Roll20 uses cookies to improve your experience on our site. Cookies enable you to enjoy certain features, social sharing functionality, and tailor message and display ads to your interests on our site and others. They also help us understand how our site is being used. By continuing to use our site, you consent to our use of cookies. Update your cookie preferences .
×
Create a free account

Defect completing the initial corner of a rectangle with the drawing tool

When I use the draw tool with held down to draw a rectangle, with the line width set to any of the greater widths, the tool does a neat job of squaring the exterior corners of any vertices I make, but when I get back to the starting point to close the figure it leave out a few pixels on the outer corner.
I noticed that too. Picture attached for the lazy who want to see what we're talking about.
1338651481
Deightine
KS Backer
Sheet Author
Just played with it some myself... looks like when you shift the cursor to create a new line, the drawing process starts the new line at the edge of the terminated line, but then ends it at the grid point it is snapped to. It always starts and ends exactly even with the grid, so the tool is doing what is intended... the issue is that the rendered line that comes up around the points in the vector drawing isn't compensating for the square of the grid that isn't actually overlapped by the render. In the drawing above, the top-left square out of the 9 squares interacting with the art. I think most vector art programs compensate for this by having the lines all bind together into a single vector object when a drawn point lands on the starting point, forcing it to re-render the whole thing, and eliminating any holes or awkward angles in the process. Essentially, you're drawing lines until they come together, at which point you've drawn a polygonal object, at least for the sake of the rendering engine. In the way the tool is working here right now, you've just been drawing lines.
Thanks for the report, it has been noted.
The obvious work-around is of course to start in the middle of a side instead of at a corner. I wish I had thought of that earlier.
I put in the suggestion to join vectors with end points that "share" a location