - in that case send me one example image and the related measured illuminance value directly (not via discussion group).

Jan

···

Am 11/26/15 um 4:59 PM schrieb [email protected]:

Dear Jan,

Thank you very much for your answer and suggestions.

I had illuminance value also at camera level, because I had spots of high luminance. I have an fish-eye lens, the specifications said that it gets 180� but it catches 160�. I hope that it can work. Ok, first I will check if calculated illuminance is the same as measured.

I understand that perhaps DGP is not the best option to describe glare in a restaurant, excuse me. However, we have problems with highly glazed fa�ade that provides very high luminance values to achieve good concentration on the taste and calm quality atmosphere, as well as, privacy. We have done surveys and usually people tolerate more visual discomfort in this activity, due to the habit probably, but when you show atmospheres with less out luminance values, they prefer. I am not sure if I am ok, but, the perception results are quite similar to the DGP results. Anyway, thank you again and I will listen to you and check all variations to demonstrate this problem.

Yours sincerely,
Urtza.

Jan Wienold <[email protected] <mailto:[email protected]>> escribi�:

Dear Urtza,

it is not 100% clear to me what you have measured besides your image. Do you have an illuminance value also at camera level? This is very much needed to check the validity of the image, especially when you have spots of high luminances (e.g. the sun or reflections of it) in the field of view. Do you have an fish-eye lens?
So the first step is to check compare the calculated illuminance from you fish-eye image with this value to check if you have problems with pixel overflow/saturation. This would explain big differences between image calculated illuminance and measured. If you don't have a fish-eye lens you cannot make this simple check. Then you need to measure with a spot luminance meter the highest luminance in the field of view and compare it with the value in the image.

If you have measurements of the Direct and Diffuse Irradiation, then it is better to use gendaylit, where you can provide both values.

For your simulated data, there must be an mistake in your workflow. If you calculate the illuminance by "rtrace -I" you should get the same value for the illuminance than calculated from the image (you can calculate the illuminance with evalglare -V . If you use the -I option in evalglare, it is just replacing the internal calculated illuminance value by the external one, all other algorithms are the same. So if you get different values, then you have a problem either with your image or with the illumiance calculation. So check first the difference in the illuminance. In case you are simulating a trans-material (e.g. for a fabric), pixel sampling, limit weight ect. might play a big role and could cause huge differences between image and rtrace calculations. But there could be many other reasons, depending on you scene you are simulating. Then parameter settings play a big role.

And another important comment is, that I don't believe that the DGP is a metrics to describe glare in a restaurant. The needs are very much different between an office and a restaurant. In a restaurant you are dealing mainly with disability glare, I'm not sure if discomfort glare is perceived so much for this kind of usage. Of course with the DGP you can compare different variants and judge which variant has more often glare than other ones. But I would say you cannot make a distinct glare evaluation in a restaurant using DGP.

Jan

Am 11/26/15 um 9:30 AM schrieb [email protected]:

Dear experts,

I am testing glare at restaurants of Barcelona for my thesis. I am using Evalglare v 1.11 with real illuminance (-I Ev) data for photographs converted to hdr by WebHDR and for visualizations got by Radiance. Furthermore, I am testing glare by DIVA. I have tested with both, Evalglare and DIVA�s Evalglare with Clear Sky without Illuminance data and with illuminance data.

# gensky 7 22 11 +s -a 41 -o -2 -m -15

# gensky 7 22 11 +s -a 41 -o -2 -m -15 -B 458.100

I have taken photograph�s Daylight Glare Probability index with real illuminance data as reference. Therefore, without Ev data, DGP�s with Clear Sky without illuminance data is more similar than Sky with illuminance data (-B):

dgp photographs= 71%
dgp DIVA Sky without illuminance data= 63%
dgp DIVA Sky with illuminance data (-B)= 100%
dgp (just Radiance-Evalglare) Clear Sky without illuminance data= 62%
dgp (just Radiance-Evalglare) Clear Sky with illuminance data (-B)= 100%

My illuminance data has both, direct and diffuse irradiance. Could someone confirm if �B parameter is only for diffuse radiance? I think, that is explained in the program specifications.
Thank you in advance.

Kind regards,
Urtza.

_______________________________________________
Radiance-general mailing list
[email protected]
http://www.radiance-online.org/mailman/listinfo/radiance-general

--
Dr.-Ing. Jan Wienold
Ecole Polytechnique F�d�rale de Lausanne (EPFL)
EPFL ENAC IA LIPID

http://people.epfl.ch/jan.wienold
LE 1 111 (Office)
Phone +41 21 69 30849

_______________________________________________
Radiance-general mailing list
[email protected]
http://www.radiance-online.org/mailman/listinfo/radiance-general

--
Dr.-Ing. Jan Wienold
Ecole Polytechnique F�d�rale de Lausanne (EPFL)
EPFL ENAC IA LIPID

http://people.epfl.ch/jan.wienold
LE 1 111 (Office)
Phone +41 21 69 30849

Dear Jan,

Ok. I checked the -V value and it scripts me for each image correctly and
adding -i measured value DGP is the same. In case, I will contact you.
Thank you very much for your attention.

Yours sincerely,
Urtza.

Jan Wienold <[email protected]> escribió:

Dear Urtza,

the 160 instead of 180 degree should not be a big problem, of course you
underestimate the illuminance a bit, but you can provide the external
measured illuminance value.

In case you have a problem in matching the calculated illuminance of the
captured image (evalglare -V) with the measured illuminance, caused by
pixel overflow, there exist since long time an undocumented option in
evalglare to adjust the overflow pixels to match the measured
illuminance. I could help you with that - in that case send me one
example image and the related measured illuminance value directly (not
via discussion group).

Jan

  Dear Jan,

Thank you very much for your answer and suggestions.

I had illuminance value also at camera level, because I had spots of high
luminance. I have an fish-eye lens, the specifications said that it gets
180º but it catches 160º. I hope that it can work. Ok, first I will check
if calculated illuminance is the same as measured.

I understand that perhaps DGP is not the best option to describe glare in a
restaurant, excuse me. However, we have problems with highly glazed façade
that provides very high luminance values to achieve good concentration on
the taste and calm quality atmosphere, as well as, privacy. We have done
surveys and usually people tolerate more visual discomfort in this
activity, due to the habit probably, but when you show atmospheres with
less out luminance values, they prefer. I am not sure if I am ok, but, the
perception results are quite similar to the DGP results. Anyway, thank you
again and I will listen to you and check all variations to demonstrate this
problem.

Yours sincerely,
Urtza.

Jan Wienold <[email protected]> escribió:

Dear Urtza,

it is not 100% clear to me what you have measured besides your image. Do
you have an illuminance value also at camera level? This is very much
needed to check the validity of the image, especially when you have
spots of high luminances (e.g. the sun or reflections of it) in the
field of view. Do you have an fish-eye lens?
So the first step is to check compare the calculated illuminance from
you fish-eye image with this value to check if you have problems with
pixel overflow/saturation. This would explain big differences between
image calculated illuminance and measured. If you don't have a fish-eye
lens you cannot make this simple check. Then you need to measure with a
spot luminance meter the highest luminance in the field of view and
compare it with the value in the image.

If you have measurements of the Direct and Diffuse Irradiation, then it
is better to use gendaylit, where you can provide both values.

For your simulated data, there must be an mistake in your workflow. If
you calculate the illuminance by "rtrace -I" you should get the same
value for the illuminance than calculated from the image (you can
calculate the illuminance with evalglare -V . If you use the -I option
in evalglare, it is just replacing the internal calculated illuminance
value by the external one, all other algorithms are the same. So if you
get different values, then you have a problem either with your image or
with the illumiance calculation. So check first the difference in the
illuminance. In case you are simulating a trans-material (e.g. for a
fabric), pixel sampling, limit weight ect. might play a big role and
could cause huge differences between image and rtrace calculations. But
there could be many other reasons, depending on you scene you are
simulating. Then parameter settings play a big role.

And another important comment is, that I don't believe that the DGP is a
metrics to describe glare in a restaurant. The needs are very much
different between an office and a restaurant. In a restaurant you are
dealing mainly with disability glare, I'm not sure if discomfort glare
is perceived so much for this kind of usage. Of course with the DGP you
can compare different variants and judge which variant has more often
glare than other ones. But I would say you cannot make a distinct glare
evaluation in a restaurant using DGP.

Jan

  Dear experts,

I am testing glare at restaurants of Barcelona for my thesis. I am using
Evalglare v 1.11 with real illuminance (-I Ev) data for photographs
converted to hdr by WebHDR and for visualizations got by Radiance.
Furthermore, I am testing glare by DIVA. I have tested with both, Evalglare
and DIVA’s Evalglare with Clear Sky without Illuminance data and with
illuminance data.

# gensky 7 22 11 +s -a 41 -o -2 -m -15

# gensky 7 22 11 +s -a 41 -o -2 -m -15 -B 458.100

I have taken photograph’s Daylight Glare Probability index with real
illuminance data as reference. Therefore, without Ev data, DGP’s with
Clear Sky without illuminance data is more similar than Sky with
illuminance data (-B):

dgp photographs= 71%
dgp DIVA Sky without illuminance data= 63%
dgp DIVA Sky with illuminance data (-B)= 100%
dgp (just Radiance-Evalglare) Clear Sky without illuminance data= 62%
dgp (just Radiance-Evalglare) Clear Sky with illuminance data (-B)= 100%

My illuminance data has both, direct and diffuse irradiance. Could someone
confirm if –B parameter is only for diffuse radiance? I think, that is
explained in the program specifications.
Thank you in advance.

Kind regards,
Urtza.

  

_______________________________________________ Radiance-general mailing
list
[email protected]://www.radiance-online.org/mailman/listinfo/radiance-general
  
-- Dr.-Ing. Jan Wienold Ecole Polytechnique Fédérale de Lausanne (EPFL)
EPFL ENAC IA LIPID http://people.epfl.ch/jan.wienoldLE 1 111 (Office)
Phone +41 21 69 30849

_______________________________________________ Radiance-general mailing
list
[email protected]://www.radiance-online.org/mailman/listinfo/radiance-general
-- Dr.-Ing. Jan Wienold Ecole Polytechnique Fédérale de Lausanne (EPFL)
EPFL ENAC IA LIPID http://people.epfl.ch/jan.wienoldLE 1 111 (Office)
Phone +41 21 69 30849

···

  Am 11/26/15 um 4:59 PM schrieb [email protected]:
  Am 11/26/15 um 9:30 AM schrieb [email protected]: