I'm wondering about the relative security of some of the multiple-loop knots.
I just discovered the French bowline, which I like better than the bowline on a bight,
because the loop sizes can be adjusted without loosening the knot,
but I suspect that it is not quite as secure. How do those two compare to each other,
the triple bowline, and to the figure 8 variations (double, equalizing)?
Again, security is my primary point of curiosity.
Let me slip in a quick & saner answer than the ping-pong
gamesmenship going on here so far.
1) What you need to test (or find assurances for, somehow) is the
knot tied in the material you expect to use (not the entire spectrum
--or just dispersed waypoints along it-- of knottable media!);
and
2) your rockclimbing concern for dbl.eyeknots is generally NOT
of slack-security (which a shake test checks) but preservation
of eye sizes when only one is loaded, other(s) slack --i.e., the
potential of *feed* of the unloaded eye through the knot.
Rockclimbers favor the
"bunny ears" fig.8 eye knot for having
less vulnerability of slippage in the case of an unloaded eye.
But there are many variations & other knots that might be
employed, which aren't (well) known.
Note that twin-eye knots might be employed, where the
duplicity isn't of the eyes --clipped qua one-- but of the
ends : each end being an independently secured anchor.
--dl*
====