[Radiance-general] Glare analysis

[Melody Wong YT]

Dear all,

I have frustrated with the command "evalglare" and "findglare" in Radiance.

As I tried with DIVA before, and the function "point-in-time glare analysis" can be used with parallel rhino saved view for Radiance "evalglare" function.

However, when I used the same function in Radiance, error message popped "view must not be parallel". Does there anyone know why there are difference in the same function of "evalglare" and "findglare" in different software?
Can Radiance evaluate glare in rhino saved view rather than only fisheye view?

Apart from the above question, the time for simulating the glare in radiance and Diva are hugely difference. Does there any one experience Diva function and Radiance function before? And have insight about the calculation method between these two simulation softwares.

I have made the summaries the calculation file as below to see if it helps with the understanding.



Radiance (glare script)

Diva (Glare)

Sky file

!gensky 7 15 13:0 +s -a 22.33 -o -114.18 -m -120.00 -t 2.00 | xform -rz 0.00

skyfunc glow sky_glow
0
0
4 .986 .986 1.2020135 0

sky_glow source sky
0
0
4 0 0 1 180

skyfunc glow ground_glow
0
0
4 1 .7 .25 0

ground_glow source ground
0
0
4 0 0 -1 180

!gensky 12 15 08 +s -a 22 -o -114 -m -120

skyfunc glow sky_mat
0
0
4
1 1 1 0

sky_mat source sky
0
0
4
0 0 1 180

skyfunc glow ground_glow
0
0
4
1 .8 .5 0

ground_glow source ground
0
0
4
0 0 -1 180

Glare evaluation

rad -v view model_M12D15T8.rif REP=1

model_M12D15T8.rif :
ZONE = I 0 1 0 1 0 1
scene = SrdBldg.rad
scene = Bridge.rad
scene = Topo.rad
scene = Skylight.rad
scene = Roof.rad
scene = Glazing.rad
scene = Wall.rad
scene = BIPV.rad
materials = materials.mat
scene = sky_M12D15T8.rad
#add illum if needed
#illum=window.rad
view= view -vf view.vf
OCTREE= model.oct
EXPOSURE= 1
UP= Z
RESOLUTION= 512 512
QUALITY= M
PENUMBRAS= F
INDIRECT= 1
DETAIL= M
VARIABILITY= M
OPTFILE= model.opt
oconv= -f
#@mkillum= -ab 3 -ad 512 -as 256
render= -st .001 -dj .02 -ad 2048
pfilt= -2
#@northarrow= nonedefined xform -rz 0
# AMBFILE= Amb_M12D15T8.amb


oconv material.rad Mikki_sky.rad Mikki.rad > Mikki.oct

rpict -t 15 -vtv -vp 131.593 -15.381 43.05 -vd 97.297 -128.425 -16.081 -vu 0 0 1 -vh 75.8026519334564 -vv 53.9156001928024 -vs 0 -vl 0 -af Mikki.amb -x 1600 -y 1200 -ps 8 -pt .15 -pj .6 -dj 0 -ds .5 -dt .5 -dc .25 -dr 0 -dp 128 -st .85 -ab 2 -aa .25 -ar 128 -ad 1024 -as 1024 -lr 4 -lw .05 Mikki.oct > Mikki_Mikki_L.overture

del Mikki_Mikki_L.overture

rpict -t 15 -vtv -vp 131.593 -15.381 43.05 -vd 97.297 -128.425 -16.081 -vu 0 0 1 -vh 75.8026519334564 -vv 53.9156001928024 -vs 0 -vl 0 -af Mikki.amb -x 1600 -y 1200 -ps 8 -pt .15 -pj .6 -dj 0 -ds .5 -dt .5 -dc .25 -dr 0 -dp 128 -st .85 -ab 2 -aa .25 -ar 128 -ad 1024 -as 1024 -lr 4 -lw .05 Mikki.oct > Mikki_Mikki_L.unf
pfilt -r .6 -x /2 -y /2 Mikki_Mikki_L.unf > Mikki_Mikki_L.pic

del Mikki_Mikki_L.unf

evalglare -c Mikki_Mikki_L_evalglare.pic Mikki_Mikki_L.pic >Mikki_Mikki_L_evalglare.txt


Thank you very much.

Regards,
Melody Wong
Building Sustainability - Building 5
Arup
Level 5 Festival Walk, 80 Tat Chee Avenue Kowloon Tong Kowloon, Hong Kong
d: [phone] (ext. 84915)



____________________________________________________________
Electronic mail messages entering and leaving Arup business
systems are scanned for viruses and acceptability of content

[Christopher Rush]

The evalgare software is intended to be used with a 180° fisheye view to capture the whole visible hemisphere.

Regarding the time difference, there seems to be different rendering parameters resulting between the two simulation methods. The Radiance .rif file for the glare image is set to be 512 pixels while the Diva script seems to be up to 1200 pixels tall. You have one indirect bounce in the .rif file, but two (-ab 2) from the Diva script. If you check the model.opt file generated via your .rif file, you could compare them to the -ar, -ad, -aa and other settings in the Diva script and will probably find other rendering parameter differences causing a faster or slower (and less or more precise) rendering of the lighting conditions.

If you view the .pic and .hdr images from each method, does either one look like a realistic representation of your scene? Or do both look a bit simplified with not enough reflections accounted for?

-Chris
____________________________________________________________
Electronic mail messages entering and leaving Arup business
systems are scanned for viruses and acceptability of content

[Lars O. Grobe]

Hi Melody,

the fact that you got a result from Diva at all indicates that an old
version (e.g. 0.9) of evalglare is included in your Diva installation.
These old versions relied on the user to input a valid view, while
current evalglare (1.x) will check the image header to prevent user errors.

As Christopher already explained, without a fisheye covering a full
hemisphere it is not possible to compute the vertical eye illuminance
from the pixel values. As long as you know the illuminance a priori,
e.g. from a measurement or a separate computation, you could still pass
it as a command line option to evalglare. This is why current evalglare
supports other view types than fisheye projections.

However, for any glare evaluation you need to know the solid angle
represented by each pixel to find the size of the glare source in the
field of view. Therefore parallel views will not work out even if
illuminance is known.

So this is not a limitation of evalglare or findglare, but you just do
not provide the information that the glare metrics you are interested in
rely on.

Cheers, Lars.

Post by Melody Wong YT
Dear all,
I have frustrated with the command "evalglare" and "findglare" in Radiance.
As I tried with DIVA before, and the function "point-in-time glare analysis" can be used with parallel rhino saved view for Radiance "evalglare" function.
However, when I used the same function in Radiance, error message popped "view must not be parallel". Does there anyone know why there are difference in the same function of "evalglare" and "findglare" in different software?
Can Radiance evaluate glare in rhino saved view rather than only fisheye view?
Apart from the above question, the time for simulating the glare in radiance and Diva are hugely difference. Does there any one experience Diva function and Radiance function before? And have insight about the calculation method between these two simulation softwares.
I have made the summaries the calculation file as below to see if it helps with the understanding.
Radiance (glare script)
Diva (Glare)
Sky file
!gensky 7 15 13:0 +s -a 22.33 -o -114.18 -m -120.00 -t 2.00 | xform -rz 0.00
skyfunc glow sky_glow
0
0
4 .986 .986 1.2020135 0
sky_glow source sky
0
0
4 0 0 1 180
skyfunc glow ground_glow
0
0
4 1 .7 .25 0
ground_glow source ground
0
0
4 0 0 -1 180
!gensky 12 15 08 +s -a 22 -o -114 -m -120
skyfunc glow sky_mat
0
0
4
1 1 1 0
sky_mat source sky
0
0
4
0 0 1 180
skyfunc glow ground_glow
0
0
4
1 .8 .5 0
ground_glow source ground
0
0
4
0 0 -1 180
Glare evaluation
rad -v view model_M12D15T8.rif REP=1
ZONE = I 0 1 0 1 0 1
scene = SrdBldg.rad
scene = Bridge.rad
scene = Topo.rad
scene = Skylight.rad
scene = Roof.rad
scene = Glazing.rad
scene = Wall.rad
scene = BIPV.rad
materials = materials.mat
scene = sky_M12D15T8.rad
#add illum if needed
#illum=window.rad
view= view -vf view.vf
OCTREE= model.oct
EXPOSURE= 1
UP= Z
RESOLUTION= 512 512
QUALITY= M
PENUMBRAS= F
INDIRECT= 1
DETAIL= M
VARIABILITY= M
OPTFILE= model.opt
oconv= -f
render= -st .001 -dj .02 -ad 2048
pfilt= -2
# AMBFILE= Amb_M12D15T8.amb
oconv material.rad Mikki_sky.rad Mikki.rad > Mikki.oct
rpict -t 15 -vtv -vp 131.593 -15.381 43.05 -vd 97.297 -128.425 -16.081 -vu 0 0 1 -vh 75.8026519334564 -vv 53.9156001928024 -vs 0 -vl 0 -af Mikki.amb -x 1600 -y 1200 -ps 8 -pt .15 -pj .6 -dj 0 -ds .5 -dt .5 -dc .25 -dr 0 -dp 128 -st .85 -ab 2 -aa .25 -ar 128 -ad 1024 -as 1024 -lr 4 -lw .05 Mikki.oct > Mikki_Mikki_L.overture
del Mikki_Mikki_L.overture
rpict -t 15 -vtv -vp 131.593 -15.381 43.05 -vd 97.297 -128.425 -16.081 -vu 0 0 1 -vh 75.8026519334564 -vv 53.9156001928024 -vs 0 -vl 0 -af Mikki.amb -x 1600 -y 1200 -ps 8 -pt .15 -pj .6 -dj 0 -ds .5 -dt .5 -dc .25 -dr 0 -dp 128 -st .85 -ab 2 -aa .25 -ar 128 -ad 1024 -as 1024 -lr 4 -lw .05 Mikki.oct > Mikki_Mikki_L.unf
pfilt -r .6 -x /2 -y /2 Mikki_Mikki_L.unf > Mikki_Mikki_L.pic
del Mikki_Mikki_L.unf
evalglare -c Mikki_Mikki_L_evalglare.pic Mikki_Mikki_L.pic >Mikki_Mikki_L_evalglare.txt
Thank you very much.
Regards,
Melody Wong
Building Sustainability - Building 5
Arup
Level 5 Festival Walk, 80 Tat Chee Avenue Kowloon Tong Kowloon, Hong Kong
d: [phone] (ext. 84915)
____________________________________________________________
Electronic mail messages entering and leaving Arup business
systems are scanned for viruses and acceptability of content
_______________________________________________
Radiance-general mailing list
https://www.radiance-online.org/mailman/listinfo/radiance-general