I have tied and tested all the possible patterns
with which the nipping and the collar structures of the 8 - 8 loop can be interweaved
--to see which particular patterns lead to loaded knots that retain the initial balanced,
nice form of the unloaded nub more than the others.
Hmmmm, I don't think that you have
I did not say that the loops shown at the oicture were ALL the possible loops !

I had only those 4 vividly coloured ropes to use, so I could nt show all the knots I tied in one picture...There were 4 more, where the returning eye leg, as it approaches the tip of the eye, goes at the one side, "over" or "under" itself, at it is approaches the Standing end - tied on the loops where the nipping structure is not "regularly" interweaved ( i.e., as in the loop made on the yellow rope ). The "blue" and the "red" knots are "one-side" colour structure variations of the "regular", "over" / "under" nipping structure, tied as in the "white" knot. There are 4 more such variations, tied on the "non-regular", "one-side" nipping structure tied as in the "yellow" knot.
However, I am glad you noticed it. It proves that you can, if you wish !
Tie and try all the possible variations by yourself, and tell me, please, if my general conclusion is correct : that,
oddly, it is the weaving pattern of the collar structure that matters more than the pattern of the nipping structure ). And if, provided that the collar structure is "regularly" weaved,
all possible knots, regardless of the particular weaving pattern of their nipping structure, lead to well formed, nice knots. So, the knot tyer should only pay attention to one thing, the "regular" collar structure s braid, and to nothing else.
(I might have long ago fiddled some end-2-end knot with a like structure, both ends "twisting" in their interlocking.)
You mean, at (1) ? Yes, I know it, but I am sure that this "regular" pattern, which, as I said, is the
minimum possible piece of textile, should have also been tied and tried
thousands of years ago !
1)
http://igkt.net/sm/index.php?topic=2694.msg17058#msg17058 P.S. I had edited my previous post, to make this clear :
P.S. I should have mentioned that the possible combinations are 12 - I had not included all of them in the (second) picture, which was not meant to be all-inclusive . The nipping structure can be "regularly" ("over"/"under"), or "non-regularly" ("one side") interweaved ( two general classes, represented by the "white" and the "yellow" knots ). Then, the collar structure can be also "regularly" ("over "/"under") interweaved, or "non-regularly" "one side"- and this side be the "upper" one, or "non-regularly "one-side"- and this side be the "lower" one. So, 2 x ( 2 + 2 + 2) = 12. One may well ask : What happens with the duplication of the total number of combinations, according to if the returning eye leg, as the collar structure, enters into the nipping structure from the one or the other side of its "lower" opening ? I believe those cases are mirror-symmetric to cases already counted, because, due to the number of openings/nodes of the nipping structure (two), the tail will always exit from it from the opposite side through which it had entered into it.