Legacy Components vs. V1.1.0 - Different Daylighting Results?

I am trying to switch to the new LBT version, and testing whether I can convert existing definitions to the new components. Here is an example of a daylighting simulation (UDI) where I am getting completely different results when using the new tools.


Legacy vs V1 - Daylighting.gh (550.5 KB)

To make sure I’m using the same RADParameters, I have converted the ones from the legacy component to a formatted string for the new component. I have noticed that there are a few parameters that the new component does not accept so I’ve filtered these out: pt, pj, xScale, yScale, av, ps, and sj.

I believe I have set identical occupancy schedules, namely from 8-18h every day.

In the above case I set the ambient bounces to 2, but the LBT V1 simulation output looks like the ambient bounces are set to 0:

If I set the ambient bounces to 0, the LBT V1 results don’t change; the legacy results do change but there is still a discrepancy between the two:

If I detach the RADParams from both simulations (i.e. go with the defaults), I get this:

FYI this is the UDI legend I am using:

What am I missing? Are there other parameters that I have forgotten about? Am I formatting them wrongly for the LBT V1 component?

I answered this a couple of times already including this one (Setting few ambient parameters crashes HB-Radiance run - #5 by mostapha). It was my poor judgement for setting the default parameters to minimum required and assuming every user adjust it based on their needs.

The main difference is that the LBT recipes are using rcontrib based commands which makes some of the parameters, including --aa, behave differently. See the discussion that I referred to and let me know if there is any other questions.

We are changing the default parameters for the LBT annual-daylight recipe in the upcoming release so it produces better results by default. :grinning:

Hi @mostapha,

based on your post and the github post that you reference in it, I’ve tried some parameter combinations but it always ends up looking something like this:

Here are the variations I’ve tried:


Legacy vs V1 - Daylighting.gh (555.6 KB)

Obviously I haven’t researched what all the parameters mean and this is probably more of a Radiance question than a Honeybee one. Still it would be great to get to the bottom of it, so if you see any obvious errors at first glance please let me know!