Legacy ImageMagick Discussions Archive

Hello,

I have followed the instructions to build ImageMagick from source on a Windows 7 box. 32-bit is fine, but I don't see a 64-bit configuration. When I naively try to create a 64-bit configuration, I get an error in CORE_ffi (among other projects) stemming from the build of win32.asm - it makes a 32-bit obj, which then causes a link error. Is a 64-bit build possible, and if so, how to do it?

Second question - what I'm looking for from ImageMagick is an interface to tiff images that allows me to read 16bpp images without losing any pixel data, and to retrieve the data in something like scan-line form (regardless of how it's stored in the image). This would be possible with libtiff, but it looks like it would be extremely fussy and error-prone. I'm hoping for a higher level interface to the pixels, without conversion to 8bpp. Is this feasible with ImageMagick, and if so, where would I look for example code?

Thanks much for any assistance!

Bill

Silly me - already answered my question about 64-bit - just run the configure program with the 64-bit option checked. So that's one down.

Still very much interested in anyone's input on the 16bpp tiff question.

Thanks,
Bill

Imagemagick should be able to read 16-bits per pixel per channel images (48-bits for RGB) if you use the Q16 install, which is the default. Whatever format it is when read, it would likely be stored in scan-line format. Imagemagick uses libtif to read and write TIFF images. So if libtif is limited, then so would Imagemagick.

Q64 means 64 bits per channel per pixel. The last time I built Q64, it was automatically HDRI (floating point). Q64 integer wasn't available, but that might have changed more recently.

I'm not sure what you want for the second question. IM can read tiff files. You can then access the pixel data from the command line, or C or C++ programs or many other interfaces.