Bug#702673: PostScript drivers won't work for HP LaserJet 1200
The best explanation why it actually makes sense that the entire page is
rendered as a bitmap I found at the ghostscript bug tracker (by Ray
Johnston) [1]:
> BTW, the reason that the entire page is rendered as an image is that the
> PDF transparency model requires any transparent region to be 'blended' on
> top of any graphics on the page. Since we don't do full analysis of the
> PDF to determine what paints where, we have to render graphics to the
> page level transparency buffers just in case some later transparency
> group (that may have a BBox that is a subset of the page) must be blended
> onto some marks already made on the page level.
>
> Trying to allow the printer PS RIP to render parts of the page outside the
> area where the transparency is actually used (as with an image with SMask)
> might work in simple cases, but rendering detail differences would make it
> impossible to have objects that cross "into" the area that is pre-rendered
> have the correct appearance (no glitches at the boundary, color matches,
> etc.)
When I print the PS file attached to my last message I see glitches at
the boundary of the bounding box of the image in the middle, the colors
(levels of gray in my case) don't match…
So as I understand it, in this case Adobe Reader produces a (relatively
small) PS file which is printed much faster but the appearance isn't
correct. Ghostscript on the other hand produces a correct appearance but
the printing speed is unacceptable.
In my opinion the ghostscript solution is better than the one Adobe
Reader chooses. There should be no compromises with the quality of the
result – and if I choose to print a page with 1200dpi I want to get a
high quality result. When I set the printing resolution to 600dpi I get
a PS file with 12.3 MB instead of 43.7 MB (@ 1.200dpi).
So in this case I'm not affected by a bug but have a problem with my
outdated hardware. Still, I'll have a look at the other issue mentioned
in the first part of Message #121 of this bug report.
And again I'd like to re-emphasize the craziness of the PostScript PPD
to be recommended for this printer.
I think 99% of all users would call it a bug if they'd have to wait 1.5
hours for a single page to get printed – even if they chose the highest
printing quality with a resolution of 1.200dpi. With the HPIJS PCL3 PPD
it takes the same printer only 18 seconds (!) to render and print the
same page in a comparable quality.
Could someone give me a hint against which package I could file this bug
(regarding the recommendation of the PostScript PPD)?
Thanks,
Stefan.
[1] http://bugs.ghostscript.com/show_bug.cgi?id=692959#c15
Reply to: