2020 Comparing Lidar Surface Creation Techniques

Hello everyone,

I am experimenting with different surface creation methods from lidar data and was noticing a few differences that I could not find the answers for through my own googling.

Method 1 - convert lidar (.las) to bare earth (.xyz) file and create point cloud in Recap, then create surface within Civil 3d from that point cloud (no filtering)
Method 2 - take same (.xyz) file and create surface from point file under Surface definitions

I used some large data sets so maybe differences would stick out a little more.

Method 1 resulted in a file size of 131,168 kb compared to Method 2 of 85,722 kb
Number of points the same (expected)
max and min and mean elevations all the same
Method 1 2d/3d area - 23,342,940.16 ft^2/23,468,623.94 ft^2 vs Method 2 23,341,991.45 ft^2/23,467,085.66 ft^2 - close but different, i imagine if i created them again C3D might solve the triangle slightly different resulting in another number?
Number of Triangles - method 1 = 3,237,049 vs method 2 = 3,237,010

Biggest difference seems to be just the size of the drawing which is pretty significant.

Anyone have an explanation? Or recommend maybe a better process for creating surfaces wit lidar? I am fairly new to this and would like to see what the rest of the industry really does.

Thanks