YOUR ACCOUNT

Login or Register to post new topics or replies
Redcap
Redcap

Posts: 1290
Filters: 100
I am making a filter than takes a single picture with a transparent background and makes a wallpaper that is seamlessly tiled. Simply enough I just offset the original image 8 times and blend all the offsets together. However, simply offsetting and blending 8 times takes more time than some of my surfaced filters.

Should it take that long to offset and blend? Aren't those two of the fastest components in FF?



If you are bored check out my unpractical math website
  Details E-Mail
Kraellin
Kraellin

Posts: 12749
Filters: 99
http://www.filterforge.com/forum/read...10&TID=237
If wishes were horses... there'd be a whole lot of horse crap to clean up!

Craig
  Details E-Mail
onyXMaster
Filter Forge, Inc.
Posts: 350
Attaching filter here (or sending a link via contact for or even private developer's e-mail if you wouldn't like to disclose the filter itself) would be a good start to finding out why your specific filter is too slow smile:)
  Details E-Mail
Redcap
Redcap

Posts: 1290
Filters: 100
It is nothing fancy so I really don't mind sharing, just connected a lot of offsets. It is interesting because if it is an external image is it quite a bit slower than if it is just a dot generated in FF. Here it is, thanks.

Seamless 2.ffxml



If you are bored check out my unpractical math website
  Details E-Mail
onyXMaster
Filter Forge, Inc.
Posts: 350
Hmm. The filter you provided renders in about 1 second on my machine with the current publicly available version (and in around 0.6 seconds using the internal performance-tuned build) -- I don't see any performance problems with it.

Maybe you're using a large input image? The rendering time usually increases proportionally to the area of the image (i.e. total pixel count), so if you specify image that is 2x larger than the original on both axes, rendering time will go up 4 times.

Also, are you applying the filter from within Photoshop (or any other host application) or from a standalone application? If it's the host application, I would like to ask you to test it from a standalone application, having the same image as input and tell me the performance difference if any, since the problem might lie in the host-to-FF interaction code, not in the rendering path.
  Details E-Mail

Join Our Community!

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!

153,533 Posts
+31 new in 30 days!

15,348 Topics
+73 new in year!

Create an Account

Online Users Last minute:

32 unregistered users.