Thank you very much. I think I will try the changes you suggest in magickcore first, and then experiment with Magickwand since it does look better suited to my needs. The API is quite big at first glance, but looks a lot friendlier to use! Is it built upon magickcore, or is it completely standalone?
Regarding magickwand usage, I see that I can create a new image with MagickNewImage() as a start. Is there a way to specify that it should be stored on disk only, or to stream pixels, rows or tiles from my image data to the file on disk in order to control memory usage when saving large images? Is there a recommended strategy for this when using imagemagick? (my image data is held in tiles internally, but it is easy for me to pull out entire rows or single pixels). Finally, is there a tutorial anyone can recommend to me for either magickwand, or magickcore usage (preferably for the C api obviously)? I have been unable to find one as yet. Best regards, Chris. Chris Garrett UNIX Administrator MISD Operations Group, University of Cambridge 01223-(7)66687 -----Original Message----- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Sent: 26 April 2007 20:08 To: Christopher Garrett; [email protected] Subject: Re: [magick-users] Saving using magickcore > does not actually produce a file at the end of it We see a few mistakes. Number 1, user MagickWand rather than MagickCore. If you must use MagickCore, you need to set the output image filename at image->filename. You also need to call SyncImagePixels() at the close of your outer loop to ensure your pixels are updated. Finally, your image assumes a Q8 version of ImageMagick. If you have a Q16 version you will need to scale your pixels froom 0..255 to 0..65535. _______________________________________________ Magick-users mailing list [email protected] http://studio.imagemagick.org/mailman/listinfo/magick-users
