2 Replies Latest reply: Jan 6, 2010 4:41 AM by Paulo Skylar RSS

    Stroke Edge & Anti-Aliasing

    Shan-Dysigns Community Member

      Maybe this is such a minor issue, and seems to only be noticeable using extreme values, but when I place an inner stroke on an object (ie. square shape), once I surpass the stroke thickness to a certain level, I begin to see adjacent pixels with an anti-aliasing effect. Instead of a stroke consisting of whole pixels, adjacent pixels have a very faint fill. The attached image should clear up any confusion from my description - it's late, I'm tired - I may not be making sense.

       

      In this example, I drew a white square shape (say, 700px x 700px - making sure the corners were snapping to pixels so the edges weren't between pixels), and placed a black stroke beginning with 10px. As the stroke thickness reached 50px, I began to notice neighboring pixels with a faint fill (as stated above). Is there a threshold of some kind with strokes? I know Photoshop needs its outer stroke to be re-vamped because the caps (corners) become beveled.

       

      Does anyone know why Photoshop may be doing this (refer to image)?

       

      Message was edited by: Shan-Dysigns I just noticed something for the first time. If I use the rectangular marquee tool and draw a large square-ish selection (instead of drawing a vector shape), then go to EDIT/STROKE, then set the thickness to 60px and inner, the resulting stroke is perfect (no anti-aliasing) - is that even the correct term to describe it? So, maybe this issue has something to do with how Photoshop defines the edges mathematically when it comes to vector shapes.

        • 1. Re: Stroke Edge & Anti-Aliasing
          Shan-Dysigns Community Member

          Anyone?

          • 2. Re: Stroke Edge & Anti-Aliasing
            Paulo Skylar Community Member

            Yes, there is an unanticipated effect showing up. It is not a simple threshold effect, however, as there are numerous large stroke width values for which the effect is non existent, but increase the stroke width by one pixel + or -  and it happens. The different intensity region is one pixel wide and the intensity may take on several values in an approximately cyclical manner as you step through the stroke width values. My first reaction is that it sounds like and internal calculation effect, but I really cannot account for it.

             

            Paulo

             

            PS: Best guess is that this is yet another artifact of high magnification. Trust appearance only at 100 %.