ECOTECT_SENSOR POINT CREATE

[arindarrc]

Hello, I am searching for a way to -generate the sensor point file *.pts with just a few points -how to import it into Ecotect -how to feed it into Daysim -how to reimport it back so I can see graphically the points with their values in Ecotect or RHino so that i get the result of a daylight simulation only on this points. This would trim down the calculation time lacking CPU.. I would kindly appreciate any hint and advices, since strolling through blogs and trying Optics, Window, Comfen, brought me very far from successful sofar. Kind Regards from Vienna, Stella

[Pennetier1 · Accepted solution]

Hello Stella,

You do not really have to create a .pts file for this if your model is already in Ecotect - and I assume it is since you wish to re-impoert the results into Ecotect.  

All you need to do to switch from a grid analysis to a single point analysis is to create these points using the Point tool which is located on the left tool panel (it looks like a Sputnik -looking object).  You can add them one by one throughout your model, or use the modify menu > surface subdivision > sensor points, if you have selected a particular plane in your model and want to space the points regularly.

Once your points are in the model, simply run the daylighting calculations as you normally would with an analysis grid except that you would select "Object in Model > Use only Selected Objects (your points)" in Step 3 of 11 of the Radiance Analysis wizard.  If you have DAYSIM installed, you can select the DAYSIM Header in the "Output Option" of step 11 of 11.

Certainly, running the simulation on single points can save you some time compared to grids with many points - especially with DAYSIM which can take a long time.  You might want to run the simulation in Radiance first to make sure things work fine.  Once the calculations are done, Ecotect will ask you if you want to import the results back in your Ecotect model.

Note that since the results will be imported to objects rather than a grid, you will have to use the Object Attribute (at bottom of the Dispay Setting control panel) controls to manipulate (show/hide, scale, export etc...) your results.

I hope it makes sense and it helps.

Let me know if you have more questions on this topic, otherwise please accept as a solution so that others can benefit from this information.

Cheers,

[arindarrc]

Hello! Thank you kindly for your response. It solves partially my problem. Yet ,I need to populate randomly points for the daylight simulation. In ecotect there ain´t no solution, Im afraid. What would you suggest or how do you proceed then? Kind Regrads from Vienna, Stella

[Pennetier1 · Accepted solution]

Hello Stella, 

You are correct, Ecotect does not offer a randomized surface subdivision option, which would be useful.

However, I have created a quick script - attached - that will do just that.

To use:

1. Create a new zone.  You can call it "Sensor Points".

2. Select the floor surface you want to create the random sensors for.

3. Run the attached script; you firs would need to rename the extension from *.txt to *.scr, then you can simply drop it in your Ecotect canvas or open it with the Script Manager.

4. Tell the script how many points you want generated over your selected floor surface.

5. Tell the script the desired vertical z offset; default value is 762 mm, which is 30" - standard practice in the US.

6. The script will randomly creates the desired number of point objects over your floor surface.  You can use these for your daylight calculation.

I hope that is useful.

Let me know if you have more questions on this topic, otherwise please accept as a solution so that others can benefit from this information.

Cheers,

[arindarrc]

Hello Olivier, Thank you for the script. It works fine until the moment of reading the boundary. From the result I assume it is as if inscribing the geometry in a rectangle regardless of the shape. How do you read it? best,Stella

[Pennetier1 · Accepted solution]

Hi Stella, 

I have created a new script that is much cleaner than the previous one.

It actually uses the existing analysis grid and randomly seeds some grid points to be displayed. This has the advantage of using the analysis grid that can be fitted to any surfaces, and it is cleaner since the analysis grid has specific grid point spacing.

You will have to have an analysis grid fitted on your floor or other surface object first.  The script will ask you what percentage of the total number of grid points you want to be randomly selected.

See attached.  

Let me know how that works for you and if you have more questions on this topic, otherwise please accept as a solution so that others can benefit from this information.

Cheers,

[arindarrc]

Thank you Olivier, for investing your time to help. The script works smooth! best, Niya

[arindarrc]

Hello, The script is working great so far for the grid. I am in the double-shuffle with another issue- I need to import exact points from dxf into Ecotect. I can import their position only with import>model data and then i get grossed lines instead of points. How can I import cleanly points into Ecotect for Sensor Points?(without manually creating them in Ecotect) Kind Regards, Niya

[Pennetier1 · Accepted solution]

Hi Nya, 

Can you send me either a screen shot of the point/lines or the dxf file you are trying to import?

I am wondering if the lines you see in Ecotect are not actually points objects but they are not sized properly after import.

Try this:

On the Display Settings control panel, there is a "Point Size" option.  If you set this value to something very small such as 10 or 20, does that fix your issue?

Let me know, 

Cheers

[arindarrc]

It worked cleanly! Thank you, Pennetier!

[arindarrc]

Hello Penetier, After reimporting the *.DA daylight Factor into Ecotect, the color carpet - values seem to be displaced and not corresponding the geometry. As far as I know, Ecotect expects the results in a separate file without any coordinates, one value per line and then reassignes the values to the calculation grid on the new sequence in the output file. Can I avoid the displacement? How do I verify whether the orientation in Ecotect and Daysim are coherent? How do you solve the problem when you need to see the results from Daysim in Ecotect? Kind Regrads, Stella

[Pennetier1]

Hi Stella, 

Did you change anything in your model before re-importing the *.dat file?

Typically, my process is to save the grid as a *.grd file by going to the Grid management button > Save, after every calculation and data import.

Doing so will keep your grid coordinates and data values.

So if running DAYSIM, once the calculation is done, import the *.dat onto the grid and save it (it is normal if the data value disappear the first time you do so).  You should be able to retrieve it at a later time by going to the File menu > Import > Model/ Analysis Data > look for your *.grd file.

I think you are correct in that the *.dat file might not keep the coordinate, only the values, so you would have to save the grid with the values on it first to be able to re-open it later.

I hope it makes sense!

Let me know if you have more questions on this topic, otherwise please accept as a solution so that others can benefit from this information.

Cheers,

[arindarrc]

Hi Pennetier, Tank you for the insight. This gives me better control over the grid yet does not solve my issue. just something confusing me - you mean the *.DA files, no? not the *. DAT files? The trigger is that the results from Daysim and Ecotect are sometimes displaced and I need to verify the North orientation of both grids. For this reason I thought of tracing back their coordinates, be it not the very elegant solution, though. So i can rotate back the .hea file in Daysim an dimport the backwards rotated points into Ecotect. Tried to open the grid #name.grd yet i get a salat of weird digits -suppose i was not using correct editor. What is the best way to get the point coordinates of the grid than? Maybe the ScriptManager within Ecotect? Kind Regrads, Stella

[arindarrc]

Hello Olivier, The reimport of the points value from Daysim in to Ecotect is giving me empty results - #after reimport the *.da file into the object attributes --> all points have 0 values(DF) I can obtain values for the selected sensor points only for RAD ILLUMINANCE. Could you, please, give e a hint what i might be missing to obtain the results for DF from Daysim or they are equivalent to the RAD ones? Best, Stella

[Pennetier1]

Hello Stella, 

I am having trouble understanding what you are after exactly, but I think I have discover a few things that pertain to your problems...

1. First things first - are you using the latest version of DAYSIM (v4)?  If so, you need to go back to the download page and use the older version ve3, as Ecotect needs the old Java GUI interface.  Not sure that applies to you, but I wanted to throw this out there.

2. You are correct, I meant to say *.DA and not *.DAT file format 

3. Running some test, I was not able to import the *.DA on the point objects either - in fact, Ecotect only allows you to import .*txt file for the Object Attributes.  This is new to me, I never knew.

4. So it looks like you have to use the Analysis Grid to run and re-import your *.DA Daylight Factor file.  I have tried it and it works as expected.  Note that sometimes there is a glitch in ecotect file import whereas after importing the *.DA file, clicking the "OK" button will erase your data.  instead, you ha to simply close the Grid Management window by clicking the "close" cross icon at the top corner.

5. I did not understand what you meant by verifying the North orientation of your grid, rotating your .*hea file, or trying to get the grid point coordinate (which are available as a .*pts file inyour Radiance files folder).  Typically, I do not have to do anything but to import the .*DA file, save it as a *.grd file and I can re-import this file in my model without a glitch.  The saved *.grd file has the coordinate of the grid extents, the point coordinates and their value.

Sorry I cannot be of more help as I do not quite understand what you are trying to achieve.

[arindarrc]

Hello Olivier, 1 Indeed, I am using the older version Daysim 3.1 exactly because of the reason you have nailed down. 3 I regret bothering with this question, yet i^d read the whole forum topics and no-target found.Is it possible to get the DF values only for selected Sensor Points and import their values into Ecotect/for graphic purpose/ at all? ..the analysis grid also with your script works really fine. still, the core issue is to be able and obtain DF for specific sensor points only... 4 i am attaching two images to depict my story - with ecotect and with daysim i get different DF distribution. this leads me to the point that the point sequence from daysim back to ecotect gets a deaf call. I ve read in the forums that this is the case when the north of the scene is different thatn the North per default/y-axis/ and Daysim rotates the scene to its North. So when I reimport the results back to Ecotect, the point count is the same/ no obvious warning/ yet the order gives misleading graphic output. here is a previous comment ive found: https://groups.google.com/forum/#!msg/daysim/p4MHauijK84/H4bI1W_TqnsJ I appreciate any hint on how to solve the problem since this is something that would benefit also others approaching this issue. Best, Niya

[arindarrc]

Hi Olivier, for some reason i cannot attach the image and visualize what i mean? could i email it to you?

[arindarrc]

Hello Olivier, I hope i can post you the image since i have an idea. maybe i can trace back how the points are displaced i f: I assign names to the sensor Points in Ecotect. Could you, please, give me some hint on how to do that or where to fin dsome info? The idea is to have named Sensor Points from Ecotect, which preserve in the name.pts file and I can see track their position back after re-import in Ecotect. Hope I could bring the idea across. Regards, Stella