Filter Forge - [bug/annoyance] Pixel coordinates (Staff: By design)

Active Topics Search

ThreeDee Lost in Space Posts: 1672 Filters: 112	In theory, this filter should show a pattern of single-pixel dots 4 pixels apart without the Offset component. However, it requires an extra (miniscule) offset to read the correct pixel. Although the Free Rectangle is set to start from 0,0 and end at 1,1, it does not show up at the pixel coordinate 0,0, requiring an offset to push the rectangle over ever so slightly so that it overlaps the 0,0 point. Don't know which Component is causing this, Free Rectangle or Lookup or Pixel X/Pixel Y or the combination thereof, but the combination errs just to the wrong side of the exact coordinate, reading 0,0 point as not being on the rectangle. Pixel Coordinates + Lookup.ffxml
	Posted: July 20, 2015 7:07 am	Details E-Mail
GMM Moderator Filter Forge, Inc Posts: 3491	It took us some time to figure out what's happening here. Your example can be simplified to Free Rectangle plugged into Lookup: the (0,0) and (1,1) points really belong to the background and not to the rectangle. The reason is that the endpoints are excluded from the (0,1) interval. This is not a bug but intended behavior, you just need to keep this in mind when designing filters. Frame (and possibly certain other components) also behave this way. Vladimir said it is possible to change the logic and include the endpoints in the interval, but, frankly, we have more urgent things to implement.
	Posted: July 28, 2015 8:52 am	Details E-Mail

Filter Forge has a thriving, vibrant, knowledgeable user community. Feel free to join us and have fun!

33,711 Registered Users
+18 new in 30 days!

153,531 Posts
+36 new in 30 days!

15,347 Topics
+72 new in year!

20 unregistered users.