This is the first time I come across this issue. I've completed an annual UDI calculation by using Diva 4 in Rhino 6. To my surprise, the output of 300-3000lux showing negative percentage (which is not possible). My first response was that I set all the windows wrongly as opaque material. No, it is actually not the reason. When I checked other ranges, <100, 100-300, >3000, I notice they all fulfill my expectation and reflect the daylight conditions correctly. Only 300-3000lux has the abnormal output with negative values. Comments? Is it a bug or my modeling mistake?
update: I re-run the model and get the correct outputs.
Something I learned:
The misleading results are from the three models running in parallel as
Although I manage to avoid using the similar names for parallel calculation, the above names still trigger mistake in Radiance calculation. The right outputs are from the three models with totally different names such as baseline, apple and banana. I am not very sure it is the reason but I will not use similar names for future parallel calculations.