center point of circle path

[Krikor]

(10-08-2023, 04:03 PM)PixLab Wrote: Alignment tool will do it, it's all natural  

Drag a guide from the ruler on the left and one from the top -> you don't care where you drop them
Now select the Alignment tool, in the tool option select "Relative to Selection" then click on the guide you want to align (2 little white square will appear one on each end of the guide) then just click to the center button and voila..

Patrice, with the advantage of not needing a plugin :-)
The Alignment Tool is much more versatile than it seems.
This reminds me of this post:
https://www.gimp-forum.net/Thread-Align-...t-of-1?pid= 17612#pid17612

(10-08-2023, 04:17 PM)Ofnuts Wrote: This is because the first three points of the ellipse are too close to each other to there are round-off errors (try to do the same with a compass and you will have the same problem).

If you do the same by making a stroke with sufficiently spaced anchors placed on the circle then the crosshair is accurate. As the doc says for the sibling Circumcircle function: "For best accuracy, the three points should be as much spread apart as possible.". Of course circles created by OPTS itself are made of 4 anchors so these are by design sufficiently well spaced (and the crosshair's anchors will be the same as the circle. I could improve the script by trying to find the three most distant anchors.

I have a vague memory of having problems with circles created with the Ellipse Tool in the past. That's when I started using plugins for this purpose.

An example is what happened to this circle seen in the first image. Created with the Elipse Tool and then converted to path, and then again from path to selection.

When the initial selection is still active, the XY values for selection and W/H are correct:
Selection x/y = 274 and W/H = 250
But after path to selection:
Selection:
x=274 ; y 275 / W=251 ; H=250

For the circle created directly for the path, this deviation does not occur, as seen in the second image of my previous post.