When I render a room on its own, the windows show up correctly, but when I render the whole building the windows don’t show up.
When I export the whole building as a gem and look at it in IES, the windows are not present either.
When I deselect the roof and render all rooms, the windows show up again.
Is this a bug or am I making any mistakes?
This only seems to happen for a few rooms where I messed with the geometry.
windows showing up when room is rendered by itself
Thank you for documenting the issue carefully. As usual can you share the PoMF with us to try to recreate the issue on our end? This might be caused by a bug that we fixed last week but hasn’t been pushed to production.
It looks like there’s something about your roofs that is either intersecting the window geometry or otherwise interfering with the windows in some way. I would recommend running the preview roofs command just to check that there isn’t something obviously wrong with the roof geometry that you have.
And, as @mostapha says, I can investigate if you can share the POMF file.
I could recreate the issue on my end and it is related to the calculation of the room volume with the roof but it’s a bug that only affects the case that the room has duplicated vertices in its floor polygon. So the only thing that you need to do to stop the error from happening is run the Join Coplanar Faces command on the problematic rooms. That will automatically remove any duplicated vertices in the floor polygon for you.
Once I do that I no longer have any issues with the windows not showing up with the roofs:
I’ll try to find a more elegant way to address in our code this since we don’t consider duplicate vertices like this invalid and the windows should show up even when you have them. But, hopefully, the recommendation above gives you what you need to unblock your current model cleanup.
Actually it was pretty easy for me to just add an extra check that just removes the duplicate vertices whenever there’s a roof that influences the room volume calculation:
So that fix will be in the next release of the model editor and it should ensure that this situation is impossible to experience going forward.
Hi @chriswmackey, thank you so much. Your advice helped me figure it out. However, I want to share some weird glitches that surfaced along the way: I had to run “delete short segments” first to run the “join coplanar faces” command. Otherwise a weird ripping through the room would happen.
Image of a room that is ripped after running “join coplanar faces” without running “delete short segements”
!! However, running both commands in a row would not always keep the previous windows intact. It’s hard to describe what I did to make things work with the room in the image. I essentially found where dots where overlaying, pulled one of the dots out of the overlaying layers. Then pressed “delete short segments”, which deleted the extra wall coming out slightly. That for some reason, didn’t break the windows. Then pressing “delete coplanar faces” also worked without breaking the windows. and the whole situation was fixed.
Your tip about coplanar face deletion definitely helped me to resolve the issue. But I couldn’t just press the button. I had to change the geometry, ensure that windows don’t get deleted every step along the way, and see what random form of changes allowed me to run the “delete coplanar faces” feature.
There seems to be some additional complication going on. Maybe that makes sense to you. I hope this helps or makes any sense.
Just sharing this in case you want / need debugging insights.
I think a lot of these questions might have an answer by understanding what version of the model editor you are using.
For this:
I think you are probably experiencing this bug that I fixed last week:
If so, the windows aren’t necessarily gone but they are assigned to the wrong wall after running “Remove Short Segments”. I also could not recreate the issue with the room becoming “ripped” in the latest development version of the model editor and the version of the model that you shared with us via PM. So I am assuming that the issue there has also been fixed.
@mostapha should be able to let you know when there is an official release with the fix.
