Difference in output between old and new DirectSunHours-component

grasshopper ladybug
, ,
1

Hello!

I’ve recently updated the ladybug components and have encountered a problem for post-processing the results. I’m trying to simulate direct sun hours on a large number of windows (around 1500 pieces). I have a post processing system of the results. For example, to interpret the results I have a work flow that calculates the mean direct sun hours per window and a colour is painted over the window depending on a set condition (0-0.5 h=red, 0.5-1h=orange, above 1h=green). The final result looks like this:

direct sun hours example
direct sun hours example520×520 33.3 KB

It seems that the new component has a different data structure. Before, the result had one branch per studied brep. With the new component I get a result list that looks flattened. I have the geometry that goes in to DirectSunHours unflattened. I have tried to have it grafted too but the simulation seems to get stuck then. The simulation normally takes 4 min with a course grid and I have tried with grafted geometry for a whole day but nothing happens.

Is there any way to get back to the “normal” structure? I need to have it separated by window so one branch per window seems to be the best approach to post process the data.

Here is the setup for the two cases:

New:

rundirectsunstudy new component
rundirectsunstudy new component883×687 57.3 KB

Old:

rundirectsunstudy old component
rundirectsunstudy old component1077×498 60.9 KB

Cheers!

2

If you graft the geometry input you get the individual results per object. If it get stucked it can be a memory problem. Test it with a couple of windows and low grid size (or high), to see if that works for you.
-A.

3

Thank you. What I am confused about is that when I use neither flattened or grafted input I get different results for the new and old component but the compution time is similar (around 4 min). When I use grafted geometry it get stuck for both components so that don’t seem to be a good option as I need it to be relatively fast (max half a day or so as I only simulate for 1 day of the year). Is there a way to get the similar structure as the old component without raising the computional time?

4

I have same issue here. Do you find the solution?