In creating and adding a grid for analysis daylighting, the grid could be created from different steps, but the results will be different, which one is correct??? and why are different?
I probably don’t really get what you mean when you say “the grid could be created from different steps”. In all your attached pictures the analysis grid is always the same and you have also created it in the same way. To me, the three results you’re showing look really similar to each other and they seem all correct.
The small difference in the color of the grid’s mashes is due to the fact that every time you run a new simulation a new set of rays depart from the analysis grid and they “bounce in the ambient” for a number of time defined by the -ab parameter, to reach the source of light (the sky and the sun).
This happens in a stochastic way, so every simulation is slightly different from each other due to that and is highly improbable that you get the same exact value at every test point with different runs.
This, however, doesn’t affect the overall value of the results, in fact, if you try to average the values for the entire grid, you will probably see that the difference between the 3 runs is really small after all.
In short, this is called backward-raytracing method and is what Radiance is based on.
I would suggest you to try to make the analysis grid a bit more accurate, switching from 1m grid to 0.5m or 0.2m and I think you will notice less the color difference between different runs.
thank you so much for your useful explanation.
Your explanation was exactly what I needed to know.