So, I agree with Compagnon (2004) on the method he used to calculate SVF: a simple radiation (or illuminance) simulation using a uniform sky. Therefore, I always refer to the following 'theoretical' Sky View Factors calculated at the centre of an infinitely long street canyon with different Height-to-width ratios in Oke's original paper (1981) as the ultimate benchmark to validate different methods to calculate SVF: Sky View Factor calculation shall be based on the first principle defining the concept: radiation exchange between urban surface and sky hemisphere: So, I think using 'geometry way' to approximate Sky View Factor is not correct. It will be affected by the proportion of sky visible from a given calculation point on a surface (vertical or horizontal) as a result of the obstruction of urban geometry, but it is not entirely associated with the solid angle subtended by the visible sky patch/patches. Sky View Factor is first and foremost defined from the discussion of radiation exchange between urban surfaces and the sky in urban heat island research (See Oke's literature list below). I'd like to share and discuss with you my understanding of Sky View Factor, considering that it is an import concept frequently misunderstood.

Make visuals great again sky