[Radiance-general] indirect/direct photometric

Sun, 25 May 2003 02:42:58 -0400

--Apple-Mail-14-375850595
Content-Transfer-Encoding: 7bit
Content-Type: text/plain;
	charset=US-ASCII;
	format=flowed

> From: John An <[email protected]>
> Date: Sun May 25, 2003  2:41:57  AM America/New_York
> To: [email protected]
> Subject: indirect/direct photometric
>
> Greg,
>
> Thanks again.
>
>
> Greg Ward wrote:
> I'll skip for now the question of why you're altering the output of
> ies2rad, and just try to answer your questions, first.
>
> I was following the instructions in Section 5.3.2.  It noted that for 
> direct/indirect luminaires, this special method should be used.  
> Besides, when I tried using the ies2rad command without modifying the 
> results, I ended up with visible geometry emitting the light.  While I 
> don't have a luminaire modeled yet, I eventually want to be able to 
> look at the luminaire once it is modeled, so I assumed that altering 
> the output of ies2rad so that an illum would eventually envelop the 
> luminaire was the way to go.  Is there a better way of doing this?
>
> Also, when I examined the dimensions of the luminaire contained in the 
> .ies file, it showed that the height was 0.  I'm assuming that the 
> height of 0 is to address the shadow banding effect.  However, a 
> non-existent height would mean that the geometry resulting from the 
> ies2rad would never enclose the geometry of the luminaire unless the 
> output of the ies2rad file was modified.
>
> So I guess my question now is, what is the best process of mapping the 
> photometric data to a 'complex' luminaire so that the luminaire is 
> visible?
>
>
>
> Greg Ward wrote:
> The -d? options affect the geometric dimensions that are output when
> you use the geometry supplied in the ies file.  When you specify your
> own illum sphere with the -i option, it assumes you mean to use the
> units you specify with the -d? option.  All that really changes is the
> first real argument to the brightdata primitive, which scales the vaues
> in the *.dat file appropriately.  The data file itself is (as you
> noted) unchanged.
>
> Understood.  Since my model is in inches, I should be using the -di 
> option.
>
>
>
> Greg Ward wrote:
> Are you visualizing the source in rview?  If so, the defaults of -ds 0
> and -dj 0 mean that all light soruces will appear as if eminating from
> a point.  If you set -ds .2 and -dj .6 or so, you should get the result
> you expect.
>
> Yes, I was using rview (rview -vf *.vf -ab 3 *.oct).  As soon as I 
> added the -ds .2 and -dj .6 options, I saw a linear profile.
>
> Thanks again, for putting up with my naive questions.
>
>
> John

--Apple-Mail-14-375850595
Content-Transfer-Encoding: 7bit
Content-Type: text/enriched;
	charset=US-ASCII

<excerpt><bold>From: </bold>John An <<[email protected]>

<bold>Date: </bold>Sun May 25, 2003  2:41:57  AM America/New_York

<bold>To: </bold>[email protected]

<bold>Subject: </bold>indirect/direct photometric

<fontfamily><param>Courier</param><bigger>Greg,

Thanks again.

<italic>

Greg Ward wrote:

I'll skip for now the question of why you're altering the output of 

ies2rad, and just try to answer your questions, first.</italic>

I was following the instructions in Section 5.3.2.  It noted that for
direct/indirect luminaires, this special method should be used. 
Besides, when I tried using the ies2rad command without modifying the
results, I ended up with visible geometry emitting the light.  While I
don't have a luminaire modeled yet, I eventually want to be able to
look at the luminaire once it is modeled, so I assumed that altering
the output of ies2rad so that an illum would eventually envelop the
luminaire was the way to go.  Is there a better way of doing this?

Also, when I examined the dimensions of the luminaire contained in the
.ies file, it showed that the height was 0.  I'm assuming that the
height of 0 is to address the shadow banding effect.  However, a
non-existent height would mean that the geometry resulting from the
ies2rad would never enclose the geometry of the luminaire unless the
output of the ies2rad file was modified.

So I guess my question now is, what is the best process of mapping the
photometric data to a 'complex' luminaire so that the luminaire is
visible?

<italic>Greg Ward wrote:

The -d? options affect the geometric dimensions that are output when 

you use the geometry supplied in the ies file.  When you specify your 

own illum sphere with the -i option, it assumes you mean to use the 

units you specify with the -d? option.  All that really changes is the 

first real argument to the brightdata primitive, which scales the
vaues 

in the *.dat file appropriately.  The data file itself is (as you 

noted) unchanged.</italic>

Understood.  Since my model is in inches, I should be using the -di
option.

<italic>Greg Ward wrote:

Are you visualizing the source in rview?  If so, the defaults of -ds 0 

and -dj 0 mean that all light soruces will appear as if eminating from 

a point.  If you set -ds .2 and -dj .6 or so, you should get the
result 

you expect.</italic>

Yes, I was using rview (rview -vf *.vf -ab 3 *.oct).  As soon as I
added the -ds .2 and -dj .6 options, I saw a linear profile.

Thanks again, for putting up with my naive questions.

John

</bigger></fontfamily></excerpt>
--Apple-Mail-14-375850595--