duplex printing cuts content

Bug #173857 reported by Johannes Meixner
72
This bug affects 6 people
Affects Status Importance Assigned to Milestone
HPLIP
Confirmed
Medium
David Suffield

Bug Description

I got a report from a colleague with a HP DeskJet 6540
for openSUSE 10.3 (i.e. HPLIP 2.7.7).
Perhaps it is already fixed in HPLIP 2.7.10?

Printing "simplex" with

echo -e 'Hello' | lp -d <queue> -o page-border=single

works correct. In particular the boder line is printed completely.

In contrast printing in duplex mode with

echo -e 'Hello\fWorld' | lp -d <queue> -o page-border=single \
 -o Duplex=DuplexNoTumble

cuts the word "Hello" and "World" and parts of the boder line
from the printout.

If I remember correctly there was a mail on the hplip mailing list
that for duplex a smaller imageable area is used inside the driver.

If this is true, it must lead to clipped printouts whenever an
application which produceds the PostScript input fills up the
whole imageable area as specified in the PPD file - i.e. when
the application works in compliance how CUPS is designed.

For example the CUPS texttops filter does it. In particular its
border line marks exactly the imageable area.

Revision history for this message
Johannes Meixner (jsmeix) wrote :

My colleague tested HPLIP 2.7.10 from
http://download.opensuse.org/repositories/home:/jsmeix/
and it is not fixed there so that the report is valid for current HPLIP.

Changed in hplip:
assignee: nobody → david-suffield
importance: Undecided → Medium
status: New → Confirmed
Revision history for this message
Aaron Albright (albrigha-deactivatedaccount) wrote :

Johannes,

Just wanted to make sure to update this so you know that we are working on this--hopefully in January. Can't promise any ETA..but Dave is looking into this.

Thanks for your support!

Aaron

Revision history for this message
Johannes Meixner (jsmeix) wrote :

Please keep me informed how you like to solve it.

My immediate idea is that you need additional paper size definitions
using so called "qualifiers" - see the Adobe PPD specification
for duplex printing in your PPDs together with matching constraints.

For example here only for A4 something like
(derived from HP-DeskJet_6540-hpijs.ppd):
-------------------------------------------------------
*UIConstraints: *PageSize A4 *Duplex DuplexNoTumble
*UIConstraints: *PageSize A4 *Duplex DuplexTumble
*UIConstraints: *Duplex DuplexNoTumble *PageSize A4
*UIConstraints: *Duplex DuplexTumble *PageSize A4
...
*PageSize A4/A4: "%% FoomaticRIPOptionSetting: PageSize=A4"
*FoomaticRIPOptionSetting PageSize=A4: " -dDEVICEWIDTHPOINTS=595 &&
-dDEVICEHEIGHTPOINTS=842"
*End
*PageSize A4.Duplex/A4 for duplex printing: "%% FoomaticRIPOptionSetting: PageSize=A4"
*FoomaticRIPOptionSetting PageSize=A4: " -dDEVICEWIDTHPOINTS=595 &&
-dDEVICEHEIGHTPOINTS=822"
*End
...
*ImageableArea A4/A4: "9.72 36 585.28 833"
*ImageableArea A4.Duplex/A4 for duplex printing: "9.72 36 585.28 813"
-------------------------------------------------------

By the way:
The above values for A4 from your HP-DeskJet_6540-hpijs.ppd
for DEVICEWIDTHPOINTS and DEVICEHEIGHTPOINTS
seem not to match to the ImageableArea (i.e. what is set for
Ghostscript seems to be bigger than what fits into the
ImageableArea).

Revision history for this message
Johannes Meixner (jsmeix) wrote :

And a nice side-effect of my proposal above is
that the DefaultPageSize should be set to A4.Duplex
because otherwise the constaints make printing
in duplex mode unnecessarily complicated for the user :-(

Revision history for this message
Jan Mynarik (jan-mynarik) wrote :

I have the same problem with my HP Photosmart C7280 in current (post-beta) Hardy.

Revision history for this message
Bernhard Gehl (bernhard-gehl) wrote :

I can confirm this (annoying) problem for a HP OfficeJet Pro 7580...
If anything has to be done for diagnostics, I'll be glad to help (it's less work than changing all those forms to fit the duplex bug... :) )

Revision history for this message
Bernhard Gehl (bernhard-gehl) wrote :

Oops: sorry, I forgot: Ubuntu 8.04 beta1, hplip 2.8.2

Revision history for this message
Antonio (antonio-moosefactory) wrote :

I just installed HPLIP 2.8.5 on CentOS 5.0 and can confirm this on a HP Photosmart D5360.

Can I fix this by editing the *ImageableArea Letter/Letter: line in the PPD for this printer? I'm sorry if that is a dumb question and it requires much much more. I'm trying to educate myself on PPD files right now.

Thanks,

Antonio

Revision history for this message
Brian (jmail524) wrote :

I am also having this problem. I am running Ubuntu 8.04 AMD64 with HPLIP-2.8.6 on a HP OfficeJet Pro L7680 (US Letter size paper).

Revision history for this message
Till Kamppeter (till-kamppeter) wrote :

Dave, any idea how to get this fixed?

Revision history for this message
Frank van Maarseveen (frankvm) wrote :

Looks like the Photosmart C6280 All-in-One printer with hplib 2.8.12.26 on
Fedora F10 has this issue too.

--
Frank

Revision history for this message
esekyavuz (esekyavuz) wrote :

Seconding Officejet L7680
Kubuntu 8.10 x64
KDE 4.1
HPLIP 2.8.7

Revision history for this message
esekyavuz (esekyavuz) wrote :

Got this error doing a double sided COPY on my L7680. I mean I put something in the auto feed paper tray and selected double sided copy on the printer. Could this still be a HPLIP problem, then? Does anyone else get this?

Still have:
Officejet L7680
Kubuntu 8.10 x64
KDE 4.1
HPLIP 2.8.7

Revision history for this message
Pete (pete123) wrote :

Also a problem with copying and printing on both L7780 and an old dx155xi. For the printing I forced margins to 0mm to see if that would make a difference, however text was still chopped as with the copy.

Testing just with the L7700 series on windows using acrobat reader to print the same PDF 'cheats' and scales the document down slightly (introducing a margin around the document when printing duplex), although I presume this scaling is probably done post Acrobat.

I guess it's possible that the full page printing mechanism inside the printer is disabled for a duplex print/copy, which the windows driver is aware of and scales the document appropriately. As this impacts both copying and printing I guess some kind of workaround is still needed inside HPLIP (to make it function more like it's windows counter part) and HP needs to do something in their firmware to fix copying.

Revision history for this message
Brian (jmail524) wrote :

I am also having this problem.

Ubuntu 9.04 AMD64, HPLIP 3.9.4b, HP OfficeJet Pro L7680.

Revision history for this message
Wolfram Riedel (taisto-web) wrote :

Confirmed for HP Photosmart Premium C309 with Jaunty. Looks like the Wind*ws driver scales to 97% for automatic duplex printing.

Revision history for this message
Johannes Meixner (jsmeix) wrote :

According how I understand
http://hplipopensource.com/hplip-web/release_notes.html
---------------------------------------------------------
HPLIP 3.9.6
...
The new native CUPS driver (HPCUPS)
is now the default printer driver.
...
HPCUPS provides new paper sizes for borderless and duplex.
This means we now have PPDs with exact printable regions
for normal, borderless and duplex paper sizes.
---------------------------------------------------------
the issue should be fixed since version 3.9.6.

If it is fixed, the crucial point is that you may have to
re-setup your printer so that a PPD for the
HPCUPS driver is used.

Those PPDs contain the line
*cupsFilter: "application/vnd.cups-raster 0 hpcups"
while the old HPIJS/Foomatic PPDs contain lines like
*cupsFilter: "application/vnd.cups-postscript 0 foomatic-rip"
*cupsFilter: "application/vnd.cups-postscript 100 foomatic-rip-hplip"
*cupsFilter: "application/vnd.cups-pdf 0 foomatic-rip-hplip"

The HPCUPS PPDs contain in their NickName entries
the word "hpcups" like "HP LaserJet 1220 pcl3, hpcups 3.9.6b.4"
while the old HPIJS/foomatic PPDs contain in their NickName
the word "hpijs" like "HP LaserJet 1220 hpijs pcl3, 3.9.6b.4"

Printer setup tools which work in compliance to CUPS
show the NickName from the PPD to the user so that
you can select the right PPD for your printer.

The PPD which CUPS actually uses for you printer is in
/etc/cups/ppd/<queue_name>.ppd

Inspect this PPD to verify that actually the HPCUPS driver
is used and not the old HPIJS/Foomatic driver.

Revision history for this message
a.h() (a.h) wrote :

I'm using the hpcups driver, with duplex paper set up, but the same problem still occurs.

Revision history for this message
BonnieJean Wiebe (bonniejean) wrote :

I'm also using the cups version 1.3.7 and hplip version 3.9.8 and am having the same duplexing problems and many above. Basically it starts printing the page 3/4" down from the top edge and cuts the bottom of the printout. When feeding the page back in through the duplexer, it starts the bottom of the next side 3/4" in and cuts the top printing off.
We have Hp Business Inkjet 1200 printer.

*PPD-Adobe: "4.3"
*% PPD file for HP Business Inkjet 1200 with CUPS.
*% Created by the CUPS PPD Compiler v1.2.3.
*% (c) 2008 Copyright Hewlett-Packard Development Company, LP
*FormatVersion: "4.3"
*FileVersion: "hpcups 3.9.8.36"

Revision history for this message
a.h() (a.h) wrote :

I tried editing my ppd, to make the printable areas and paper size for A4 duplex the same as for normal A4, and... sadly no change.

Revision history for this message
a.h() (a.h) wrote :

Also, under Windows the problem doesn't occur. Also, the problem didn't occur on my old HP Deskjet 990cxi, but that was with hplip 2.4, I'm now on 3.9.8 (had to install 3.9.8 for my Officjet J6410). I can't however test the 990cxi until probably October.

Revision history for this message
martin_jacobs (martin-jacobs) wrote :

My environment is affected too, it's a own compilation based on SuSE Linux 8.1 and CPUS 1.3.11, HPLIP 3.9.10. My printer is HP PhotoSmart C7280, connected via WLAN.

Iv'e done three tests:

1. echo -e 'Hello\n\nhello-simplex-001.pdf' | lp -d litfass -o page-border=single
2. echo -e 'Hello\fWorld\n\nhello-duplex-001.pdf' | lp -d litfass -o page-border=single -o Duplex=DuplexNoTumble
3. echo -e 'Hello\fWorld\n\nhello-duplex-002.pdf' | lp -d litfass-duplex -o page-border=single

After that I scanned each result and saved them with given file name, see attached archive (173857-scan-and-ppd.tgz) which contains these files:

-rw-r----- mjacobs/users 27940 2009-11-14 09:25:12 hello-duplex-001.pdf
-rw-r----- mjacobs/users 25839 2009-11-14 09:33:07 hello-duplex-002.pdf
-rw-r----- mjacobs/users 16521 2009-11-14 09:19:25 hello-simplex-001.pdf
-rw-r--r-- root/root 28086 2009-11-08 14:53:37 etc/cups/ppd/litfass-duplex.ppd
-rw-r--r-- root/root 28139 2009-11-07 21:49:27 etc/cups/ppd/litfass.ppd

By the way, scan is done with that hpio/xsane/PhotoSmart C7280

Revision history for this message
Jason Callough (jcallough) wrote :

Martin:

We have been able to confirm that this problem is corrected with the latest release (3.9.10) if you set the duplexer to installed on the cups localhost page. Then run the following command:

echo -e 'Hello\fWorld' | lp -d HP_Officejet_Pro_8000_A809 -o page-border=single -o Duplex=DuplexNoTumble -o media=Letter.Duplex

We are not able to see any border on the printout because lp is not creating it and we do not see it on the input to the driver.

Jason

Revision history for this message
martin_jacobs (martin-jacobs) wrote : Re: [Bug 173857] Re: duplex printing cuts content

Hi Jason,

thanks for your advise. First, I used my printer "litfass"
during my tests. This printer has duplexer marked as
installed. At least CUPS web frontend shows Duplexer as
installed. I did not find out how and where this is reflected
in /etc/cups/ppd/litfass.ppd.

On Fri, 20 Nov 2009, Jason Callough wrote:

> Martin:
>
> We have been able to confirm that this problem is corrected with the
> latest release (3.9.10) if you set the duplexer to installed on the cups
> localhost page. Then run the following command:
>
> echo -e 'Hello\fWorld' | lp -d HP_Officejet_Pro_8000_A809 -o page-
> border=single -o Duplex=DuplexNoTumble -o media=Letter.Duplex
>
> We are not able to see any border on the printout because lp is not
> creating it and we do not see it on the input to the driver.
>
> Jason
> ...

Regards

Martin Jacobs

--
Dipl-Ing. Martin Jacobs * Windsbach * <email address hidden> und
<email address hidden> * ICQ:153921785
Registered Linux User #87175, http://counter.li.org/

Revision history for this message
martin_jacobs (martin-jacobs) wrote :

Hi Jason,

I've made two additional test with these commands:

echo -e 'Hello\nlp -d litfass -o page-border=single -o
Duplex=DuplexNoTumble -o media=A4.Duplex\fWorld' | lp -d
litfass -o page-border=single -o Duplex=DuplexNoTumble -o
media=A4.Duplex

echo -e 'Hello\nlp -d litfass-duplex -o page-border=single -o
Duplex=DuplexNoTumble -o media=A4.Duplex\fWorld' | lp -d
litfass-duplex -o page-border=single -o Duplex=DuplexNoTumble
-o media=A4.Duplex

Both create a double sided print out with mentionable
differences. I've scanned both outputs and will upload them to
launchpad. Filenames are given below:

First command produces two pages, first cropped at the bottom,
rear side with a full frame (see file
launchpad-173857-20091121-001.pdf).

Second command produces two pages, front-side with full frame
but with "large" empty space at top and bottom, rear-side with
"smaller" frame but shifted towards bottom of page and
cropped (see file launchpad-173857-20091121-002.pdf).

Here is diff output of

 diff /etc/cups/ppd/litfass.ppd /etc/cups/ppd/litfass-duplex.ppd

339c339
< *DefaultDuplex: None
---
> *DefaultDuplex: DuplexNoTumble

And, as mentioned earlier, CUPS web front-end shows duplexer
marked as installed.

On Fri, 20 Nov 2009, Jason Callough wrote:

> Martin:
>
> We have been able to confirm that this problem is corrected with the
> latest release (3.9.10) if you set the duplexer to installed on the cups
> localhost page. Then run the following command:
>
> echo -e 'Hello\fWorld' | lp -d HP_Officejet_Pro_8000_A809 -o page-
> border=single -o Duplex=DuplexNoTumble -o media=Letter.Duplex
>
> We are not able to see any border on the printout because lp is not
> creating it and we do not see it on the input to the driver.
>
> Jason
> ...

Regards

Martin

Revision history for this message
martin_jacobs (martin-jacobs) wrote :

Here is scanned result of command:

echo -e 'Hello\nlp -d litfass -o page-border=single -o
Duplex=DuplexNoTumble -o media=A4.Duplex\fWorld' | lp -d
litfass -o page-border=single -o Duplex=DuplexNoTumble -o
media=A4.Duplex

Revision history for this message
martin_jacobs (martin-jacobs) wrote :

Here is scanned result of command:

echo -e 'Hello\nlp -d litfass-duplex -o page-border=single -o
Duplex=DuplexNoTumble -o media=A4.Duplex\fWorld' | lp -d
litfass-duplex -o page-border=single -o Duplex=DuplexNoTumble
-o media=A4.Duplex

Revision history for this message
a.h() (a.h) wrote :

Hi all,

I was wondering whether any progress is being done on this rather major bug? It shouldn't be too much of a bother to fix (2.X worked fine...), so is anything likely to happen, or do those needing double sided printing need to continue with the 2.X series, which isn't included in most distributions any more? (3 years now?)

Revision history for this message
florianr (florianr-gmx) wrote :

I installed 3.10 with hope for a fix of this problem. But things get even worth ... see bug report: https://bugs.launchpad.net/hplip/+bug/551239

Revision history for this message
a.h() (a.h) wrote :

Hmm... Maybe I'm just going senile/paranoid/crazy... but is the hplip team actually continuously breaking ever more of the printing system (and actually releasing these backward steps!)? I know they are working on making a good printing system, but is it really acceptable for a (rather major) bug which didn't exist in older versions, which makes the current versions unusable for some, which is now present in most linux distributions, to just be ignored?

Revision history for this message
Andreas Klöckner (inform) wrote :

Does not appear fixed with

HP LaserJet p3005 pcl3, hpcups 3.10.2

on letter paper.

Revision history for this message
martin_jacobs (martin-jacobs) wrote :

I upgraded to hplip 3.10.5 and renewed my ppd files. CUPS list my C7280 printer now as hpcups 3.10.4.16 and repeated my simple tests:

1. echo -e 'Hello\n\nhello-simplex-001.pdf' | lp -d litfass -o page-border=single
2. echo -e 'Hello\fWorld\n\nhello-duplex-001.pdf' | lp -d litfass -o page-border=single -o Duplex=DuplexNoTumble
3. echo -e 'Hello\fWorld\n\nhello-duplex-002.pdf' | lp -d litfass-duplex -o page-border=single

Result: first page is still cut at the bottom.

All three results and used ppd files are attached in archive 173857-scan-and-ppd-3_10_5.tgz Content of this archive is

hello-duplex-001.pdf
hello-duplex-002.pdf
hello-simplex-001.pdf
etc/cups/ppd/litfass-duplex.ppd
etc/cups/ppd/litfass.ppd

By the way, scans are done using scanimage using scanner of C7280 using hpio device connecte via WLAN.

Revision history for this message
Alvaro Fernando García (alvarofernandogarcia) wrote :

I am also having this problem.

Arch x86_64, HPLIP 3.11.7 , HP Photosmart c6280 .

Revision history for this message
Jakob Wa (jawalkob) wrote :

I am having the same problem on a HP Officejet 6700 Premium , hplip 3.20.3 , Ubuntu 20.04

To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Duplicates of this bug

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.