If I understand jitter angle setting, it causes loops to start at different places to avoid seams forming?
Correct.
Any travel over 5mm should trigger a retraction, and it nearly always does. However, there will always be one travel move that will not trigger a retraction, even though it's over a long distance. In the pic of the attached parts, it's a move from one corner to the opposite corner. No retraction, so there's almost always a blob tree forming there. I can't think of any logical reason why it wouldn't retract for that move, it retracts on all other moves over 5mm.
Which version of KS are you running? Here's what used to happen:
If the long travel is all "inside" the part (does not cross the perimeter) then KISSlicer will not trigger a destring operation. In the latest Beta on the forums, you can now specify a "Trigger" distance...any jump longer than this will do a destring retract / prime, even if the whole move is inside the part.
Jonathan
Sounds like the Trigger isn't influencing the Perimeter Crossing code.
I set jitter to 45 degrees for this print of 5 of these parts. I didn't see it starting loops at different positions on different layers though, it always started on one corner.
Must be a later one or Jitter wouldn't be present. How are the parts grouped on the bed, one STL with multiples set in KS or a gang STL?
EDIT: Just tried a test with the Y Bearing Mount that I repaired for karabas yesterday:
http://kisslicer.com/forum/index.php?topic=819.msg6125#msg6125and grouping by either method makes no difference. With Jitter set to 360, they all start in the same random positions, as if in synchronized swimming. However, this part has thin, high aspect ratio sections in the upper levels. Once the layers reach this altitude and the aspect ratio passes some threshold, the start point
only appears at alternating corners around the "circle". That you have Jitter set to 45 will make it select one corner every time.
EDIT II: Just took a longer look at the slices to see where the aspect ratio comes into the mix and see that, if there are no curved Perimeters, Jitter selects corners--no matter what. I thought that the answer to this was in two Alphas that are here on the forum:
http://kisslicer.com/forum/index.php?topic=754.msg5661#msg5661http://kisslicer.com/forum/index.php?topic=759.msg5732#msg5732Win32, with other OS versions farther down in the 2nd thread. Beware that some other stuff may be broken in these, so use at your own risk.
BUT Jitter still only selects corners on non-quasi-circular Perimeter sections. No matter what the Layer Start angle is set to, only non-straight paths show intermediate start points. This is due to it selecting a random circle segment as a start point. (Circles are chains of small straight paths.) On rectilinear paths, there is only one start point per side, so there's nothing else to select. On a rectangle, all you get are four changes of direction. With 360 deg Jitter that translates to four random start points. With 45 deg Jitter; one.
