Proof that BEVEL is under performing ?

[ggaliens]

Proof that BEVEL is under performing ?

I've done some work-flow varations that approximate edge-bevel way work-around. Reasonable speeds can be attained on 15,000 polygon models being say ... edge-beveled.

I often do a) All edges bevel b) invert selection to get between faces c) Intrude.

The result is that you get a CAGE mesh from your original model. I'd really like to find the bottleneck with plain old edge bevel setup process. Anyone wants to help or has ideas ... please provide here.

Thanks.