hp3300backend-devel Mailing List for HP3300c SANE backend (Page 3)

SourceForge Headquarters 225 Broadway Suite 1600 San Diego, CA 92101 +1 (858) 422-6466

Where i can get the sane-backends 1.0.9 sources ?

Hi all,

Here are a couple of my thoughts about the status of the niash backend.

1) Rework of the transfer method layer
Currently each transfer method (USB using kernel module, USB using libusb,
USB using uscanner, libieee1284 transfers) only has an init which both
initialises the transfer method and tries to open the first available
device. I intend to extend that a bit (actually I have already started).
The new init function will report each device that it thinks may be used
to access a scanner, but not open it yet. There will be a new function
("open") that actually opens the scanner.
This way it should be possible to show the actually present scanners
(instead of always reporting a hp3300c even if no scanners are
connected) and it should even be possible to use more than one scanner
at once.

2) Sane 1.0.9 will support both libusb and the kernel module
I think this will simplify a lot of things, meaning we can use the
sanei_usb transfer method for the SANE backend and forget about the
specific kernel module, libusb transfer method and uscanner transfer
method.
The only remaining transfer methods will be sanei_usb and the libieee1284
method.

3) Image still stretched for HP4300c ?
I got an error report that the image scanned using the SANE backend
was still stretched (although I don't have actual images showing this).
I thought I fixed this a couple of version ago, but perhaps it got
broken again in one of the recent releases.

4) Other open issues
* we still don't know how to set the vertical starting position of the
scan for HP3400/4300 scanners
* we don't know how to properly set the vertical resolution for HP3400/
4300 scanners (what is implemented now, is rather tricky)
* there is no coarse calibration done to set the gain of the analog parts
(we still use a set of fixed values, which seem to work quite well for
most people. Although some other people report vertical stripes which is
a typical symptom)
* parallel port access test
I think there is a reasonable chance that the backend will work over the
parallel port, but it's hardly ever been tested.
* timer for the scanner lamp?
It would be nice to have some mechanism that keep track of whether the
scanner lamp has already warmed up or not.

If you there is anything that you can more-or-less independently
work on, please let me know.

All the best,
Bertrik

NetNoise wrote:

>I cannot understand why my hp3300c work if I use the patch in sane1.0.7 but
>if I use the same patch in sane1.0.8 it tells me:
>$ scanimage > ~testfile
>scanimage: no sane device found
>
>$ scanimage -d niash > ~testfile
>scanimage: open of device niash failed: invalid argument
>
>I can't understand why, since the configuration file are the same.
>
>NetNoise
>  
>

Things I can image that can be the problem are:
* multiple dll.conf config files. Make sure that all dll.conf files on 
your system contain a line with niash
* out-of-date cache for dynamic linking which causes libsane-niash not 
to be found.
Run ldconfig to bring it up-to-date.

Bertrik

Henning Meier-Geinitz wrote:

 >[Cc'ed to sane-devel and the authors of the mentioned backends]
 >
 >Hi developers,
 >
 >Are there any backends that should be included into SANE 1.0.9?
 >
 >TODO currently lists:
 >
 >    - hp4200 http://hp4200-backend.sourceforge.net/
 >    - Genius Vivid Pro USB, Colorado USB 19200, Visioneer OneTouch 7600
 >      IBM IdeaScan 2000 USB
 >      http://homepages.paradise.net.nz/stevenel/scanner/
 >    - Hewlett-Packard ScanJet 3300C / 3400C and 4300C (wip)
 >      http://sourceforge.net/projects/hp3300backend
 >
 >I haven't heard much about the first one for long time now. The viceo
 >backend uses c++ and it's license situation is unclear. I don't think
 >we can include it therefore.
 >
 >I intend to remove both from the TODO list if there aren't any news
 >about them.
 >
 >But we should add a .desc file for the viceo backend so it's listed on
 >the website. I would be happy if the backend author can provide one
 >(see doc/descriptions-external/ and doc/descriptions.txt of current
 >CVS) or at least sends a list of scanners and their status (stable,
 >beta, alpha, untested, unsupported). Thanks.
 >
 >What about the last one (niash backend)?

I'd like to hold off inclusion into Sane for a while longer.

Some things that are not OK yet. For example, all scanners recognised by the
backend are identified to the sane api as a hp3300c. It is currently not
possible
to handle more than one scanner at a time.
Although Hewlett-Packard  ScanJet 3300c and Agfa Snapscan Touch scanners
now seem to work, there are still issues with the HP3400 and HP4300 (not
possible
to set the vertical position of the scan area).
Further I think the sources will need a good cleanup before inclusion
(renaming
functions and files, better comments, uniform coding style etc.) and I
have not
really looked at the backend writing guidelines yet.

Kind regards,
Bertrik

I cannot understand why my hp3300c work if I use the patch in sane1.0.7 but
if I use the same patch in sane1.0.8 it tells me:
$ scanimage > ~testfile
scanimage: no sane device found

$ scanimage -d niash > ~testfile
scanimage: open of device niash failed: invalid argument

I can't understand why, since the configuration file are the same.

NetNoise

Hi all,

Here's another update on recent events and plans for the future.

1) new developers
Two more developers are now part of the hp3300backend project:
Jean Goffinet helped with support for the Agfa Snapscan Touch.
It turns out now that the Agfa Snapscan Touch is very similar
to the HP3300c, except that the scanner head is mounted the
other way around (this makes the software actually easier).
Marcin Cieslak helped with splitting up the various transfer
methods (libusb, scanner kernel module, libieee1284 etc.),
which should make it easier to port stuff to other platforms
among other things.

2) memory corruption bug
I just found quite a serious bug in the current sources. It is
related to how data is juggled around when compensating for the
fact that the red green and blue layers of an image are slightly
offset vertically. This bug seems to be quite old and may explain
some mysterious segfaults people experienced earlier.
Anyway, it's fixed now in the CVS version.

3) contributors
A lot of people have helped since the start of the project by
sending USB snoopy logs, debugging, reporting bugs, sending fixes
or code suggestions. Until now I have gladly accepted this help,
but without paying much attention to giving due credit.
I'll make a list of all who made a significant contribution and
start paying more attention to this in the future.

4) renaming source files
I plan to rename everything with 'hp3300c' in it to 'niash',
since the code is getting less hp3300c-specific with every release.
So for example, the file hp3300c.c may be renamed to niash_core.c
or something like that. And hp3300c_xfer.c to niash_xfer.c.
Also I plan to merge the hp3400.c|h file with the other hp3300c
core code.

5) HP5400/HP5470 and niash00033
It turns out that the HP5400 and HP5470 contain a niash00033 chip.
I did a quick analysis of some USB snoopy logs taken when the
scanner was plugged in. The USB protocol is completely different
from the one used by the HP3300/3400/4300 and I think it's
unlikely that the niash00033 will be supported by out backend.

That's all for now,

All the best
Bertrik Sikken

123

Hi all,

Here's another update on recent events and plans for the future.

1) new developers
Two more developers are now part of the hp3300backend project:
Jean Goffinet helped with support for the Agfa Snapscan Touch.
It turns out now that the Agfa Snapscan Touch is very similar
to the HP3300c, except that the scanner head is mounted the
other way around (this makes the software actually easier).
Marcin Cieslak helped with splitting up the various transfer
methods (libusb, scanner kernel module, libieee1284 etc.),
which should make it easier to port stuff to other platforms
among other things.

2) memory corruption bug
I just found quite a serious bug in the current sources. It is
related to how data is juggled around when compensating for the
fact that the red green and blue layers of an image are slightly
offset vertically. This bug seems to be quite old and may explain
some mysterious segfaults people experienced earlier.
Anyway, it's fixed now in the CVS version.

3) contributors
A lot of people have helped since the start of the project by
sending USB snoopy logs, debugging, reporting bugs, sending fixes
or code suggestions. Until now I have gladly accepted this help,
but without paying much attention to giving due credit.
I'll make a list of all who made a significant contribution and
start paying more attention to this in the future.

4) renaming source files
I plan to rename everything with 'hp3300c' in it to 'niash',
since the code is getting less hp3300c-specific with every release.
So for example, the file hp3300c.c may be renamed to niash_core.c
or something like that. And hp3300c_xfer.c to niash_xfer.c.
Also I plan to merge the hp3400.c|h file with the other hp3300c
core code.

5) HP5400/HP5470 and niash00033
It turns out that the HP5400 and HP5470 contain a niash00033 chip.
I did a quick analysis of some USB snoopy logs taken when the
scanner was plugged in. The USB protocol is completely different
from the one used by the HP3300/3400/4300 and I think it's
unlikely that the niash00033 will be supported by out backend.

That's all for now,

All the best
Bertrik Sikken

Ian Timshel wrote:

>Could someone help make some sense out of this for me..  I'm likely not 
terribly well equiped to be playing with this thing but the scanner is the 
only thing that ties me to windoz... I thought  it time I contribute 
something for myself and others.
>
>       I run a RH 7.2 with the stock RH kernel 2.4.9-31.  I live on kde 
2.2.2  The scanner I have is the hp scanjet 3400c  When I tried to finish the 
install of the testtool I got the following result and I don't know where to 
go from here.  
>
>       Would someone please post a link to some information to help me learn 
to do the problem solving with this please?  I downloaded and installed the 
"libusb version 0.1.4"  
>
>       
>
>[root@localhost testtool]# make all
>gcc -Wall    -c -o main.o main.c
>gcc -Wall    -c -o hp3300c_xfer.o hp3300c_xfer.c
>hp3300c_xfer.c:49:17: usb.h: No such file or directory
>make: *** [hp3300c_xfer.o] Error 1
>  
>
You need libusb 0.1.5 plus the development package.
Even better is downloading libusb sources and compiling it yourself.

Kind regards,
Bertrik Sikken

Ian Timshel wrote:

>Could someone help make some sense out of this for me..  I'm likely not terribly well equiped to be playing with this thing but the scanner is the only thing that ties me to windoz... I thought  it time I contribute something for myself and others.
>
>	I run a RH 7.2 with the stock RH kernel 2.4.9-31.  I live on kde 2.2.2  The scanner I have is the hp scanjet 3400c  When I tried to finish the install of the testtool I got the following result and I don't know where to go from here.  
>
>	Would someone please post a link to some information to help me learn to do the problem solving with this please?  I downloaded and installed the "libusb version 0.1.4"  
>
>	
>
>[root@localhost testtool]# make all
>gcc -Wall    -c -o main.o main.c
>gcc -Wall    -c -o hp3300c_xfer.o hp3300c_xfer.c
>hp3300c_xfer.c:49:17: usb.h: No such file or directory
>make: *** [hp3300c_xfer.o] Error 1
>  
>
You need libusb 0.1.5 plus the development package.
Even better is downloading libusb sources and compiling it yourself.

Kind regards,
Bertrik Sikken

Could someone help make some sense out of this for me..  I'm likely not terribly well equiped to be playing with this thing but the scanner is the only thing that ties me to windoz... I thought  it time I contribute something for myself and others.

	I run a RH 7.2 with the stock RH kernel 2.4.9-31.  I live on kde 2.2.2  The scanner I have is the hp scanjet 3400c  When I tried to finish the install of the testtool I got the following result and I don't know where to go from here.  

	Would someone please post a link to some information to help me learn to do the problem solving with this please?  I downloaded and installed the "libusb version 0.1.4"  

[root@localhost testtool]# make all
gcc -Wall    -c -o main.o main.c
gcc -Wall    -c -o hp3300c_xfer.o hp3300c_xfer.c
hp3300c_xfer.c:49:17: usb.h: No such file or directory
make: *** [hp3300c_xfer.o] Error 1

Hi people,

Here's what happened since my last e-mail to this list.

Bertrik Sikken wrote:
> First, as many of you probably already know, there appear to me more
> than just two niash chips (the niash00014 for the hp3300c and the
> niash00019 for the hp3400c/4300c). I've had reports of hp3300c
> scanners with niash00012 and niash00013 chips. Apparently the
> hp5400 is even using a niash00033.
> In the past, support for these different chips has been added as
> lots of little kludges, mostly on how to initialise the chip
> before starting to read the image data. My plan is now to
> separate the specifics for each chipset in a separate function
> (perhaps even a separate file) for each. This should make it easier
> to make changes / add features / fix bugs for a specific chip,
> without breaking the code for the other chips.

The code is now a little bit more organised now, by separating out
the niash00019 specific code. I tried to avoid duplicating code.
It's in CVS. If there are no major bugs, I think I'll make another
release soon.

During the process of reorganising I found out some new stuff about
the register that determines how much data is allowed to be stored
in the scanner's internal buffer (too little means a lot of
backtracking, too much results in sideways shifted images).
This new knowledge is now also in the sources in CVS. It affects
only HP3300c scanners.

For the HP3400/4300 scanners I added some experimental code that
should make it possible to also scan at 150 and 600 dpi.
It uses a trick with a register that makes the scanner skip every
other line, pointed out by Lev Shamardin.
I have not been able to test this (I don't have such a scanner).

Finally you may have noticed that sane 1.0.8 was just released
this week. Fortunately Ricardo Martin's patch-sane.sh script
still does its magic.

All the best,
Bertrik

Hi all,

Here's an update on what's going on.
(This list can definitely use some traffic, so here is some :P )

First, as many of you probably already know, there appear to me more
than just two niash chips (the niash00014 for the hp3300c and the
niash00019 for the hp3400c/4300c). I've had reports of hp3300c
scanners with niash00012 and niash00013 chips. Apparently the
hp5400 is even using a niash00033.
In the past, support for these different chips has been added as
lots of little kludges, mostly on how to initialise the chip
before starting to read the image data. My plan is now to
separate the specifics for each chipset in a separate function
(perhaps even a separate file) for each. This should make it easier
to make changes / add features / fix bugs for a specific chip,
without breaking the code for the other chips.

There are now 5 developers registered with the hp3300backend
project at sourceforge. Here's a short introduction
(please correct me if I'm wrong about something):
Ricardo Martins: has helped me greatly fixing lots of things,
wrote the patch-script, cleaned up the build system (makefile etc.)
wrote documentation and is investigating support for the *BSD OSs.
Brian Beattie has recently joined, has experience with figuring
out protocols and is looking into the HP4300.
Robert Jones has looked into parallel port scanning with the
Hp3400.
Frederic Leuba is looking at HP4300 specifics.

BTW, I have not heard anything lately from HP regarding Linux
support for scanner with niash chipsets.

All the best,
Bertrik

Hi!

I have just downloaded and compiled testtool from sourceforge.
It works quite nice with usblib, but there are some things, that I would
like to fix (If I knew how) ;)
- the coordinates where it starts scanning are off (too much to the left
  and up
- the scan does not scan the whole A4 (DIN) area, but only less than
  halve of it - can be fixed with a change in parse_args to this:
  int parse_args (int argc, char **argv, struct arguments *arguments ){
    int width = 1325;             /* default value */
    int height = 3500;            /* default value */ /* <-- changed */
  ...

- and the other things like setting resolution - but I know that's on
  the buglist.

I thank all of the developers for this fine work, so I can finally ditch
that pesci WinDOS.
And I would be really gratefull if someone could please either tell me
how to fix the first two problems or (preferably - since I lack the
coding experience) fix it themselves.

Thank you in advance!

Regards/Lep pozdrav

Boštjan Müller
-- 
[*] Boštjan Müller - neo...@ne... - http://neonatus.net/~neonatus [*]
[*]    GPG/PGP key -> finger: neo...@ne..., DSA id: 0x9B2FF108    [*]
[*] Celular: +386(0)41243189, Powered by Debian GNU/LiNUX - ICQ #:7506644  [*]
            Give a man a fish and you feed him for a day;
     teach him to use the Net and he won't bother you for weeks.

Hi Brian,

Brian Beattie wrote:
> I was setting up my scanner under windows and I can not locate the
> software CD that can with my scanner.  Anybody got one they woudl be
> willing to copy or loan me?  I can burn my own, so the raw bits over the
> network would work as well.

I don't have the CD but I do have a driver package from HP
(sj565en.exe). It's about 30 MB and last time I looked they
still offer it for download.

I tried these drivers once on my HP3300c (by editing the .inf file
I can make Windows think that my HP3300c is a HP3400c) and they
actually did something. They turned on the scanner lamp and
seemed to do some kind of simple calibration, then the carriage
went of to the bottom clicking loudly as it reached the end.

Bertrik

Just to clarify, I'm doing this to run snoopy to see if I can figure out
the initialization and configuration stuff.

On Sat, 2002-04-27 at 09:37, Brian Beattie wrote:
> I was setting up my scanner under windows and I can not locate the
> software CD that can with my scanner.  Anybody got one they woudl be
> willing to copy or loan me?  I can burn my own, so the raw bits over the
> network would work as well.
> -- 
> Brian Beattie            | This email was produced using professional
> quality,
> be...@be... | standards based software.  Users of Microsoft
>                          | products or other substandard software should
> www.beattie-home.net     | contact the author about receiving a Free
> upgrade to
>                          | FreeBSD or Linux.   "FreeBSD: The power to
> serve"
> 
> 
> _______________________________________________
> Hp3300backend-devel mailing list
> Hp3...@li...
> https://lists.sourceforge.net/lists/listinfo/hp3300backend-devel
-- 
Brian Beattie            | This email was produced using professional
quality,
be...@be... | standards based software.  Users of Microsoft
                         | products or other substandard software should
www.beattie-home.net     | contact the author about receiving a Free
upgrade to
                         | FreeBSD or Linux.   "FreeBSD: The power to
serve"

I was setting up my scanner under windows and I can not locate the
software CD that can with my scanner.  Anybody got one they woudl be
willing to copy or loan me?  I can burn my own, so the raw bits over the
network would work as well.
-- 
Brian Beattie            | This email was produced using professional
quality,
be...@be... | standards based software.  Users of Microsoft
                         | products or other substandard software should
www.beattie-home.net     | contact the author about receiving a Free
upgrade to
                         | FreeBSD or Linux.   "FreeBSD: The power to
serve"

Hi Brian,

Brian Beattie wrote:
>      I've gor an HP4300C and would like to know if anbody is curently
> working of getting this working.  Anything I can do to help.  I'm a
> decent kernel hacker and have done som ework on reverse engineering
> protocols (e.g. the dimera driver for gphoto).  Anyway I'd really like
> to get this thing working fully.

You're very welcome to help with HP3400C support.

Although many people sent me USB logs (or offered to do so), there
is currently no-one actively working on support for this scanner.
It would be great if you could do some investigation on it.

Please tell me your sourceforge id (beattie I suppose) and I'll
add you as a developer to the projects as soon as possible.

Bertrik

Hi,
     I've gor an HP4300C and would like to know if anbody is curently
working of getting this working.  Anything I can do to help.  I'm a
decent kernel hacker and have done som ework on reverse engineering
protocols (e.g. the dimera driver for gphoto).  Anyway I'd really like
to get this thing working fully.

Hi Dave,

Dave Pilar wrote:
> 
> Thanks for the tremendous efforts of everyone involved!
> 
> Unfortunately, I haven't been able to get the head to move.  I can toggle the
> lamp using either the scanner module or the libusb methods, but the head
> doesn't move during scans, it hums a little.  The scanner is fully functional
> under Win98.  My 3300C seems to have a slightly different chipset:
> 
> 1x: NIASH00013
>      F69A5-000
>      9928
> 1x: IS61C1024-15JT
>      JT471941L 9917
> 2x: UT611024  JC-15
>      9928
> 1x: NIPEDC0804

The key to the symptoms you experience is indeed the version of
the niash chip (mine has a niash00014 and there are even hp3300
scanners with a niash00012 chip).

The niash00013 seems to control the stepper motor a little
differently: with the same settings is runs twice as fast
compared to a niash00014 chip. Also the gamma tables are different
and some unknown registers are written with different values.

I have a quick-and-dirty version of the sources that works on hp3300
scanners with a niash00012 chip. This has been reported to work with
niash00013 chips too, although I had one negative report.
I'm looking into integrating the modifications of this modified
version into the main source code.

Bertrik

Thanks for the tremendous efforts of everyone involved!

Unfortunately, I haven't been able to get the head to move.  I can toggle the 
lamp using either the scanner module or the libusb methods, but the head 
doesn't move during scans, it hums a little.  The scanner is fully functional 
under Win98.  My 3300C seems to have a slightly different chipset:

1x: NIASH00013
     F69A5-000
     9928
1x: IS61C1024-15JT
     JT471941L 9917
2x: UT611024  JC-15
     9928
1x: NIPEDC0804

I am running SuSE 7.3 with kernel 2.4.16 (I also tried Mandrake 8.1 with 
2.4.13) on an AMD-K6, 350Mhz.
below I included some testool results:

with -l:

Initializing scanner
Bus 001 Dev 003 : 0x03F0-0x0205
Matching USB device 0x03F0-0x0205 ... found HP3300C
Toggling scanner lamp

-c seems to work (makes noise) I had let the lamp warm up first:

Initializing scanner
Bus 001 Dev 003 : 0x03F0-0x0205
Matching USB device 0x03F0-0x0205 ... found HP3300C
Creating calibration file
_iBytesPerLine = 3975
_iLinesPerCircBuf = 3
Allocated 11925 bytes for circular buffer
_iLinesPerXferBuf = 15
Xfer block size = 59625
buffer level =  24, <reading 59625 bytes>, buffer level =  28
Black level 38 142 45, White level 38 136 43

second run:
Initializing scanner
Bus 001 Dev 003 : 0x03F0-0x0205
Matching USB device 0x03F0-0x0205 ... found HP3300C
Creating calibration file
_iBytesPerLine = 3975
_iLinesPerCircBuf = 3
Allocated 11925 bytes for circular buffer
_iLinesPerXferBuf = 15
Xfer block size = 59625
buffer level =  22, <reading 59625 bytes>, buffer level =  25
Black level 39 138 45, White level 38 137 42

benchmark:

Initializing scanner
Bus 001 Dev 003 : 0x03F0-0x0205
Matching USB device 0x03F0-0x0205 ... found HP3300C
Benchmarking: 6000 kB...
Elapsed time (real time): 6.9800 seconds
c1048918-a:/tmp/scanner/testtool #

then -p

Initializing scanner
Bus 001 Dev 003 : 0x03F0-0x0205
Matching USB device 0x03F0-0x0205 ... found HP3300C
Showing all registers

0000 04 00 00 19 07 00 00 00
0008 00 00 00 00 00 00 00 00
0010 00 00 00 00 00 00 00 00
0018 00 00 00 00 00 00 00 00
0020 00 00 00 00 00 00 00 00
0028 00 00 00 00 00 00 00 00
0030 00 00 00 00 00 00 00 00
0038 00 00 00 00 00 00 00 00

scanning with -v -s 300,300 | display: (I've tried many other combos)

Initializing scanner
Bus 001 Dev 003 : 0x03F0-0x0205
Matching USB device 0x03F0-0x0205 ... found HP3300C
Scanning area (300, 300) to stdout
Using calibration file
_iBytesPerLine = 900
_iLinesPerCircBuf = 3
Allocated 2700 bytes for circular buffer
_iLinesPerXferBuf = 68
Xfer block size = 61200
buffer level =  84, <reading 61200 bytes>, buffer level =  34

produces a green garbled fuzz/snow with brighter vertical lines--but, as I 
said, the scanner head did not move under the picture I was trying to scan.

Thanks again for all your work!

On Sat, Jan 19, 2002 at 01:01:36PM +0100, Bertrik Sikken wrote:
> Joshua N Pritikin wrote:
> > On Sat, Jan 19, 2002 at 04:04:20PM +0530, Joshua N Pritikin wrote:
> > > i am using the testtool from CVS with the kernel usb scanner
> > > driver (2.4.16).
> > >
> > > It seems like i had to run "./testtool calib2" twice.  The first scan
> > > i did after calibration had horrible artifacts.  After re-calibration
> > > the colors was accurate and the scan clear.
> 
> I think this is because the lamp had not warmed up completely
> during the first calibration run. This causes the software to
> set the contrast too high, possibly resulting in an overflow
> in the pixel calculation. You probably saw some vertical stripes
> consisting of one of the primary colors (especially on the sides
> of the scan), right?

Yes.

> > 300,300 and 600,300 gave the same result -- horizontal 150dpi, vertical 300dpi
> > 300,600 and 600,600 gave the same result -- horizontal 150dpi, vertical 600dpi
> 
> Thanks a lot for the info.
> 
> You are referring to the numbers given to the two calls to the
> _ConvertMotorTable routine, I suppose? (Just asking to be sure)

Yes.

i'll try to answer your other questions later.

-- 
Victory to the Divine Mother!!         after all,
  http://sahajayoga.org                  http://why-compete.org

Hi Joshua,

Joshua N Pritikin wrote:
> On Sat, Jan 19, 2002 at 04:04:20PM +0530, Joshua N Pritikin wrote:
> > i am using the testtool from CVS with the kernel usb scanner
> > driver (2.4.16).
> >
> > It seems like i had to run "./testtool calib2" twice.  The first scan
> > i did after calibration had horrible artifacts.  After re-calibration
> > the colors was accurate and the scan clear.

I think this is because the lamp had not warmed up completely
during the first calibration run. This causes the software to
set the contrast too high, possibly resulting in an overflow
in the pixel calculation. You probably saw some vertical stripes
consisting of one of the primary colors (especially on the sides
of the scan), right?

> > 300,300 and 600,300 gave the same result -- horizontal 300dpi, vertical 150dpi
> > 300,600 and 600,600 gave the same result -- horizontal 600dpi, vertical 150dpi
> 
> Oh sorry, i got confused between horz and vert.  Here are the corrected
> results:
> 
> 300,300 and 600,300 gave the same result -- horizontal 150dpi, vertical 300dpi
> 300,600 and 600,600 gave the same result -- horizontal 150dpi, vertical 600dpi

Thanks a lot for the info.

You are referring to the numbers given to the two calls to the
_ConvertMotorTable routine, I suppose? (Just asking to be sure)

Your numbers seem to reinforce the idea that the second motor
table determines motor speed during the scan (and thus the vertical
resolution, since the exposure time for one line is fixed).

The suspicion is that the first table determines motor speed when
the scanner is not scanning (for example when moving the scanner
head to the top of the scan area or when moving it back to the
home position).
Can you hear any difference in motor sound when comparing the
300,300 and the 600,300 case?

> 300,150 gave an unpleasant noise so i aborted the scan.

Ok, 150 lpi probably will need to be emulated for now I think
(by averaging two 300 lpi lines for example).
I guess the motor accelerates too quickly for this resolution,
causing it to jam (it attempts to do 1234 steps/second !).
On my HP3300 I have noticed that writing the value 0x01 to
register 0x06 causes the scanner to skip every other line,
thereby halving the resolution. This may also be used to get
down to 150 lpi on the HP3400/4300.

Regards,
Bertrik

On Sat, Jan 19, 2002 at 04:04:20PM +0530, Joshua N Pritikin wrote:
> i am using the testtool from CVS with the kernel usb scanner
> driver (2.4.16).
> 
> It seems like i had to run "./testtool calib2" twice.  The first scan
> i did after calibration had horrible artifacts.  After re-calibration
> the colors was accurate and the scan clear.
> 
> 300,300 and 600,300 gave the same result -- horizontal 300dpi, vertical 150dpi
> 300,600 and 600,600 gave the same result -- horizontal 600dpi, vertical 150dpi

Oh sorry, i got confused between horz and vert.  Here are the corrected
results:

300,300 and 600,300 gave the same result -- horizontal 150dpi, vertical 300dpi
300,600 and 600,600 gave the same result -- horizontal 150dpi, vertical 600dpi

-- 
Victory to the Divine Mother!!         after all,
  http://sahajayoga.org                  http://why-compete.org

i am using the testtool from CVS with the kernel usb scanner
driver (2.4.16).

It seems like i had to run "./testtool calib2" twice.  The first scan
i did after calibration had horrible artifacts.  After re-calibration
the colors was accurate and the scan clear.

300,300 and 600,300 gave the same result -- horizontal 300dpi, vertical 150dpi
300,600 and 600,600 gave the same result -- horizontal 600dpi, vertical 150dpi

300,150 gave an unpleasant noise so i aborted the scan.

Any clue how to change the vertical DPI?

-- 
Victory to the Divine Mother!!         after all,
  http://sahajayoga.org                  http://why-compete.org

2001	Jan	Feb	Mar	Apr	May	Jun	Jul	Aug	Sep	Oct	Nov (2)	Dec (10)
2002	Jan (5)	Feb (1)	Mar (3)	Apr (3)	May (2)	Jun (4)	Jul (3)	Aug	Sep (4)	Oct	Nov (5)	Dec
2003	Jan (2)	Feb	Mar	Apr (4)	May (3)	Jun (5)	Jul	Aug (3)	Sep (1)	Oct (2)	Nov (3)	Dec (6)
2004	Jan (1)	Feb	Mar (7)	Apr	May	Jun	Jul	Aug (1)	Sep (6)	Oct	Nov (1)	Dec (1)

hp3300backend-devel Mailing List for HP3300c SANE backend (Page 3)

hp3300backend-devel — Development of SANE backend for HP ScanJet 3300/3400/4300 scanners