Closed michaelrsweet closed 19 years ago
CUPS.org User: mike
Hmm, I'm guessing that the PDF file includes shaded polygons, which could explain things a bit. I'll do some more investigation and let you know...
CUPS.org User: mike
Nope, it looks like there are a bunch of small images that are repeated many, many times. Xpdf does not try to store and reference the image data, but Acrobat Reader does.
I suspect if you turned off the optimization options in the Acrobat dialog that you would see similar output size...
I can bump this down to an RFE, however it looks like this is not a bug but an optimization issue.
CUPS.org User: steve.p.walsh.hp
I'm happy with that though I'm not sure where the multiple small images come from unless the two pie charts are made up of them. The document is not typical anyway and the filter does in general create postscript of comparable size to acrobat.
Many thanks.
CUPS.org User: mike
OK, we've funded an update from Derek (author of Xpdf) that fixes this particular issue - the Xpdf 3.x code will be integrated into 1.2cvs sometime next week. Will let you know when the final changes are in...
CUPS.org User: darkmind971
as of 26/04/2005 with cups 1.1.23, this bug is still valid but this time when the PDF contains big images. I will try to provide a PDF as example.
CUPS.org User: mike
Fixed in Subversion repository.
It appears this is resolved with the Xpdf 3.01 baseline code, which is now part of CUPS 1.2...
New output size for your sample file is 260468 bytes, vs. 31810086 bytes in the CUPS 1.1.23 version of pdftops.
Version: 1.2-feature CUPS.org User: steve.p.walsh.hp
Solaris 8
pdf file attached created postscript output in excess of 30MB when printed via pdftops whilst acroread (5) created postscript output of around 275K. Is this a bug or just a limitation of pdftops ?