Scanline rasterization of a large circles seems inefficient

I’m doing a large circle engrave, having used inkscape’s object to path and stroke to path.

When engraving, it starts at the bottom, then goes back and forth. at some point, it starts doing the sides. the issue is that it does the scanline form one side, then the other, then reverses direction and does it again. wondering if anyone knows thy it doesn’t just continue up one side, do the top, then descend down the other side. would save a lot of movement and likely make the engrave go much faster.

I figure there is a good reason, just interested to know what it would be.

You can test this right now by breaking your circle in half vertically and setting the two halves to be different colors, and therefore different engrave jobs. I suspect you’ll end up with a visible seam in the final product though, so ymmv.

My gut says that it’ll take longer to do it this way but I could be wrong. It’ll probably depend a bit on the size of your piece and how “thick” your circular engraved area is. (As in the “walls” of your circle. )

You might be interested in this thread:

To truly optimize this you’d slice the bottom and top off tangentially to the inner path, setting them up as separate jobs, and therefore not duplicating any vertical engrave distance unless 100% necessary.

1 Like

It’s because turning the head around takes so long that it almost always wins to just go all the way across so when software people look at the optimization you propose, they’ll generally be inclined to go with the simpler algorithm that is faster in most cases anyway.


Yeah what @markwal said. That’s what I was referring to about the whole “depends on how fat your circle is” comment. If you’re engraving a really thin circle, you might come out ahead, but I’m suspect. If you give it a whirl, let us know how it turns out!

If it’s really just a circle, a score will be MUCH faster than an engrave. You might need to play with the power settings to get it to just engrave (i.e. minimal depth) instead of score (which cuts 1/2 way through the material).


Also, if you want a wider line than a score, set the focus height higher than it really is. The score will be wider that way. You may have to take the crumb tray out, but there’s plenty of info on the forum about that.