BSDF-material error "unexpected row count header" by 5-phase analysis



Hallo everybody,

I hope someone could help me solve a problem I am incurring into by preparing BSDF materials for a 5-phase analysis to test some blind geometry on a simple volume before applying them to a real case model.

To define the BSDF material I used both WINDOW 7.4 and 7.6 (wasn’t sure it was a compatibility problem) and followed a tutorial to generate a BSDF xml file for a particular window and shading system I developed within the programs. WINDOW generates the xml file without problems, and compared with the sample files attached in the forum by @mostapha they look similar (of course more layers in the first), at least for the organization of the data through the different headers.

You can find below a very simple BSDF ‘double glass clear’ definition produced with WINDOW that can be used in any 5-phase sample file such as the ones (11 and 12) attached as reference in the forum.
Double Clear Air.xml (138.2 KB)

You can give it a try by just replacing the ‘double glass clear.xml’ file above with the ‘clear.xml’ in the reference case 11_five_phase_I_gridbased of the sample files attached in the forum and here below again.

(thanks to @mostapha and the Ladybug tools team for uploading these very useful examples)

When I try to use this to generate a .rad material to feed into the window group component, there is no problem and the xml file is correctly read by the window group component.

However, the simulation of the 5-phase state output the following error:

I understand there is a problem with the distribution of data in the header. Apart from the fact that the reference files are created with Radiance genBSDF.exe and mine with WINDOW, contents look completely the same. I also tried to reconstruct the .xml file by importing the data of the double clear glass system in the ‘clear.xml’, but to no avail. A solution would be to create geometries in GH, then export the HB surfaces to RAD file and use genBSDF.exe to create a file which could work. Problem is, my Radiance installation is missing genBSDF.exe and I tried to install also a more recent version to check if its was a problem of my built, but again, no luck at all.

Therefore if someone would have any idea or hint how to solve this problem it would be much appreciated. I can still run the simulation with the sample files “clear.xml” and “diffuse50.xml”, but i would also like to study different and more complex blind systems, but until this problem is not resolved I cannot go on.

Many thanks indeed for any help!



This was a limitation of Honeybee[+] which has been addressed before. If you update your Honeybee[+] installation the simulation should run successfully.

Also unless you use high resolution BSDF for the 5th Phase, which is not currently supported in Honeybee[+] via the interface but it will be supported by the end of August, you better off using the daylight coefficient recipe instead of 5 phase.


Hey @mostapha,

Sorry for my late reply to your always immediate response to the different topics. I am really thankful for your prompt action. Unfortunately I really missed that particular topic, I don’t really know why it didn’t pop up when I searched for the “row count header” error when it was clearly addressed in the thread you attached.

As you suggested, the update of the ladybug tools worked like a charm and BSDF files could be loaded both in the 5-phase and daylight coefficient simulations.

Thanks very much indeed again!