[Radiance-general] Three-phase Method dctimestep

[Greg Ward]

It's a little bit complicated by whether or not reciprocity is needed, but the details are contained in the routines of src/util/cmbsdf.c.

When reciprocity is not needed, you must still multiply each direction by the projected solid angle of the incident directions (columns).

-Greg

> From: jinsook kang <kjstotoro at naver.com>
> Subject: Re: [Radiance-general] Three-phase Method dctimestep
> Date: September 1, 2014 11:14:56 AM GMT+01:00
> 
> Thank you very much for your answer Andrew.
> 
> Can you explain me more in detail about the required "conversion" for BSDF xml ?
> 
> Thank you 
> 
> Jinsook
> 
>  
> 
> -----Original Message-----
> From: "Andrew McNeil"<amcneil at lbl.gov> 
> To: "Radiance general discussion"<radiance-general at radiance-online.org>; 
> Cc: "jinsook kang"<kjstotoro at naver.com>; 
>  
> 
> When dctimestep reads the BSDF xml, there is a conversion required for it to be a true coefficient matrix. If you're giving it a BSDF, I suspect this might be one of the causes of the discrepency you're observing.
> Andy
> 
> 
> On Thu, Aug 28, 2014 at 9:39 AM, CHI-German Molina <gmolina at hdlao.com> wrote:
> Jinsook,
>  
> DCTIMESTEP does multiply matrices, a pretty naive method, actually. I have no idea what might be causing the dismatch between your results and DCTIMESTEP's.
>  
> Nothing too helpful... sorry.
>  
> Bye!
>  
>  
> 
> 
> 2014-08-28 12:27 GMT-04:00 jinsook kang <kjstotoro at naver.com>:
> Hi everyone.
> 
> Is there anybody who knows how dctimestep works in three phase method?
> 
> When I tried to multiply as a normal matrix (with python language), the illuminance result is not matched with dctimestep result.
> 
> So, I wonder if dctimestep doesn't use normal matrix multiplication.
> 
> If there is a specific method in dctimestep, please teach me.
> 
> * let you know,
> 
> 1. I picked up one column out of the every RGB(view matrix and daylight matrix), so that I made a simple matrix multiplication.
> 
> 2. I created daylight matrix as a ASCII data instead of binary.
> 
> Thank you.
> 
> Jinsook
> 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.radiance-online.org/pipermail/radiance-general/attachments/20140901/ac6bbc4f/attachment.html>