[Radiance-general] gendaymtx returns -1.#J -1.#J -1.#J

[Greg Ward]

Hi Mostapha,

If you run the command I gave you at the end of my e-mail, minus the "uniq" function that takes out the repeated values, you will get 145 solid angles for the 145 patches, starting as I said from the ZENITH patch.  You just need to find the klems_ang.cal file, which comes in the ray/src/cal/cal directory with the distribution.

The Reinhart sky source generator is called reinsrc.cal, and you can use it like this:

	cnt 2305 | rcalc -e MF:4 -f reinsrc.cal -e 'Rbin=recno;$1=Romega' > rh_sang.txt

You will 2305 values, get one solid angle for each patch.  If you change the MF setting, you will need to use a different number in cnt according to (using an example subdivision of 2):

	rcalc -n -e MF:2 -f reinhart.cal -e '$1=Nrbins'

The output of gendaymtx has 145 rows of 8760 values each for a full year, which may be less convenient for you.  You may swap rows and columns using the new rcollate command if you have it in your installation:

	gendaymtx year.wea | rcollate -i > transpose.txt

Cheers,
-Greg

> From: Mostapha Sadeghipour <sadeghipour at gmail.com>
> Date: October 4, 2013 7:07:24 AM PDT
> 
> Hi Greg,
> 
> I know it may not be the normal use of gendaymtx but I'm not using it to run 5Phase or 3Phase method.
> 
> What I'm looking for is the values for each sky patch itself (e.g. 577 numbers for each time step for Reinhart sky). The numbers are already generated from gendaymtx and I just want to multiply them with stradian angles to have the values in w/m2.
> 
> I assume it should be 15 numbers for 14 rows + zenith patch.
> 
> Cheers,
> Mostapha
> 
> PS: I wish I knew how to write a line like this on Windows then I could run it for reinhart.cal I assume. :-\
> 
> 
> On Thu, Oct 3, 2013 at 10:43 PM, Greg Ward <gregoryjward at gmail.com> wrote:
> Here are the solid angles (in steradians) for the full Klems rows:
> 
> 0.023909417
> 0.0237731163
> 0.0234119493
> 0.0273807841
> 0.0293333813
> 0.0349581625
> 0.0478687897
> 0.051801754
> 0.135517335
> 
> Note that we start at the zenith in the Klems subdivision, not at the horizon the way Tregenza's goes.
> 
> -Greg
> 
> P.S.  I used the command:
> 
> 	cnt 145 | rcalc -f klems_ang.cal -e 'kbin=recno;$1=omega' | uniq 
> 
>> From: Mostapha Sadeghipour <sadeghipour at gmail.com>
>> Date: October 3, 2013 10:21:55 AM PDT
>> 
>> Hi Greg,
>> 
>> Thank you for checking. My bad! I use my own written epw2wea convertor so the user don't need to download epw2wea file separately. That was a bug indeed. Sorry about that.
>> 
>> One more question about conversion of the results of gendaymtx. Where can I find  solid angle subtended for each patch of Reinhart sky? I already have the values for Tregenza sky as (0.0435, 0.0416, 0.0474, 0.0407, 0.0429, 0.0445, 0.0455, 0.0344) for the 8 rows. Is it fine to use the same numbers for every two rows?
>> 
>> Thanks again,
>> Mostapha
>> 
>> 
>> 
>> 
>> On Thu, Oct 3, 2013 at 12:21 PM, Gregory J. Ward <gregoryjward at gmail.com> wrote:
>> Hi Mostapha,
>> 
>> Thanks for sending your files.  For some reason, your WEA file has the month and day columns swapped, which is the source of your error.  What did you use to translate the EPW file?  The latest HEAD includes Christoph Reinharts epw2wea program, which does the correct translation.  With the correct input, gendaymtx seems to produce appropriate output.
>> 
>> Best,
>> -Greg
>> 
>>> From: Greg Ward <gward at lmi.net>
>>> Date: October 3, 2013 8:37:15 AM PDT
>>> 
>>> Hi Mostapha,
>>> 
>>> This sounds like a bug in gendaymtx.  Some versions of the standard library print out "-1.#J" for NaN (Not a Number), which can result from a divide-by-zero or similar illegal math operation.  If you send me your weather file in a separate message (not to the mailing list), I will see if I can reproduce the problem.  There may be a corner case or two that gendaymtx still doesn't deal with gracefully.
>>> 
>>> Cheers,
>>> -Greg
>>> 
>>>> From: Mostapha Sadeghipour <sadeghipour at gmail.com>
>>>> Date: October 2, 2013 1:37:49 PM PDT
>>>> 
>>>> Hi all,
>>>> 
>>>> I tried to read the results of the gendaymtx with a custom script and I noticed that there are multiple values of  -1.#J -1.#J -1.#J for RGB. Does anybody know why this is happening?
>>>> 
>>>> I'm trying to get the total solar radiation for sky patches (-O1). I'm running RADIANCE on Windows and here is the line:
>>>> 
>>>> gendaymtx -m 1 -O1 .\USA_AK_Anchorage.Intl.AP.702730_TMY3.wea> .\USA_AK_Anchorage.Intl.AP.702730_TMY3.mtx
>>>> 
>>>> Thanks in advance,
>>>> Mostapha
> http://www.radiance-online.org/mailman/listinfo/radiance-general
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.radiance-online.org/pipermail/radiance-general/attachments/20131004/ace67579/attachment-0001.html>