I am currently using the Cyclops Grasshopper plugin in Grasshopper for solar analysis, and I am having difficulty understanding how to correctly set the upper and lower limits when visualizing results.
After running a Cumulative Radiation analysis, I get a range of values (for example around 200 → 1300 kWh/m²). When I directly use these values as the legend bounds, the visualization becomes misleading — most of the geometry appears in a single color (either too red or too blue), and the gradient does not clearly represent differences.
In some example files, I noticed that the upper limit is manually set (for example around 700), which produces a much more readable result. However, I don’t understand:
What is the correct logic behind choosing these limits?
Should the upper limit be based on a percentage of the maximum value?
Is there a recommended method (e.g. percentile-based clipping, normalization, etc.)?
How can I ensure consistency when comparing multiple design options?
In Ladybug Tools, the legend scaling is handled automatically and always produces balanced visualizations, but in Cyclops this step is manual and unclear.
I would really appreciate if someone could explain:
A reliable workflow for setting upper/lower bounds
Or best practices for achieving a clear and comparable radiation diagram
In Cyclops, you shouldn’t use raw min/max values for the legend because outliers distort the gradient. A clearer workflow is to clip extremes—set the lower bound near zero or the 5th percentile, and the upper bound around the 95th percentile—so most values spread across the color scale. For comparing design options, fix consistent bounds (e.g. 0–700 kWh/m²) across all runs; if you only care about relative differences within one model, normalize each dataset separately. This way you avoid misleading visuals and ensure fair, readable comparisons.
So sorry for asking this but this is my first time using cyclops.I wonder that is it okey to fix the upper and lower limits to a certain bounds for comparing design options ? I am asking this because in this way it gives very different visuals and i am not sure that i can able to compare them with the visual results.
You have fix the bounds if you want to compare options. Otherwise the same colour might represent different values for different analyses/options which makes it impossible to compare.
An option is to create a set of domains like 0-100,100-200 and so on, pair each domain with a colour and then check which domain each score sits within. Thereby you can get a stepped visualisation instead of a continuous gradient.
