[ Resent for Milan -jhnc ]
----------------------------------------------------------------------------
I'm replying to the mailing list, because the details can be interesting
for tracking my problem with Epson Perfection 1200 Photo that I describe
below.
>>>>> "JHNC" == Jonathan H N Chin <jc254@newton.cam.ac.uk> writes:
JHNC> To anyone who is succesfully using an Epson 1200 Photo
JHNC> (including the TPU) under Linux with USB on a "PC":
Well, the scanner mostly works, including TPU.
JHNC> 1) version of linux kernel 1a) any kernel patches you have
JHNC> applied 1b) is it SMP?
2.2.17pre6 compiled from Debian 2.2 sources, USB backport
usb-2.4.0-test2-pre2-for-2.2.16-v3.diff.gz, ReiserFS patch, no SMP.
JHNC> 2) version of SANE epson backend 3) version of SANE
sane 1.0.3 with the included Epson backend.
JHNC> 4) version of xsane, if you use it
0.61 with a hack not to get the list of devices but to use Epson driver
directly. This is because SANE freezes on an attempt to list devices
(yes, I'm going to send an xsane patch to Oliver Rauch sometimes).
JHNC> 5) motherboard model 5a) processor(s) model 5b) if USB is
JHNC> not built-in, what additional hardware you use
TopGun Pentium board with AMD K6/200, general USB PCI card, UHCI.
JHNC> 6) any problems you had (USB or SANE related) 6a) and how
JHNC> you overcame them
I have two problems:
1) The insignificant one is `scanimage --help -d epson' freezes after
writing out the appropriate information, as well as `sane --list-devices'.
I have to switch off/on the scanner. I overcame this by saving the
help into a file for future reference, specifying the epson driver
to SANE always directly and hacking xsane, the problem doesn't appear
anymore. :-)
2) The significant one is that in most cases (but not always) scanimage,
the driver and the scanner all freeze during scan at higher
resolutions. I have to switch off the scanner and reboot my computer
(killing scanimage doesn't help, the `scanner' module remains busy)
to fix the problem. My current observations of the problem are:
- It never happens during 600 dpi (or lower) scans.
- It happened once during 1200 dpi scan; it very often happens with
2400 dpi.
- It seems to happen only when using TPU (I tried about 5 scans with
flatbed and it never happened).
- It seems to happen only when gamma correction is used
(--gamma-red-table etc.) but I'm not sure about this.
- If TPU is used and resolution is 2400 dpi and gamma correction is
specified, then:
* If it is the very first scan after switching on the scanner, it's
very likely to succeed. It has frozen only once, when I had
switched off the TPU, but specified it as a source (it switched
on automatically and the scan has frozen).
* Otherwise it always freezes.
- It always freezes at the beginning of the scan, without outputting
any image data.
Note these observations aren't based on large amount of statistical data.
The following information from drivers is available:
- SANE report with SANE_DEBUG_EPSON=1:
[epson] error in receive - status = 5
scanimage: sane_read: Invalid argument
[epson] error in receive - status = 5
[epson] error in receive - status = 5
... <many times repeating>
- Kernel log:
Sep 3 00:02:42 blackbird kernel: usb_control/bulk_msg: timeout
Sep 3 00:02:52 blackbird kernel: usb-uhci.c: interrupt, status 3, frame# 804
Sep 3 00:02:52 blackbird kernel: scanner.c: read_scanner(0): funky result:-32. Please notify the maintainer.
Sep 3 00:02:52 blackbird kernel: scanner.c: read_scanner(0): funky result:-32. Please notify the maintainer.
- On the very first scan after switching on the TPU (and using it)
SANE reports:
[epson] fatal error - Status = 92
[epson] retrying ESC G - 1
[epson] fatal error - Status = 92
[epson] retrying ESC G - 2
[epson] fatal error - Status = 92
[epson] retrying ESC G - 3
[epson] fatal error - Status = 92
[epson] retrying ESC G - 4
[epson] fatal error - Status = 92
[epson] retrying ESC G - 5
After reporting these messages the performing scan continues
happily.
I'd appreciate any help, because the vision of rebooting before each
high resolution TPU scan isn't much fine. :-(
Milan Zamazal
--
"Having GNU Emacs is like having a dragon's cave of treasures."
Robert J. Chassell
--
Source code, list archive, and docs: http://www.mostang.com/sane/
To unsubscribe: echo unsubscribe sane-devel | mail majordomo@mostang.com
This archive was generated by hypermail 2b29 : Tue Sep 05 2000 - 03:43:26 PDT