[cups.general] Colors and accounting with HP LJ 4600

[Helge Blischke]

Martin Drab wrote:
> 
> On Thu, 3 Jun 2004, Helge Blischke wrote:
> 
> > Martin Drab wrote:
> > >
> > > On Wed, 2 Jun 2004, Helge Blischke wrote:
> > >
> > > > Martin Drab wrote:
> > > > >
> > > > > Hi,
> > > > >
> > > > > I have a problem, here, with the HP LJ 4600dn using
> > > > > CUPS 1.1.20 + SAMBA 3.0.2. When I use the PPD from the Wihdows HP drivers
> > > > > that identifies itself as "HP Color LaserJet 4600 PS v3010.107", the
> > > > > printing works, but the print accounting (i.e. logging into the
> > > > > /var/log/cups/page_log) does not! I thought I solved the problem by using
> > > > > the Foomatic hpijs driver. Accounting works there, but I've discovered
> > > > > another serious problem. The color tones are not correct, in fact they are
> > > > > _way_ off. So now it seems both PPD drivers are unusable.
> > > > >
> > > > > Can anybody help me on that one? I think the easiest way would be to fix
> > > > > the "HP Color LaserJet 4600 PS v3010.107" PPD to produce something (??)
> > > > > that would make it do the accounting. However, I don't know what should it
> > > > > be. Another way would be to fix the colors within the hpijs PPD, also no
> > > > > clue. :(
> > > > >
> > > > Isn't your printer a native PostScript printer?
> > > > If yes, don't use things like Foomatic.
> > >
> > > It is, but as I said, when using the original HP drivers (which
> > > communicates in PostScript directly), the page accounting in CUPS isn't
> > > working, and I really need that. The Foomatic drivers were just a
> > > workaround for this problem. It worked fine until I realized, that it
> > > totally messes the colors. That's my problem, I'd like to force the
> > > original HP PPD to do the page accounting with CUPS, somehow. And,
> > > frankly, I don't know why it isn't working.
> > >
> > Look into HP's PPD for a *JobPatchFile entry. If there is one, it probably contains
> > PostScript code to set up y CRD (color rendering dictionary) tailored for just this
> > printer, so you should use the vendor supplied PPD.
> 
> Unfortunatelly, this line isn't there. (See the attached PPD.)
> 
> > In addition, how is your samba configured with respect to printing? You should use
> > printing = cups and, on the windows side, set your printer instance preferences
> > to "optimize for portability". Then - at leas with most applications under Win - the
> > print job gets fed through CUPS' pstops filter which does (a rudimentary) page accounting.
> 
> It is configured exactly like that. The problem isn't with pstops not
> running. Every job _is_ passed through the pstops filter, but the filter
> simply doesn't do the accounting if the source is generated with the
> HP PPD.
> 
> I've attached the HP PPD that I'm using, so that you could take a look if
> prehaps there is something missing or wrong (?). It is not exactly the
> original HP, I had to comment out the lines 46 through 70, because it
> generated a "@PJL ..." header to the resulting PostScript and as such it
> was allways totally rejected by the CUPS, so that nothing was printed.
> This variant works fine, except for the accounting.
> 
> > Another easy to implement way would be to insert an additional filter
> > (using the *cupsFilter entry in the printer's PPD) which runs the PS job through
> > Ghostscript's bbox device and uses what Ghostscript talks to count the pages.
> 
> Could you please tell me how exactly should this line look like?
> 

Sorry, there is no attachment. Perhaps you could post an URL to the PPD
and a sample file? (As a last resort, you could e-mail me those two, but
don't
forget to use the above subject).

Helge


-- 
H.Blischke at srz-berlin.de
H.Blischke at srz-berlin.com
H.Blischke at acm.org