IM 6.7.1 Q16 / Win XP / convert -layers OptimizePlus

[jverne]

The frame doubling optimization for animated GIFs doesn't seem to be working on IM 6.7.1 Q16 / Win XP. For instance, when I try to use:

Code: Select all

convert moving_hole.gif  -layers OptimizePlus   moving_hole_oplus.gif

on the image moving_hole.gif, the image is here:

http://www.imagemagick.org/Usage/anim_o ... g_hole.gif

no optimization is produced.

The following code doesn't seem to be working also:

Code: Select all

  convert moving_hole.gif  \( -clone 0--1 -set delay 0 \) \
          +delete -insert 2 -insert 1 -insert 0 \
          -layers OptimizeFrame  moving_hole_double.gif

The following code "does":

Code: Select all

  convert moving_hole.gif  \( -clone 0--1 -set delay 0 \) \
          +delete -insert 2 -insert 1 -insert 0 \
          -layers Optimize  moving_hole_double.gif

but the optimized image is not as small as the one on the manual:

http://www.imagemagick.org/Usage/anim_opt/

Am I missing some kind of configuration?

[anthony]

It should be working... But you are right it is no longer working for linux either.

The question then becomes what changed to break it!

I have a feeling the lower level test functions are breaking, rather that optimizeplus itself.

Shame I don't have time to look into it right now, but I'll put it on my list.

[jverne]

Optimizeframe doesn't seem to be working as advertised either, as the following command:

Code: Select all

convert  canvas_prev.gif -coalesce  -layers OptimizeFrame  optframe.gif

for the example provided at:

http://www.imagemagick.org/Usage/anim_opt/

produces a smaller image, but not as small as optframe.gif shown in the previous link.

Could it be that OptimizeFrame is broken?

[anthony]

Internally they are the same function. Just a flag is passed for a extra optimization test.

OptimizePlus tests to see if it can reduce the total number of pixels by separating frame deletion from the next frame overlay, as discussed on the IM examples page.

[jverne]

What is the status of this bug?