Xerox 8400 printing b+w with colour consumables

[Helge Blischke]

Anonymous wrote:
> 
> > Anonymous wrote:
> > >
> > > > > Anonymous wrote:
> > > > > >
> > > > > > > Anonymous wrote:
> > > > > > > >
> > > > > > > > Hi
> > > > > > > >
> > > > > > > > I'm running a cups server from cups 1.1.22 on Scientific Linux 4 and printing from other linux clients.
> > > > > > > >
> > > > > > > > Our Xerox 8400 color printer (using the Xerox supplied ppd file) is using colour consumables to print most black and white files instead of the
> > > > > > > > black ink consumable.
> > > > > > > >
> > > > > > > > What I'd like to happen is for colour postscript files to print in colour and black and white postscript files to print using the black ink only without the user having to specify whether to print in black and white or colour.
> > > > > > > >
> > > > > > > > I notice that printing plain text files does seem to use only black ink, so the problem seems to be particular to postscript documents.
> > > > > > > >
> > > > > > > > Can anyone suggest how to obtain the desired behaviour for black and white postscript files.
> > > > > > > >
> > > > > > >
> > > > > > > Where do those jobs come from? And could you post (an URL to) a sample
> > > > > > > file? It may be there is a chance
> > > > > > > to print in b/w if the color components (be it RGB or CMY) are all
> > > > > > > equal.
> > > > > > >
> > > > > > > Helge
> > > > > > >
> > > > >
> > > > > >
> > > > > > Helge
> > > > > >
> > > > > > Thanks for your reply. I've put an example file at: http://www-xray.ast.cam.ac.uk/~rmj/pgplot.ps
> > > > > >
> > > > > > That file was made by the pgplot plotting package, but there are also files from dvips that use the colour consumables to render black print.
> > > > > >
> > > > > > Roderick
> > > > >
> > > > > I just downloaded the PPDs for those 8400 printers. It seems that what I
> > > > > suggested in my previous post
> > > > > isn't as easy as I thought (the PPDs already contain a job patch file
> > > > > ....), but using the
> > > > > command line option "-o XRXColor=BW" (without the quotes) or equivalent
> > > > > should switch to b/w
> > > > > printing. Anything else I don't dare to propose as I haven't such a
> > > > > printer at hand to play with.
> > > > >
> > > > > Helge
> > > > >
> > > >
> > > >
> > > > Helge
> > > >
> > > > Thanks, that works, but I was hoping for a way to automatically make the printer not use colour inks for a black and white postscript file.
> > > >
> > > > Since the printer can print plain text with only the black ink there must be a way for it to know automatically whether to use the colour inks or not.
> > > >
> > > > I'd be happy to try out any ideas you might have.
> > > >
> > > > Roderick
> > >
> > > What I have found out today is that the text files can be printed with only black ink by printing them with -o raw. My guess is that they are printed with the printer in PCL mode, which somehow defaults to black ink. I don't have any colour pcl files to try.
> > >
> > > Postscript files or plain text files printed without -o raw (which I guess get wrapped in postscript) always seem to use colour inks to render blacks unless I explicitly use lpr -o XRXColor=BW. So, I guess the printer can't determine whether the file is b+w or colour itself (or at least the postscript interpreter can't) and just does as its told in this case.
> > >
> > > So, what I need to do is find a way of detecting whether a file is colour ps or not and then do the right thing in terms of printer commands. I can think of sort of how to do this with a shell script by redefining the lpr and lp commands to check whether a file is destined for this printer then grep it for "color" (or similar) and construct a command line with the right setting for XRXColor, but this is a horrible way to do it and I don't know if it would work for binary ps files, not to mention files of other type.
> > >
> > > Can anyone suggest the elegant way to achieve this please? ie firstly to determine whether the file needs to be rendered in colour or B+W and secondly how send the appropriate XRXColor settings to the printer. I have an idea that its possible to run a shell script from a custom ppd file, but maybe there is a better way to do it than that.
> > >
> > > Thanks
> > >
> > > Roderick
> >
> > To check for using color in a PostScript job requires a full PostScript
> > interpreter (like Ghostscript or Distiller)
> > running a set of redefined PS operators in order to reliably distinguish
> > betweenb b/w and color. I think it
> > is quite a task (but perhaps in a couple of weeks it may be I have to
> > write such a beast for use at our site ...).
> >
> > Helge
> >
> > --
> > Helge Blischke
> > Softwareentwicklung
> > SRZ Berlin | Firmengruppe besscom
> > http://www.srz.de
> 
> Hi
> 
> A quick update to say how I resolved this issue, with the help of the Xerox support team.
> 
> It turns out that one of the values that the "OutputMode" setting in the ppd file can take causes the printer to use only black ink when printing black. Legal values are Fastcolor, Standard, Enhanced and HiResPhoto. The Enhanced setting (and only this setting) has the desired effect. Although this is not a 'general case solution', its good enough for our case. In my tests, this setting took only slightly more interpreter time to render postscript files and used more colour inks for the colour parts of the documents than Fastcolor or Standard, but this is not a major problem for us as our documents are typically mostly black writing with a few colour diagrams or figures.
> 
> Thanks for all your helpful comments.
> 
> Roderick

Strange ...
The funny thing is that the only differences mentioned in the PPDs for
the various 8400 printers are
the resolution settings. Xerox should at least place a comment into the
PPDs as a hint to this
weird behaviour.

Helge

-- 
Helge Blischke
Softwareentwicklung
SRZ Berlin | Firmengruppe besscom
http://www.srz.de