[Radiance-general] IES files and MGF geometry

[Axel Jacobs]

Hi Greg,

I only know about this MGF embedding from the ies2rad man page, and
thought it might solve my problem:

" -i rad    Ignore the crude geometry given by the IES input file and
use  instead  an  illum
                 sphere  with radius rad.  This option may be useful
when the user wishes to add a
                 more accurate geometric description to the light
source model, though  this  need
                 is  obviated by the recent LM-63-1995 specification,
which uses MGF detail geome-
                 try.  (See -g option below.)"

This sounds rather more positive than the situation appears to be now.  Oh well.

As you say--I would need near-field for my bollard.  Thank for sending
the projection formula.  With the NF-correction for boxes already in
the code, would a square bollard (again w/o bottom and top) be any
easier to describe?  A square imposter is still a lot better than a
hovering crystal ball.

Many thanks

Best
Axel




On 5 November 2014 16:00, Greg Ward <gregoryjward at gmail.com> wrote:
> Hi Axel,
>
> Regarding MGF, I wasn't aware that IES ever gave it's official support to the LUMINOUSGEOMETRY tag, and I don't know of any manufacturer's that included this information.  After working on the format for many months, I was rather disappointed that they declined to use it, so I'd be very happy to know otherwise.  The intent of MGF was to provide detail geometry for the fixture's appearance, but it does not technically interact with the photometry.  That would be handled by "imposter" geometry (an illum sphere or similar) enclosing the MGF description.
>
> Regarding your actual problem of representing your luminous geometry as an open cylinder rather than two disks, you will need to do this by hand.  The IES luminaire standard had some odd ways of specifying geometry, involving negative dimensions in the (height,width,depth) fields, and all geometry was assumed to be closed.  The current source.cal file has projection calculations needed to convert candelas (as given in the IES file) to radiance (watts/sr/m^2) as needed for rendering, but only for closed cylinders.  You would have to roll your own if you wanted to remove the end caps to have the correct factors in there.  The current projected area is approximated as:
>
> cylprojection = A2*A3*sqrt(1-Dz*Dz) + PI/4*A2*A2*abs(Dz);
>
> This is only valid for far-field -- I don't have a local projection factor for cylinders in there, like I do for boxes.  You probably need one in your case, but I'm not sure off-hand how to compute it.  The far-field approximation for just the cylinder section would be the first term, above.
>
> Hope this helps!
>
> Cheers,
> -Greg
>
>> From: Christopher Rush <Christopher.Rush at arup.com>
>> Subject: Re: [Radiance-general] IES files and MGF geometry
>> Date: November 5, 2014 7:14:18 AM PST
>>
>> Do I understand that the problem is not generating geometry much taller than the actual lens, but rather that that it generates a cylinder with top and bottom circles, while only the shell of the cylinder is the actual light emitter? I had started to reply regarding adjusting the height in the IES file, but maybe someone else knows a solution for the top/bottom surface problem - which would be good to know.
>
> ++++++
>
>> From: "Guglielmetti, Robert" <Robert.Guglielmetti at nrel.gov>
>> Subject: Re: [Radiance-general] IES files and MGF geometry
>> Date: November 5, 2014 7:41:17 AM PST
>>
>> Ah, I "see". So you have this thing that throws down a great deal, so a
>> lot of the candelas toward nadir are being mapped to that bottom ring in
>> the ies2rad translation. My sense is that for the best result you'd want
>> to use the illum sphere with a large-ish radius ‹ large enough so that the
>> output gets mapped to points on the sphere that would fall outside the
>> supporting post impostor geometry, but small enough that the sphere is
>> high enough off the ground that the projected light looks "right".
>> Obviously for critical illuminance calcs you could do one run with just
>> the emitters (which seem to have been used in the images on the
>> manufacturer's website), and then for visualization you could use the
>> illum sphere trick, or get fancy with lboxcorr...
>
> -----Original Message-----
>
>> The product I'm looking at is this:
>> http://products.iguzzini.com/iway_round
>> which is why I want to get rid of the top and bottom bits of the
>> ies2rad geometry.
>>
>> What I was hoping to be able to do with MGF is tell it what the
>> fitting looks like.  In this particular case, I'd model 3 stacked
>> cylinders, enclosed with a disk at the top. Only the middle cylinder
>> would act as an emitter.  Does that make sense?
>>
>> Best
>> Axel
>
>
> _______________________________________________
> Radiance-general mailing list
> Radiance-general at radiance-online.org
> http://www.radiance-online.org/mailman/listinfo/radiance-general