[Radiance-general] Sun Matrix by gendaymtx and genskyvec

[Greg Ward]

Hi Tommy,

Looks like no one responded to your question.  I'm not sure what the correct response is, as the 5-phase method only applies to annual calculations.  There is a "-d" option to genskyvec that outputs zero values except for the sun, but that sun is spread over 3 adjacent patches, which is what the "-5" option in genskymtx avoids.

If you really want to reproduce the 5-phase calculation at a single time point for some reason, why don't you edit the input file to gendaymtx to contain only that time point?  You can run the rest of the calculation as you would for an annual simulation, I think.

-Greg

> From: "WEI, Minchen [BSE]" <minchen.wei at polyu.edu.hk>
> Date: December 9, 2017 5:20:15 AM PST
> 
> When we are generating the sun matrix for the Five Phase and the improved version of the Daylight Coefficient Method, gendaymtx is used in Sarith’s tutorial file. “-5 0.533” is used.
>  
> Is there a simply way to generate the sun matrix by using genskyvec? Or it can only be applied to gendaymtx.
>  
> Thanks,
>  
> Tommy
>  
> --
> Minchen (Tommy) Wei, PhD
> The Hong Kong Polytechnic University
> (+852)-3400 3606 minchen.wei at polyu.edu.hk
>  
> From: Sarith Subramaniam [mailto:sarith at psu.edu] 
> Sent: Friday, 8 December 2017 10:30 PM
> To: Radiance general discussion <radiance-general at radiance-online.org>
> Subject: Re: [Radiance-general] Proxy geometry in the five-phase method
>  
> The Daylight Coefficient and Three Phase examples use a sky with 145 patches in which the sun accounts for 3-4 patches. In these examples, the sky and sun are both represented through the "glow" primitive and the ray-tracing is stochastic. The additional steps in the Five Phase example subtract out the direct component of the results from the glow-based oversized sun patches and replace them with replace them with accurately calculated direct sun-results. In this case, the sun is represented using the "light" primitive (implying a deterministic calculation) and is also more appropriately sized (0.533 degrees).
> 
> The exercise files also contain an improved version of the Daylight Coefficient Method that features the direct-sun correction. I think the results from that calculation and the Five Phase calculation will be similar as they both involve the direct-sun correction.
> 
> The final part of the Five-Phase method requires Tensor-Tree BSDFs. The methodology for creating BSDFs from Radiance primitives is discussed in the genBSDF tutorial. If your glazing system is just a simple glazing comprised of something like "glass" or "trans", you can incorporate the glazing geometry directly. This process would be similar to what is discussed for the improved Daylight Coefficient calculation.
> 
>  
> 
> Sarith
> 
>  
> 
>  
> On 12/8/2017 4:11 AM, WEI, Minchen [BSE] wrote:
> Hi All,
>  
> I have questions regarding the proxy geometry in the five-phase method.
>  
> I tried the tutorial using a room with a glazing (no venetian blinds). The daylight coefficient and three-phase method can produce similar results, but the five-phase method produce smaller illuminance values.
>  
> I think the problem is coming from the “proxy geometry”. In my “proxy geometry” I only included a glazing geometry with the transmittance I used in the daylight coefficient method. Is that right?
>  
> I looked at the proxy geometry included in Sarith’s tutorial which included all the blinds, the glazing, and a BSDF (“blindsT4c.xml”) for maxPoly and minPoly. How should I generate these BSDF if I have the blinds?
>  
> Thanks,
>  
> Tommy
>  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.radiance-online.org/pipermail/radiance-general/attachments/20171215/16a57851/attachment-0001.html>