[vox-tech] postscript printer: light goes on, nobody home.

[Peter Jay Salzman]

On Sun 01 Feb 04,  2:12 PM, Jonathan Stickel <jjstickel@sbcglobal.net> said:
> Peter Jay Salzman wrote:
> >hi all,
> >
> >i have an HP laserjet 6MP.  it works great on linux.  it's shared with
> >other linux and windows systems.  no problems.
> >
> >well, almost no problems.
> >
> >every blue moon, i'll queue something with lpr, and the printer LEDs
> >will flash like they normally do to indicate "i'm receiving data".
> >
> >but after the data gets trasfered to the printer, the "data LED" and
> >"ready LED" will remain on.  my printer manual says this state is
> >called:
> >
> >    "data is in the printer memory (form feed)"
> >
> >and says to press the "go button" to start printing.  i hit the "go
> >button", but nothing prints.
> >
> >eventually, the data LED goes out indicating the printer is in the
> >"ready for printing" state.  the job is gone.
> >
> >it's like the printer receives data but just gives up on printing it.
> >this happens on all the operating systems in the house: debian, suse,
> >redhat, win2k and win98.
> >
> >has anybody seen this happen before?  any guesses as to what causes it?
> 
> This sounds very similar to something I've experienced with an HP 4050, 
> a networked double-sided laserjet in my department's computer lab.  It 
> will hang with the "data" led flashing.  Since other people use it, I 
> can't wait around so I cancel the job.  It seems to happen mostly with 
> journal article pdfs with 3D graphics.  Strangely, I can print them just 
> fine to a HP 4100 laserjet (single-sided).
> 
> >the only thing i can think of is perhaps a flaw in the postscript or
> >perhaps the printer's postscript interpreter.
> 
> Yep, that's my guess too.  Not sure how to fix it.  Next time I might try 
> bitmapping just the pages with the 3d grahpics to confirm my hypothesis.
> 
> Jonathan

jon, please let me know.  i'm interested in hearing what happens,
although i strongly suspect it'll work for bitmaps.

i have some more information to share:

* it seems to happen with postscript more often than other things.  i
  can't recall it ever happening for bitmaps.

* recently, i was trying to print something with gnumeric (which like MUCH
  better than open office) and was having this problem.  it just
  wouldn't print.

  it did print from within gimp when i created a ps and pdf file.

  also, i loaded the spreadsheet with openoffice, and it printed fine
  from within open office!

  i think that's a strong indication that there must be broken
  postscript being generated by some applications.

* rhonda reminded me of a very interesting fact.

  she was trying to print a normal text file, and the same thing was
  happening.  she couldn't print the text file.

  i fiddled and fiddled and fiddled.  spent the entire afternoon at it,
  and was able to determine that the problem was a percent sign before
  a number at the left column of the file.  without that percent sign
  in the text document, the file printed fine.

  my belief is that the percent sign was tricking the printer's
  PCL interpreter into thinking the document was postscript, which
  of course it wasn't.  not even close.  not hard to imagine why
  the printer would give up.

now, that last story doesn't really say why the printer will give up on
certain postscript files, but i guess something similar must be
happening.

i recall reading there are many variants of postscript.  even some
variants that have been semi-broken on purpose, like mathematica's
version of postscript.  i wish i could get some definite answers (and
better, solutions) about this.   :(

pete

-- 
Make everything as simple as possible, but no simpler.  -- Albert Einstein
GPG Instructions: http://www.dirac.org/linux/gpg
GPG Fingerprint: B9F1 6CF3 47C4 7CD8 D33E 70A9 A3B9 1945 67EA 951D