Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Gray artifacts in large gif when using -layers optimize #57

Closed
wheany opened this issue Nov 20, 2015 · 15 comments
Closed

Gray artifacts in large gif when using -layers optimize #57

wheany opened this issue Nov 20, 2015 · 15 comments

Comments

@wheany
Copy link

wheany commented Nov 20, 2015

When creating a large .gif, using -layers optimize causes gray artifacts to appear all over the image.
The source frames I'm using in these commands are in this zip file: http://www26.zippyshare.com/v/rOwQZL3D/file.html and they are originally from this cartoon: https://youtu.be/_2WMLV1UeuM

The version I'm using is the newest available atm, but this bug was also present in some 6.8 version I was using before I updated:

$ convert -version
Version: ImageMagick 6.9.2-6 Q16 x64 2015-11-14 http://www.imagemagick.org
Copyright: Copyright (C) 1999-2015 ImageMagick Studio LLC
License: http://www.imagemagick.org/script/license.php
Visual C++: 180040629
Features: Cipher DPC HDRI Modules OpenMP
Delegates (built-in): bzlib cairo freetype jng jp2 jpeg lcms lqr openexr pangocairo png ps rsvg tiff webp xml zlib

I'm using Windows 7 and cygwin terminal to run these commands, they won't work as-is with the default windows command prompt. I haven't tested if the problem happens in Linux as well.

This command works properly and creates a 1,72 MB gif with no artifacts:
convert -delay 4 frames/mickey_mustache0{08..74}.png -colorspace rgb -resize 50% -colorspace srgb -quantize lab -fuzz 3% -layers optimize +dither mickey_mustache_small.gif

This command demonstrates the error:
convert -delay 4 frames/mickey_mustache0{08..74}.png -colorspace rgb -resize 94% -colorspace srgb -quantize lab -fuzz 3% -layers optimize +dither mickey_mustache_big.gif

The only difference between those two is the amount of scaling from the 1080p source files (50% vs 94%)

If you don't use -layers optimize, you get a very big, but working, gif:
convert -delay 4 frames/mickey_mustache0{08..74}.png -colorspace rgb -resize 94% -colorspace srgb -quantize lab -fuzz 3% +dither mickey_mustache_big_no-optimize.gif

If you don't use fuzz, you get a fairly big, but extremely corrupted gif:
convert -delay 4 frames/mickey_mustache0{08..74}.png -colorspace rgb -resize 94% -colorspace srgb -quantize lab -layers optimize +dither mickey_mustache_big_no-fuzz.gif

Here is a zip of the resulting gifs: http://www62.zippyshare.com/v/CD2stFbG/file.html

@dlemstra
Copy link
Member

This is not a bug in -layers optimize but an issue with your command order. You should put the +dither before your optimize your layers. We get the correct result with the following command:

convert -delay 4 frames/mickey_mustache0{08..74}.png -colorspace rgb -resize 94% -colorspace srgb -quantize lab -fuzz 3% +dither -layers optimize mickey_mustache_big.gif

@wheany
Copy link
Author

wheany commented Nov 21, 2015

Confirmed.

Is there an explanation for why it works the way it does, because I'm confused why it works with the first command and doesn't with the second?

I have seen this before but I don't think I've ever tried moving the +dither command as a fix.

@fmw42
Copy link

fmw42 commented Nov 21, 2015

In IM 6, settings (+dither) should come before operators (such as -colors); otherwise, the operator does not know what setting to use. see http://www.imagemagick.org/Usage/quantize/#colors. Command lines are mostly parsed in sequence. See http://www.imagemagick.org/Usage/basics/#syntax

However, I am puzzled, since I do not see any operator in your command that would be affected by: -quantize, -fuzz or +dither. I am not aware that -layers optimize is sensitive to any of those

@wheany
Copy link
Author

wheany commented Nov 21, 2015

I just threw that there because I though that only the color reduction to 256 would be affected by dithering. I haven't had a "set" place for the setting in my gif making before, but I guess I should just put it in the front like with the -delay setting.

Then again I just tried leaving the +dither out entirely, and then explicitly specifying -dither FloydSteinberg as the first option, before the -delay, and I'm getting gray artifacts in both cases.
convert -dither FloydSteinberg -delay 4 frames/mickey_mustache0{08..74}.png -colorspace rgb -resize 89% -colorspace srgb -quantize lab -fuzz 3% -layers optimize mickey_mustache_big_dither.gif

@fmw42
Copy link

fmw42 commented Nov 21, 2015

I am totally puzzled as I said above, since I do not know what operator in your command line is dither sensitive.

What do you get when you leave all that off? Is it the same as putting +dither at the beginning?

convert -delay 4 frames/mickey_mustache0{08..74}.png -colorspace rgb -resize 89% -colorspace srgb -layers optimize mickey_mustache_big_dither.gif

What version of IM are you using?

@wheany
Copy link
Author

wheany commented Nov 21, 2015

the command was convert -delay 4 frames/mickey_mustache0{08..74}.png -colorspace rgb -resize 89% -colorspace srgb -quantize lab -fuzz 3% -layers optimize mickey_mustache_big_dither.gif
Version is 6.9.2-6 Q16 x64 2015-11-14, the full version output is in the first message.

@fmw42
Copy link

fmw42 commented Nov 21, 2015

I realize that. But I do not see why you have any need for -quantize, -fuzz and +dither. There is no operator there that should be affected to my knowledge. So I wanted to see if you got a decent result or any different result from what you had before with +dither at the beginning. So I asked you to run

convert -delay 4 frames/mickey_mustache0{08..74}.png -colorspace rgb -resize 89% -colorspace srgb -layers optimize mickey_mustache_big_dither.gif

and there report back any difference.

@dlemstra
Copy link
Member

The -layers optimize option will do a color quantization as the last step and that is why +dither affects this. And it seems you get the best result with dithering disabled.

@wheany
Copy link
Author

wheany commented Nov 21, 2015

@fmw42 Running convert -delay 4 frames/mickey_mustache0{08..74}.png -colorspace rgb -resize 89% -colorspace srgb -layers optimize mickey_mustache_big_dither.gif causes artifacts.

I use -quantize lab because afaik it treats perceptually similar lighting and color change equally, so it's "the fairest" quantization method. Usually I use ordered dither in the lab colorspace when making gifs, but with some cartoons I've found that it's better to just use quantization.

Also running this command (without -fuzz) balloons the file size from 3.99MB (which is just under my maximum file size of 4MB) to 15.6MB, because every tiny color change in every pixel is treated as a meaningful change. With cartoony source like this, I can sometimes use up to 4% or 5% fuzz without a significant drop in the visual quality.

@fmw42
Copy link

fmw42 commented Nov 22, 2015

Sorry, I did not realize that -layers optimize does a color quatization, as dlemstra points out above. Thus dither, fuzz, quantize would be needed prior to -layers optimize.

@dlemstra
Copy link
Member

Do you still have some questions or can we close your issue?

@wheany
Copy link
Author

wheany commented Nov 29, 2015

I'm confused about your question. Is the gray artifacting intended behaviour? Especially in this case I added earlier: convert -dither FloydSteinberg -delay 4 frames/mickey_mustache0{08..74}.png -colorspace rgb -resize 89% -colorspace srgb -quantize lab -fuzz 3% -layers optimize mickey_mustache_big_dither.gif

This is the resulting gif, btw: https://i.imgur.com/E93Y4Xl.gif

If gifs made with that command like are supposed to have terrible gray smears all over, then I guess this issue can be closed, but I intended this to be a bug report, because that seems like a bug to me.

@wheany
Copy link
Author

wheany commented Nov 30, 2015

It is somehow related to the lenght of the gif, if I only use the first 19 frames, i.e.
convert -dither FloydSteinberg -delay 4 frames/mickey_mustache0{08..27}.png -colorspace rgb -resize 89% -colorspace srgb -quantize lab -fuzz 3% -layers optimize mickey_mustache_big_short27.gif
the artifacts are barely visible with the naked eye. Areas of the image with motion have a very slight grayish color to them.

Could this be caused by some overflow happening in the quantization code? There is a "big number" of pixels. Adding their numbers to the histogram could fail. I haven't looked at Imagemagick's quantization code, but I assume a histogram is built at some point of it.

@urban-warrior
Copy link
Member

Try your command with ImageMagick 6.9.2-8.

@wheany
Copy link
Author

wheany commented Dec 6, 2015

Hey, it works! No more artifacts.

@dlemstra dlemstra closed this as completed Dec 7, 2015
jsonn pushed a commit to jsonn/pkgsrc that referenced this issue Apr 29, 2016
2016-03-22  6.9.3-8 Cristy  <quetzlzacatenango@image...>
  * Respect gravity when rendering text (e.g. convert -gravity center my.txt)
  * Return empty string for %d property and no directory (reference
    https://www.imagemagick.org/discourse-server/viewtopic.php?f=3&t=29365)
  * Return filename for the %i property (reference
    https://www.imagemagick.org/discourse-server/viewtopic.php?f=2&t=29364).
  * Fixed lost pixels in frequency space (reference
    https://www.imagemagick.org/discourse-server/viewtopic.php?f=3&t=29374).

2016-03-06  6.9.3-7 Cristy  <quetzlzacatenango@image...>
  * New version 6.9.3-7, GIT revision 10628:810a9fe:20160306

2016-03-05  6.9.3-7 Dirk Lemstra <dirk@lem.....org>
  * -size can now be used the specify the maximum width/height for the
    following vector formats:
      * SVG (when librsvg and cairo are being used).
      * EMF (did not work on Windows).

2016-02-29  6.9.3-7 Cristy  <quetzlzacatenango@image...>
  * Fixed text alignment issue for label: / coder: (reference
    https://www.imagemagick.org/discourse-server/viewtopic.php?f=3&t=29223).

2016-02-27  6.9.3-6 Cristy  <quetzlzacatenango@image...>
  * New version 6.9.3-6, GIT revision 10597:83584a1:20160228.

2016-02-25  6.9.3-6 Cristy  <quetzlzacatenango@image...>
  * Ensure TXT: image format survives round-trip (reference
    https://www.imagemagick.org/discourse-server/viewtopic.php?f=3&t=29206).

2016-02-21  6.9.2-6 Dirk Lemstra <dirk@lem.....org>
  * Added wavelet denoise.

2016-02-20  6.9.3-5 Cristy  <quetzlzacatenango@image...>
  * New version 6.9.3-5, GIT revision 10572:cb24a69:20160219.

2016-02-16  6.9.3-5 Cristy  <quetzlzacatenango@image...>
  * Too many bytes when writing raw image formats (reference
    ImageMagick/ImageMagick#132).

2016-02-12  6.9.3-4 Cristy  <quetzlzacatenango@image...>
  * New version 6.9.3-4, GIT revision 10560:09f5f5d:20160212.

2016-02-12  6.9.3-4 Dirk Lemstra <dirk@lem.....org>
  * Fixed memory leak when reading an SVG without width/height (reference
    http://www.imagemagick.org/discourse-server/viewtopic.php?f=27&p=130197).

2016-02-09  6.9.3-4 Cristy  <quetzlzacatenango@image...>
  * Fix compiler error for complext text layout.

2016-02-05  6.9.3-3 Cristy  <quetzlzacatenango@image...>
  * New version 6.9.3-3, GIT revision 10540:6b4c918:20160206.

2016-02-04  6.9.3-3 Cristy  <quetzlzacatenango@image...>
  * Numerous bugs fixed from fuzzed images (reference
    ImageMagick/ImageMagick#123).
  * The -bias option is no longer a noop (reference
    ImageMagick/ImageMagick#125).

2016-01-25  6.9.3-3 Dirk Lemstra <dirk@lem.....org>
  * Read XCF layer name (reference
    http://www.imagemagick.org/discourse-server/viewtopic.php?f=3&p=129638).

2016-01-23  6.9.3-2 Cristy  <quetzlzacatenango@image...>
  * New version 6.9.3-2, GIT revision 10496:315028c:20160122.

2016-01-17  6.9.3-2 Cristy  <quetzlzacatenango@image...>
  * Don't reuse lock after free (reference
    http://www.imagemagick.org/discourse-server/viewtopic.php?f=3&t=29031).

2016-01-21  6.9.3-2 Dirk Lemstra <dirk@lem.....org>
  * Fixed writing DocClass in CALS coder (reference
    http://www.imagemagick.org/discourse-server/viewtopic.php?f=3&p=129478).

2016-01-17  6.9.3-1 Cristy  <quetzlzacatenango@image...>
  * New version 6.9.3-1, GIT revision 10488:e4f046e:20160116.

2016-01-06  6.9.3-1 Cristy  <quetzlzacatenango@image...>
  * Pair AcquireMagickResource with calls to RelinquishMagickResource when
    resource limits are hit (reference
    ImageMagick/ImageMagick#75).
  * Return non-zero for the -format %b option (reference
    http://www.imagemagick.org/discourse-server/viewtopic.php?f=1&t=28969).

2016-01-01  6.9.3-0 Cristy  <quetzlzacatenango@image...>
  * New version 6.9.3-0, GIT revision 10445:620328b:20160101.

2015-12-31  6.9.3-0 Cristy  <quetzlzacatenango@image...>
  * Don't break on euro-style numbers (reference
    http://www.imagemagick.org/discourse-server/viewtopic.php?f=1&t=28910).

2015-12-22  6.9.3-0 Glenn Randers-Pehrson <glennrp@image...>
  * 16-bit pnm images have a max value of 65535.

2015-12-18  6.9.2-10 Cristy  <quetzlzacatenango@image...>
  * New version 6.9.2-10, GIT revision 10415:e31c09d:20151219.

2015-12-19  6.9.2-10 Cristy  <quetzlzacatenango@image...>
  * Fixed compile error when POSIX threads are not defined.

    http://www.imagemagick.org/discourse-server/viewtopic.php?f=3&t=28791).
2015-12-18  6.9.2-9 Cristy  <quetzlzacatenango@image...>
  * New version 6.9.2-9, GIT revision 10402:0027295:20151219

2015-12-05  6.9.2-9 Dirk Lemstra <dirk@lem.....org>
  * Fixed memory leak when reading incorrect PSD files (reference
    http://www.imagemagick.org/discourse-server/viewtopic.php?f=3&t=28791).

2015-12-05  6.9.2-9 Cristy  <quetzlzacatenango@image...>
  * Enhance PDF to properly handle unicode titles (reference
    http://www.imagemagick.org/discourse-server/viewtopic.php?f=3&t=28786).
  * Fix memory leak in icon coder (reference
    http://www.imagemagick.org/discourse-server/viewtopic.php?f=6&t=28848).

2015-12-05  6.9.2-8 Cristy  <quetzlzacatenango@image...>
  * New version 6.9.2-8, GIT revision 10374:ac7d144:20151205.

2015-12-04  6.9.2-8 Cristy  <quetzlzacatenango@image...>
  * Gray artifacts in large gif when using -layers optimize (reference
    ImageMagick/ImageMagick#57).

2015-11-26  6.9.2-7 Cristy  <quetzlzacatenango@image...>
  * New version 6.9.2-7, GIT revision 10370:eeb7d4e:20151126.

2015-11-24  6.9.2-7 Dirk Lemstra <dirk@lem.....org>
  * The DICOM reader now handles the rescale intercept and slope.
  * Added 'bmp3:alpha' option for including the alpha channel when writing
    an image in the BMP3 format. (reference
    http://www.imagemagick.org/discourse-server/viewtopic.php?f=1&t=28733).
  * Only one thread permitted when delegating to the Ghostscript library
    under Windows.

2015-11-14  6.9.2-6 Cristy  <quetzlzacatenango@image...>
  * New version 6.9.2-6, GIT revision 10341:f408196:20151113.

2015-11-08  6.9.2-6 Cristy  <quetzlzacatenango@image...>
  * PixelColor off by one on i386 (reference
    ImageMagick/ImageMagick#54).
  * Eliminate stack overflow (reference
    http://www.imagemagick.org/discourse-server/viewtopic.php?f=1&t=28658).

2015-11-03  6.9.2-6 Dirk Lemstra <dirk@lem.....org>
  * Added local contrast enhancement.
  * Fixed bug in SetPixelCacheExtent that made images all black (reference
    http://www.imagemagick.org/discourse-server/viewtopic.php?f=3&t=28514).
  * Added support for reading DICOM images with implicit vr little-endian
    transfer syntax.

2015-11-02  6.9.2-6 Samuel Thibault <samuel.thibault@...org>
  * Added 6dot variant for unicode and iso braille formats (reference
    http://www.imagemagick.org/discourse-server/viewtopic.php?f=2&p=127111).

2015-10-31  6.9.2-5 Cristy  <quetzlzacatenango@image...>
  * New version 6.9.2-5, GIT revision 10316:86a76b6:20151031.

2015-10-15  6.9.2-5 Dirk Lemstra <dirk@lem.....org>
  * Fixed alpha blending issue with semi-transparent pixels in the merged
    image of PSD files. This can be disabled by setting the option
    'psd:alpha-unblend' to 'off' (reference
    http://www.imagemagick.org/discourse-server/viewtopic.php?f=3&t=28169).
  * Fixed issue in jpeg:extent that prevented it from working (reference
    http://www.imagemagick.org/discourse-server/viewtopic.php?f=3&p=126638).
  * Fixed memory leak when reading Photoshop layers in a TIFF file.

2015-10-12  6.9.2-5 Cristy  <quetzlzacatenango@image...>
  * Support gradient:bounding-box, gradient:vector, gradient:center, and
    gradient:radius to shape the gradient rendering.
  * Recognize label:@- as stdin (reference
    http://www.imagemagick.org/discourse-server/viewtopic.php?f=3&t=28501).
  * Make commas optional for coordinates (e.g. 0,0 256,256).

2015-10-10  6.9.2-4 Cristy  <quetzlzacatenango@image...>
  * New version 6.9.2-4, GIT revision 10273:e00cf21:20151010

2015-10-08  6.9.2-4 Dirk Lemstra <dirk@lem.....org>
  * Fixed accessing subimage in a TIFF photoshop layer (reference
    http://www.imagemagick.org/discourse-server/viewtopic.php?f=3&t=28469).
  * Fixed out of bounds error in -splice (reference
    http://www.imagemagick.org/discourse-server/viewtopic.php?f=3&t=28466).

2015-09-25  6.9.2-4 Eric McConville <emcconville@emccon...>
  * Created Manhattan Interpolate method for -sparse-color (reference
    ImageMagick/ImageMagick#36).

2015-09-22  6.9.2-4 Cristy  <quetzlzacatenango@image...>
  * Prevent null pointer access (reference
    ImageMagick/ImageMagick#34).
  * Specify histogram code image format implictedly (reference
    http://www.imagemagick.org/discourse-server/viewtopic.php?f=3&t=28412).
  * Don't round up for JPEG image resolution (reference
    http://www.imagemagick.org/discourse-server/viewtopic.php?f=1&t=28422).
  * Read the whole image @ image.jp2[0] or an individual tile @ image.jp2[1],
    image.jp2[2].
  * The -caption option no longer fails for filenames with @ prefix (reference
    ImageMagick/ImageMagick#43).
  * Honor $XDG_CONFIG_HOME and $XDG_CACHE_HOME (reference
    ImageMagick/ImageMagick#44).

2015-09-19  6.9.2-3 Cristy  <quetzlzacatenango@image...>
  * New version 6.9.2-3, GIT revision 10235:c40ee57:20150919

2015-09-17  6.9.2-2 Dirk Lemstra <dirk@lem.....org>
  * Added extra checks to avoid out of bounds error when parsing the 8bim
    profile (reference ImageMagick/ImageMagick#31).
  * Fixed size of memory allocation in RLE coder to avoid segfault (reference
    ImageMagick/ImageMagick#32).

2015-09-16  6.9.2-2 Cristy  <quetzlzacatenango@image...>
  * Add Bootstrap MIT license to ImageMagick web CSS and JS.
  * Use modern idiom for autoconf
    Fix a forgotten dir when regenerating html doc (reference
    http://www.imagemagick.org/discourse-server/viewtopic.php?f=3&t=28363).
  * The -colorspace gray option no long leaves a ghostly shadow (reference
    http://www.imagemagick.org/discourse-server/viewtopic.php?f=3&t=28350).

2015-09-14  6.9.2-1 Cristy  <quetzlzacatenango@image...>
  * New version 6.9.2-1, GIT revision 10201:5879612:20150913.

2015-09-13  6.9.2-1 Dirk Lemstra <dirk@lem.....org>
  * Preserving image origin in TGA files (reference
    http://www.imagemagick.org/discourse-server/viewtopic.php?f=1&t=28310).

2015-08-15  6.9.2-1 Cristy  <quetzlzacatenango@image...>
  * Support color compliance for TXT format (e.g. -define txt:compliance=css)
    (bug report from Anthony).
  * The first argument of a hsl() color is integer, not percentage (reference
    http://www.imagemagick.org/discourse-server/viewtopic.php?f=2&t=26072).
  * Limit -fx recursive to avoid stack overflow (reference
    http://www.imagemagick.org/discourse-server/viewtopic.php?f=3&t=28200).
  * Support inline images with data URLS.
  * Report progress less often (reference
    http://www.imagemagick.org/discourse-server/viewtopic.php?f=3&t=28299).
  * Don't set image colorspace to gray for -alpha copy option.
  * Fixed a possible memory leak associated with the -sampling-factor option.
  * GetImageType() no longer has side-effects to match behavior of IMv7.

2015-08-15  6.9.2-0 Cristy  <quetzlzacatenango@image...>
  * New version 6.9.2-0, GIT revision 10143:3534f6d:20150815.

2015-08-10  6.9.2-0 Cristy  <quetzlzacatenango@image...>
  * Swap pixels for -spread command-line option (reference
    http://www.imagemagick.org/discourse-server/viewtopic.php?f=1&t=28043).
  * fx logical operations short circuit (reference
    http://www.imagemagick.org/discourse-server/viewtopic.php?f=3&t=28179).
  * Type cast intensity variable (reference
    http://www.imagemagick.org/discourse-server/viewtopic.php?f=3&t=28182).
  * Fix ModulusAdd & ModulusSubstract for HDRI compositing (reference
    http://www.imagemagick.org/discourse-server/viewtopic.php?f=3&t=28186).

2015-07-25  6.9.2-0 Dirk Lemstra <dirk@lem.....org>
  * Added "-set colorspace:auto-grayscale false" that will prevent automatic
    conversion to grayscale inside coders that support grayscale.
  * Fixed -list weight and the options for -weight.
  * Added fontFamily, fontStyle, fontWeight and textUnderColor to the Image
    class of Magick++.
  * Fixed reading Photoshop layers of LSB TIFF files (reference
    http://www.imagemagick.org/discourse-server/viewtopic.php?f=1&t=28045).
  * Fixed memory leak in font cache under Windows (reference
    http://www.imagemagick.org/discourse-server/viewtopic.php?f=6&t=28191).

2015-07-25  6.9.1-10 Cristy  <quetzlzacatenango@image...>
  * New version 6.9.1-10, SVN revision 19269.

2015-07-20  6.9.1-10 Glenn Randers-Pehrson <glennrp@image...>
  * coders/magick.c: added "-define h:format=FMT" and
    "-define magick:format=FMT" options. FMT can be any output format
    supported by ImageMagick except "H" or "MAGICK". If this define is
    omitted, the encoder uses GIF by default for pseudoclass images and
    PNM for directclass images, as previously.  Made "MAGICK" (read-write)
    and "H" (write-only) visible in the format list.

2015-07-18  6.9.1-10 Dirk lemstra <dirk@lem.....org>
  * Removed incorrect EOF check in the DDS reader (reference
    http://www.imagemagick.org/discourse-server/viewtopic.php?f=3&t=28065).

2015-07-18  6.9.1-10 Cristy  <quetzlzacatenango@image...>
  * Fixed undefined behaviors (reference
    http://www.imagemagick.org/discourse-server/viewtopic.php?f=3&t=28067).
  * Return exception message for unknown image properties.
  * Color shift removed when reading transparent gray images (reference
    http://www.imagemagick.org/discourse-server/viewtopic.php?f=3&t=28081)

2015-07-18  6.9.1-9 Cristy  <quetzlzacatenango@image...>
  * New version 6.9.1-9, SVN revision 19200.

2015-07-12  6.9.1-9 Dirk Lemstra <dirk@lem.....org>
  * Fixed issue with radial gradient in MVG (reference
    http://www.imagemagick.org/discourse-server/viewtopic.php?f=3&t=27995).

2015-07-11  6.9.1-8 Cristy  <quetzlzacatenango@image...>
  * New version 6.9.1-8, SVN revision 19167.

2015-06-25  6.9.1-8 Cristy  <quetzlzacatenango@image...>
  * Correct install location of the Magick++ headers (reference
    https://github.com/ImageMagick/ImageMagick/pull/17/commits).
  * Different gif cropping behavior between versions (reference
    http://www.imagemagick.org/discourse-server/viewtopic.php?f=3&t=28013).
  * Cannot read properly simple psd file (reference
    http://www.imagemagick.org/discourse-server/viewtopic.php?f=3&t=28002).
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Development

No branches or pull requests

4 participants