Custom Spanish Roof Tile Hatch
That looks to me like a case of X- and Y-offset values being figured relative to zero rotation, instead of relative to the direction of each lineset separately for itself. It might be possible to troubleshoot if you can post the .DWG file that I assume is the source of your "correct" image.
And I kind of wonder whether it could be simplified somewhat without losing too much of the "curvature" appearance. You have linesets having offset values like this:
....,1161.742879,0.00082279,....
with the X offset being more than 1,400,000 times the Y offset, and/or ending with this kind of pair:
....,0.04657701,-1808.6968728
with the pen up for close to 39,000 times the distance that it's down, both of which I would think could be simplified to something a little less extreme, with some slight changes in direction angles.
EDIT: I managed to get the pattern to work in a limited area and was able to make my own basis. BUT it raises other questions, just within the area where it is close to working right:
I notice that the "under" tiles are at different levels in relation to the rest of the pattern [the red ones are higher than the yellow ones]. And the "over" tiles are not consistent, either -- the green one is higher than those I left white, and the magenta one should be doubled like the cyan ones, with the second one there presumably above the magenta one, meaning the close-to-vertical lines going upward from there should be shorter. Etc., etc. Maybe that's part of the reason the .PAT file has 235 lines....
Also, at a scale of one, these are the "cycle" dimensions in drawing units:
Is that a real size in meters, perhaps? If not, how do you decide on a Hatch pattern scale?
Can you make a .DWG version that is consistent about all this in the way that you want?
@caoelizabeth1 wrote:
....When zoom in, they don't connect ....
Is it better if you set the Hatch's origin somewhere within it or very close to it? That kind of "wander" is usually a matter of getting farther from the origin.
