canonscanner-devel

[Canonscanner-devel] Re: Good news from N656U (R N dev)

[Albert P. <ap...@ch...>]

> Hi guys,
> Now xsane starts! the scanner has been found!
> the problem was on the SCANNER_IOCTL_VENDOR and
> SCANNER_IOCTL_PRODUCT definitions for ioctl.
> The correct one is the second one. Now I have to
> find the reason why with xsane the device results
> always busy and scanimage says:
>
> [angelo@naspa angelo]$ scanimage
> scanimage: open of device canon_usb:/dev/usb/scanner0
> failed: Invalid argument
>
Try it as user root, you probably don't have the rights to use it as a no=
rmal=20
user. Try (as root): chmod 666 /dev/usb/scanner0

Albert

[Canonscanner-devel] Good news from N656U

[R N d. <ran...@ya...>]

Hi guys, 
Now xsane starts! the scanner has been found!
the problem was on the SCANNER_IOCTL_VENDOR and
SCANNER_IOCTL_PRODUCT definitions for ioctl. 
The correct one is the second one. Now I have to
find the reason why with xsane the device results
always busy and scanimage says:

[angelo@naspa angelo]$ scanimage
scanimage: open of device canon_usb:/dev/usb/scanner0
failed: Invalid argument

I'm going to try tomorrow. 

If you have any suggestion it's welcome! :-)

Thanks 
Angelo



__________________________________________________
Do you Yahoo!?
New DSL Internet Access from SBC & Yahoo!
http://sbc.yahoo.com

Re: [Canonscanner-devel] N656U and sane-backend problem

[R N d. <ran...@ya...>]

Hi, I'd like to add new info on my problem, i.e.
the ct1220 works, so the problem might be on only 
the sane integration...

--- R N dev <ran...@ya...> wrote:
> Hi guys,
> I added into canon_usb.c the device data for
> my N656U, I rebuilt all (also into rpm for
> Mandrake 8.2) but it seems not to work.
> If i compile and install the result is a an 
> error of Segmentation fault when Xsane starts.
> And if i rebuild the rpm the result is that no
> device
> is found. (modprobe scanner is done and the device
> is at /dev/usb/scanner0).
> 
> Any help?
> 
> thanks 
> Angelo
> 
> Ps i wanted to post the diff file for my
> scanner bu till it's not working...
> However the scanner data are
> vendor Id 0x04A9
> Product Id 0x2206
> Revision Id 0x0100
> I think the are the same as N650U
> 
> 
> 
> __________________________________________________
> Do you Yahoo!?
> New DSL Internet Access from SBC & Yahoo!
> http://sbc.yahoo.com
> 
> 
>
-------------------------------------------------------
> This sf.net email is sponsored by:ThinkGeek
> Welcome to geek heaven.
> http://thinkgeek.com/sf
> _______________________________________________
> Canonscanner-devel mailing list
> Can...@li...
>
https://lists.sourceforge.net/lists/listinfo/canonscanner-devel


__________________________________________________
Do you Yahoo!?
New DSL Internet Access from SBC & Yahoo!
http://sbc.yahoo.com

[Canonscanner-devel] Narrowing down the scanners?

[Albert P. <ap...@ch...>]

It may be interesting to narrow down the supported scanners?

The following have an LM9832 chip, CIS scansystem, 1 button, 3x14 bits co=
lor

650/656U, id=3D0x2206, 600x1200
1220U, id=3D0x2207, 1200x2400

The following have an LM9833 chip, CIS scansystem, 3 buttons, 3x16 bits c=
olor:

670/676U, id=3D0x220d, 600x1200
1240U, id=3D0x220e, 1200x2400

The 9832 and 9833 are pin and software compatible (except for the 14/16 b=
its=20
resolution).=20

So the different parameters are:
- resolution (600x1200, 1200x2400)
- color definition (14/16 bits, mainly used for calibration)
- 1 or 3 buttons

There are also a few other canon scanners, 646U, 660U, 1250U2, but

the 1250U2 has USB 2.0, 1200x2400, 48 bits and a CRT (a real lamp),so it =
must=20
have an other chip than 9832/9833 (at least for the USB 2.0 stuff).

the 660U has 600x1200, 42 bits and a CRT

the 646 has 600x1200, 42 bits and a CCD (so no CIS element)

I think the best would be to concentrate on the 650/656/1220/670/676/1240=
=20
range CIS scanners.

Just my two cents,

Albert

[Canonscanner-devel] N656U and sane-backend problem

[R N d. <ran...@ya...>]

Hi guys,
I added into canon_usb.c the device data for
my N656U, I rebuilt all (also into rpm for
Mandrake 8.2) but it seems not to work.
If i compile and install the result is a an 
error of Segmentation fault when Xsane starts.
And if i rebuild the rpm the result is that no device
is found. (modprobe scanner is done and the device
is at /dev/usb/scanner0).

Any help?

thanks 
Angelo

Ps i wanted to post the diff file for my
scanner bu till it's not working...
However the scanner data are
vendor Id 0x04A9
Product Id 0x2206
Revision Id 0x0100
I think the are the same as N650U



__________________________________________________
Do you Yahoo!?
New DSL Internet Access from SBC & Yahoo!
http://sbc.yahoo.com

[Canonscanner-devel] Read configuration file

[Christian <c.r...@ro...>]

Hi,

I have written the configurationfile part in sane_get_devices(...).

The configfile is named "canonusb.conf". It may contain comments beginnin=
g=20
with # and blank lines. Example:

=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
# canonusb.conf
#

/dev/usbscanner
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D

Hope, it is okay. I wanted to do something productive to the project inst=
ead=20
of always posting problems ;-)

Please give me a feedback

Regards

Christian

[Canonscanner-devel] More calibration thinking

[Albert P. <ap...@ch...>]

The idea of calibration didn't let me go. Reading through the documentati=
on,
including flowdiagrams!, I came up with the following idea:

Start scanning in little steps from the very start until you'll find the=20
calibration strip. It should be a black and a white one to get the black=20
level and white level. Get the black level (which gives the noise level
in the lowest bits!) from the black strip for every color, and get the wh=
ite=20
levels for every color with several intensities to get the maximum levels=
 and=20
a sort of gamma (if I understand the concept of gamma well).

One other thing that the example code does is reading a plunk of scandata=
 in
(what it actually is is not important) and figure out the optimum for the=
 data=20
stream through the USB bus (i.e. the settings for motor stepping etc). Th=
is=20
is probably why the Windows driver only stalls scanning on the highest DP=
I=20
settings, but the transport runs smoothly in lower scans. I haven't given=
=20
this a thoroug thought though.

Food for thought,

Albert

Re: [Canonscanner-devel] Status report

[R N d. <ran...@ya...>]

How stupid i am! :-)
Sorry i downloaded the wrong project!
Now i can rebuild the project again. 
I'm going to read as soon as
possible all readme files and try with my scanner.

Thank you (and sorry)
Angelo

 
> --- R N dev <ran...@ya...> wrote:
> > Here are the data of my system:
> > Mdk 8.2 with libusb0.1_4 (mdk)
> > 
> > and this is the error message:
> > 
> > [angelo@naspa ucanoscan]$ gmake
> > usbio-ugen.c:28:25: dev/usb/usb.h: No such file or
> > directory
> > usbio-ugen.c:28:25: dev/usb/usb.h: No such file or
> > directory
> > gcc -c -o ucanoscan.o -g -Wall -DDEBUG_DATASAVE
> > -DUGEN_DRIVER -I/usr/local/include ucanoscan.c
> > gcc -c -o usbio-ugen.o -g -Wall -DDEBUG_DATASAVE
> > -DUGEN_DRIVER -I/usr/local/include usbio-ugen.c
> > usbio-ugen.c:28:25: dev/usb/usb.h: No such file or
> > directory
> > gmake: *** [usbio-ugen.o] Error 1
> > [angelo@naspa ucanoscan]$
> > 
> > Trying gmake UCANOSCAN_DRIVER=1
> > 
> > [angelo@naspa ucanoscan]$ gmake UCANOSCAN_DRIVER=1
> > usbio-ucanoscan.c:28:25: dev/usb/usb.h: No such
> file
> > or directory
> > usbio-ucanoscan.c:28:25: dev/usb/usb.h: No such
> file
> > or directory
> > gcc -c -o ucanoscan.o -g -Wall -DDEBUG_DATASAVE
> > -DUCANOSCAN_DRIVER -I/usr/local/include
> ucanoscan.c
> > gcc -c -o usbio-ucanoscan.o -g -Wall
> > -DDEBUG_DATASAVE
> > -DUCANOSCAN_DRIVER -I/usr/local/include
> > usbio-ucanoscan.c
> > usbio-ucanoscan.c:28:25: dev/usb/usb.h: No such
> file
> > or directory
> > gmake: *** [usbio-ucanoscan.o] Error 1
> > [angelo@naspa ucanoscan]$
> > 
> > What's wrong in it?
> > 
> > thanks 
> > Angelo
> > 
> > --- R N dev <ran...@ya...> wrote:
> > > Hi Albert,
> > > I have the N656U, i'm sorry but i can't do any
> > > test since i can't compile the sources. 
> > > I've tryed the old code with Mandrake 8.1 and 
> > > old usblib (i cannot remember the version and
> i'm
> > > not at home now to give you more info) and it
> was
> > > good. Now i've MDK 8.2 and i cannot compile the
> > > source the error is on "dev/usb/usb.h" that
> cannot
> > > be
> > > included (i don't have it, i have usb.h under 
> > > /usr/include, but trying to use that one seems
> to
> > > give more errors...) I'm quite convince myself
> to 
> > > download the MDK 9.0 rc3 to try if something
> goes
> > > better.... but at this moment i've not so much
> > free 
> > > time. Hope somebody had the same problem to
> solve
> > > the compiling problem so i can do some tests on
> my
> > > scanner...
> > > Anyway i should be a little bit free next week,
> > > could
> > > it be too late?
> > > 
> > > Angelo
> > > 
> > > 
> > > --- Albert Pauw <a....@wa...> wrote:
> > > > After following very hopefull messages on this
> > > list
> > > > I was wondering
> > > > what the actual status is. Maybe that would be
> > > > handy. Also
> > > > which scanners are actually supported (or
> > working)
> > > > or should
> > > > be working.
> > > > 
> > > > There seems to be differences between the
> > > different
> > > > models:
> > > > 
> > > > 42 bits (14 bits per color):
> > > > 
> > > > N650U/N656U (600x1200 dpi)
> > > > N1220U (1200x2400 dpi)
> > > > 
> > > > 48 bits (16 bits per color):
> > > > 
> > > > N670U/N676U (600x1200 dpi)
> > > > N1240U (1200x2400 dpi)
> > > > 
> > > > How are all these supported?
> > > > 
> > > > The 656 and 676 seem to be identical to the
> 650
> > > and
> > > > 670 model, resp.
> > > > Only difference is the metallic finish
> (although
> > > > it's plastic).
> > > > 
> > > > The software of Jason Millward works very well
> > for
> > > > my N650U scanner,
> > > > although the scanner wants to go a bit further
> > > than
> > > > it can (which makes
> > > > an uneasy noise, but it's just a little bit).
> > > > 
> > > > Could someone give some sort of a status
> report?
> > > > 
> > > > That would be very interesting.
> > > > 
> > > > Keep up the good work!
> > > > 
> > > > Albert
> > > > 
> > > > 
> > > > 
> > > > 
> > > > 
> > > >
> > >
> >
>
-------------------------------------------------------
> > > > This sf.net email is sponsored by:ThinkGeek
> > > > Welcome to geek heaven.
> > > > http://thinkgeek.com/sf
> > > >
> _______________________________________________
> > > > Canonscanner-devel mailing list
> > > > Can...@li...
> > > >
> > >
> >
>
https://lists.sourceforge.net/lists/listinfo/canonscanner-devel
> > > 
> > > 
> > >
> __________________________________________________
> > > Do you Yahoo!?
> > > New DSL Internet Access from SBC & Yahoo!
> > > http://sbc.yahoo.com
> > > 
> > > 
> > >
> >
>
-------------------------------------------------------
> > > This sf.net email is sponsored by:ThinkGeek
> > > Welcome to geek heaven.
> > > http://thinkgeek.com/sf
> > > _______________________________________________
> > > Canonscanner-devel mailing list
> > > Can...@li...
> > >
> >
>
https://lists.sourceforge.net/lists/listinfo/canonscanner-devel
> > 
> > __________________________________________________
> > Do you Yahoo!?
> > New DSL Internet Access from SBC & Yahoo!
> > http://sbc.yahoo.com
> > 
> > 
> >
>
-------------------------------------------------------
> > This sf.net email is sponsored by:ThinkGeek
> > Welcome to geek heaven.
> > http://thinkgeek.com/sf
> > _______________________________________________
> > Canonscanner-devel mailing list
> > Can...@li...
> >
>
https://lists.sourceforge.net/lists/listinfo/canonscanner-devel
> 
> 
> __________________________________________________
> Do you Yahoo!?
> New DSL Internet Access from SBC & Yahoo!
> http://sbc.yahoo.com
> 


__________________________________________________
Do you Yahoo!?
New DSL Internet Access from SBC & Yahoo!
http://sbc.yahoo.com

[Canonscanner-devel] Fwd: Re: Canon USB Sane module: calibration

[Johann.Philipp <joh...@we...>]

Hi,

I just had a quick look on the geocities.com site you mentioned, Albert. It 
looks like its quite cool stuff in the "Software Engineers' section"...

Johann

----------  Weitergeleitete Nachricht  ----------

Subject: Re: Canon USB Sane module: calibration
Date: Tue, 24 Sep 2002 18:05:49 +0200
From: Albert Pauw <ap...@ch...>
To: "Johann.Philipp" <joh...@we...>

Hi Johann,

I have been digging into the data sheet of the LM9832 and
found some things which could be related to your noise picture.

Section 10.2 Soft Reset

A Soft Reset is generated by setting bit 5 of register 07. A Soft
Reset:

(left some bits out)

- Stops DRAM refresh. This will corrupt all the gamma, offset,
gain values, as well as any image data, stored in the external
DRAM.

Then a procedure to prevent this follows.

I also found the site: http://www.geocities.com/SiliconValley/Mouse/5403/

Which contains quite a lot of interesting materials (if you've not found it
already) for programmers (which I am not for a long time anymore).

In the LM9831 Software Designers Guide (it's applicable to the 983x family):

In table 1 they talk about a calibration strip, could this be the thing I was
suggesting in my previous e-mail, just under the bottom border?

Bye for now,

Albert

-------------------------------------------------------

[Canonscanner-devel] hints

[Johann.Philipp <joh...@we...>]

Hi everyone,

thanks for your information, Albert. I think you inspired me for a solution 
for the snowy picture / false color / no action problem when doing the first 
scan after a boot. Try this:
in the soft_reset(...) function put the following line _*before*_ any other 
command:
	reg_w(c_h->fd, REG_SIGNAL_SAMPLE_POSITION, 5);
I don't know, why this would stop the strange effects, but some quick tests 
gave good results for me.

Anyway, in the next few weeks, possibly even months, I will be away from home 
quite a lot, so don't worry, if I don't reply immediately. It doesn't mean, 
that I'm not interested anymore. Just don't stop trying and reading, 
everything can be done!

Johann

[Canonscanner-devel] Re: Canon USB Sane module

[Johann.Philipp <joh...@we...>]

Hi Albert,

no, I don't have an idea, what the scanner does when being plugged into a 
Windows machine. When doing the research with the usbsniffer in the 
beginning, I concentrated more or less on the initializing procedure, but 
didn't follow all the actual data that came from the scanner (the sniffer I 
used then couldn't catch all of it). One thing is sure, the replugging action 
is initiated by the windows driver software. But my TWAIN-software has a 
special calibration button. Pressing this, it takes a lot more time than the 
replugging action. So I think, either when replugging the software makes a 
short functional test, or it validates the somewhere stored calibration data 
with a few samples.
This is, how I think the calibration works: on the glas plate is glued a 
narrow, colored strip for scanning a reference color. I took once my scanner 
apart and I think there was such a small strip in white. The software scans 
these few lines with different light colors and changing the offset and gain 
coefficients, until black and white are OK. The rest is gamma correction.
I'm no expert in optical sciences and such, but this will be my first try...
To get to your last idea: I think soon we need to contact the sane community. 
Even without calibration, this has to my opinion already alpha status. But on 
the other hand, don't let us hurry this, I like to get some more to the snowy 
picture thing and other probably related problems.
For the Canon people: at least I have never asked them....

Bye for now
Johann

> Hi Johann,
>
> > that sounds well, I will add the N650U support in the next release, but
> > tell me, does it have also a 600pdi sensor?
>
> Yes, I think so. It is said to be a 600x1200 dpi scanner (the 656 is
> identical actually). So the optical resolution is 600, only the stepping
> motor can do better.
>
> > I just read through the LM9832 datasheet. There it says, that the modes
> > with higher color depths are only used for accurate gain and offset
> > calibration. Calibration is not done yet by the code, so the maximum
> > color depth is not so important yet. I believe it will anyway have no
> > effect on the code at all, because 14 bit pixel data is send by the
> > scanner as 2 bytes, same as 16 bit data.
>
> In that case it indeed doesn't make a difference for the code. Have you any
> idea what the scanner actually does when you plug it into an Windows
> machine? It does some stepping and goes through the colors of the leds, is
> that the calibration you mean?
>
> Do you have any idea when an alpha/beta release can be done? For such
> a nice and small scanner it's a shame that Canon doesn't cooperate, that
> would make things much easier.
>
> Bye for now,
>
> Albert

Re: [Canonscanner-devel] Status report

[Johann.Philipp <joh...@we...>]

Hi Angelo,

I was wondering, why the sourcecode has a file called "ucanoscan.c". I think 
I remember, that from the beginning there was a module "ucanoscan" on cvs, 
but it was for freeBSD... But: My sane code is in the cvs-module "canon", the 
same as all the ct_.... test programs are in. The files for the sane code are 
named "canon_usb.....". Please download these files, especially the 
"README.canon_usb", which contains some installation instructions.
I hope, this helps.

Greetings
Johann

> Here are the data of my system:
> Mdk 8.2 with libusb0.1_4 (mdk)
>
> and this is the error message:
>
> [angelo@naspa ucanoscan]$ gmake
> usbio-ugen.c:28:25: dev/usb/usb.h: No such file or
> directory
> usbio-ugen.c:28:25: dev/usb/usb.h: No such file or
> directory
> gcc -c -o ucanoscan.o -g -Wall -DDEBUG_DATASAVE
> -DUGEN_DRIVER -I/usr/local/include ucanoscan.c
> gcc -c -o usbio-ugen.o -g -Wall -DDEBUG_DATASAVE
> -DUGEN_DRIVER -I/usr/local/include usbio-ugen.c
> usbio-ugen.c:28:25: dev/usb/usb.h: No such file or
> directory
> gmake: *** [usbio-ugen.o] Error 1
> [angelo@naspa ucanoscan]$
>
> Trying gmake UCANOSCAN_DRIVER=1
>
> [angelo@naspa ucanoscan]$ gmake UCANOSCAN_DRIVER=1
> usbio-ucanoscan.c:28:25: dev/usb/usb.h: No such file
> or directory
> usbio-ucanoscan.c:28:25: dev/usb/usb.h: No such file
> or directory
> gcc -c -o ucanoscan.o -g -Wall -DDEBUG_DATASAVE
> -DUCANOSCAN_DRIVER -I/usr/local/include ucanoscan.c
> gcc -c -o usbio-ucanoscan.o -g -Wall -DDEBUG_DATASAVE
> -DUCANOSCAN_DRIVER -I/usr/local/include
> usbio-ucanoscan.c
> usbio-ucanoscan.c:28:25: dev/usb/usb.h: No such file
> or directory
> gmake: *** [usbio-ucanoscan.o] Error 1
> [angelo@naspa ucanoscan]$
>
> What's wrong in it?
>
> thanks
> Angelo
>
> --- R N dev <ran...@ya...> wrote:
> > Hi Albert,
> > I have the N656U, i'm sorry but i can't do any
> > test since i can't compile the sources.
> > I've tryed the old code with Mandrake 8.1 and
> > old usblib (i cannot remember the version and i'm
> > not at home now to give you more info) and it was
> > good. Now i've MDK 8.2 and i cannot compile the
> > source the error is on "dev/usb/usb.h" that cannot
> > be
> > included (i don't have it, i have usb.h under
> > /usr/include, but trying to use that one seems to
> > give more errors...) I'm quite convince myself to
> > download the MDK 9.0 rc3 to try if something goes
> > better.... but at this moment i've not so much free
> > time. Hope somebody had the same problem to solve
> > the compiling problem so i can do some tests on my
> > scanner...
> > Anyway i should be a little bit free next week,
> > could
> > it be too late?
> >
> > Angelo
> >
> > --- Albert Pauw <a....@wa...> wrote:
> > > After following very hopefull messages on this
> >
> > list
> >
> > > I was wondering
> > > what the actual status is. Maybe that would be
> > > handy. Also
> > > which scanners are actually supported (or working)
> > > or should
> > > be working.
> > >
> > > There seems to be differences between the
> >
> > different
> >
> > > models:
> > >
> > > 42 bits (14 bits per color):
> > >
> > > N650U/N656U (600x1200 dpi)
> > > N1220U (1200x2400 dpi)
> > >
> > > 48 bits (16 bits per color):
> > >
> > > N670U/N676U (600x1200 dpi)
> > > N1240U (1200x2400 dpi)
> > >
> > > How are all these supported?
> > >
> > > The 656 and 676 seem to be identical to the 650
> >
> > and
> >
> > > 670 model, resp.
> > > Only difference is the metallic finish (although
> > > it's plastic).
> > >
> > > The software of Jason Millward works very well for
> > > my N650U scanner,
> > > although the scanner wants to go a bit further
> >
> > than
> >
> > > it can (which makes
> > > an uneasy noise, but it's just a little bit).
> > >
> > > Could someone give some sort of a status report?
> > >
> > > That would be very interesting.
> > >
> > > Keep up the good work!
> > >
> > > Albert
>
> -------------------------------------------------------
>
> > > This sf.net email is sponsored by:ThinkGeek
> > > Welcome to geek heaven.
> > > http://thinkgeek.com/sf
> > > _______________________________________________
> > > Canonscanner-devel mailing list
> > > Can...@li...
>
> https://lists.sourceforge.net/lists/listinfo/canonscanner-devel
>
> > __________________________________________________
> > Do you Yahoo!?
> > New DSL Internet Access from SBC & Yahoo!
> > http://sbc.yahoo.com
>
> -------------------------------------------------------
>
> > This sf.net email is sponsored by:ThinkGeek
> > Welcome to geek heaven.
> > http://thinkgeek.com/sf
> > _______________________________________________
> > Canonscanner-devel mailing list
> > Can...@li...
> > https://lists.sourceforge.net/lists/listinfo/canonscanner-devel
>
> __________________________________________________
> Do you Yahoo!?
> New DSL Internet Access from SBC & Yahoo!
> http://sbc.yahoo.com
>
>
> -------------------------------------------------------
> This sf.net email is sponsored by:ThinkGeek
> Welcome to geek heaven.
> http://thinkgeek.com/sf
> _______________________________________________
> Canonscanner-devel mailing list
> Can...@li...
> https://lists.sourceforge.net/lists/listinfo/canonscanner-devel

Re: [Canonscanner-devel] Status report

[R N d. <ran...@ya...>]

Here are the data of my system:
Mdk 8.2 with libusb0.1_4 (mdk)

and this is the error message:

[angelo@naspa ucanoscan]$ gmake
usbio-ugen.c:28:25: dev/usb/usb.h: No such file or
directory
usbio-ugen.c:28:25: dev/usb/usb.h: No such file or
directory
gcc -c -o ucanoscan.o -g -Wall -DDEBUG_DATASAVE
-DUGEN_DRIVER -I/usr/local/include ucanoscan.c
gcc -c -o usbio-ugen.o -g -Wall -DDEBUG_DATASAVE
-DUGEN_DRIVER -I/usr/local/include usbio-ugen.c
usbio-ugen.c:28:25: dev/usb/usb.h: No such file or
directory
gmake: *** [usbio-ugen.o] Error 1
[angelo@naspa ucanoscan]$

Trying gmake UCANOSCAN_DRIVER=1

[angelo@naspa ucanoscan]$ gmake UCANOSCAN_DRIVER=1
usbio-ucanoscan.c:28:25: dev/usb/usb.h: No such file
or directory
usbio-ucanoscan.c:28:25: dev/usb/usb.h: No such file
or directory
gcc -c -o ucanoscan.o -g -Wall -DDEBUG_DATASAVE
-DUCANOSCAN_DRIVER -I/usr/local/include ucanoscan.c
gcc -c -o usbio-ucanoscan.o -g -Wall -DDEBUG_DATASAVE
-DUCANOSCAN_DRIVER -I/usr/local/include
usbio-ucanoscan.c
usbio-ucanoscan.c:28:25: dev/usb/usb.h: No such file
or directory
gmake: *** [usbio-ucanoscan.o] Error 1
[angelo@naspa ucanoscan]$

What's wrong in it?

thanks 
Angelo

--- R N dev <ran...@ya...> wrote:
> Hi Albert,
> I have the N656U, i'm sorry but i can't do any
> test since i can't compile the sources. 
> I've tryed the old code with Mandrake 8.1 and 
> old usblib (i cannot remember the version and i'm
> not at home now to give you more info) and it was
> good. Now i've MDK 8.2 and i cannot compile the
> source the error is on "dev/usb/usb.h" that cannot
> be
> included (i don't have it, i have usb.h under 
> /usr/include, but trying to use that one seems to
> give more errors...) I'm quite convince myself to 
> download the MDK 9.0 rc3 to try if something goes
> better.... but at this moment i've not so much free 
> time. Hope somebody had the same problem to solve
> the compiling problem so i can do some tests on my
> scanner...
> Anyway i should be a little bit free next week,
> could
> it be too late?
> 
> Angelo
> 
> 
> --- Albert Pauw <a....@wa...> wrote:
> > After following very hopefull messages on this
> list
> > I was wondering
> > what the actual status is. Maybe that would be
> > handy. Also
> > which scanners are actually supported (or working)
> > or should
> > be working.
> > 
> > There seems to be differences between the
> different
> > models:
> > 
> > 42 bits (14 bits per color):
> > 
> > N650U/N656U (600x1200 dpi)
> > N1220U (1200x2400 dpi)
> > 
> > 48 bits (16 bits per color):
> > 
> > N670U/N676U (600x1200 dpi)
> > N1240U (1200x2400 dpi)
> > 
> > How are all these supported?
> > 
> > The 656 and 676 seem to be identical to the 650
> and
> > 670 model, resp.
> > Only difference is the metallic finish (although
> > it's plastic).
> > 
> > The software of Jason Millward works very well for
> > my N650U scanner,
> > although the scanner wants to go a bit further
> than
> > it can (which makes
> > an uneasy noise, but it's just a little bit).
> > 
> > Could someone give some sort of a status report?
> > 
> > That would be very interesting.
> > 
> > Keep up the good work!
> > 
> > Albert
> > 
> > 
> > 
> > 
> > 
> >
>
-------------------------------------------------------
> > This sf.net email is sponsored by:ThinkGeek
> > Welcome to geek heaven.
> > http://thinkgeek.com/sf
> > _______________________________________________
> > Canonscanner-devel mailing list
> > Can...@li...
> >
>
https://lists.sourceforge.net/lists/listinfo/canonscanner-devel
> 
> 
> __________________________________________________
> Do you Yahoo!?
> New DSL Internet Access from SBC & Yahoo!
> http://sbc.yahoo.com
> 
> 
>
-------------------------------------------------------
> This sf.net email is sponsored by:ThinkGeek
> Welcome to geek heaven.
> http://thinkgeek.com/sf
> _______________________________________________
> Canonscanner-devel mailing list
> Can...@li...
> https://lists.sourceforge.net/lists/listinfo/canonscanner-devel

__________________________________________________
Do you Yahoo!?
New DSL Internet Access from SBC & Yahoo!
http://sbc.yahoo.com

Re: [Canonscanner-devel] CanoScan N670U - x,y position of scanned pic

[Christian <c.r...@ro...>]

Am Sonntag, 22. September 2002 21:20 schrieb Johann.Philipp:
> Okay,
>
> I made a mistake when calculating the tl_x_pixels variable: I rounded u=
p to
> the next inch, but should have rounded to the next pixel. Fix this simp=
ly
> by moving the ")" of the round function behind the
> c_h->model_parms[PAR_MAX_HZ_RES].i factor.

Great! It works.

> Some more to the snowy-picture-problem: I have this effect, too. It app=
ears
> also, when un- and replugging the scanner and reloading the kernel modu=
le.
> The effect seems to be of random nature, so the initializing procedure =
is
> not totally OK. Maybe the ordering. I'll try some more.

There is some other problem with scanner init. If you abort the previewsc=
an=20
while you have seen the picture is going to be snowy and you do a second=20
previewscan, the scanner makes noise. It seems that the scanner tries to=20
bring the sensor back to home position, although it is already there.

Maybe something like a position reset is required, isn't it?

Regards

Christian

Re: [Canonscanner-devel] Status report

[R N d. <ran...@ya...>]

Hi Albert,
I have the N656U, i'm sorry but i can't do any
test since i can't compile the sources. 
I've tryed the old code with Mandrake 8.1 and 
old usblib (i cannot remember the version and i'm
not at home now to give you more info) and it was
good. Now i've MDK 8.2 and i cannot compile the
source the error is on "dev/usb/usb.h" that cannot be
included (i don't have it, i have usb.h under 
/usr/include, but trying to use that one seems to
give more errors...) I'm quite convince myself to 
download the MDK 9.0 rc3 to try if something goes
better.... but at this moment i've not so much free 
time. Hope somebody had the same problem to solve
the compiling problem so i can do some tests on my
scanner...
Anyway i should be a little bit free next week, could
it be too late?

Angelo


--- Albert Pauw <a....@wa...> wrote:
> After following very hopefull messages on this list
> I was wondering
> what the actual status is. Maybe that would be
> handy. Also
> which scanners are actually supported (or working)
> or should
> be working.
> 
> There seems to be differences between the different
> models:
> 
> 42 bits (14 bits per color):
> 
> N650U/N656U (600x1200 dpi)
> N1220U (1200x2400 dpi)
> 
> 48 bits (16 bits per color):
> 
> N670U/N676U (600x1200 dpi)
> N1240U (1200x2400 dpi)
> 
> How are all these supported?
> 
> The 656 and 676 seem to be identical to the 650 and
> 670 model, resp.
> Only difference is the metallic finish (although
> it's plastic).
> 
> The software of Jason Millward works very well for
> my N650U scanner,
> although the scanner wants to go a bit further than
> it can (which makes
> an uneasy noise, but it's just a little bit).
> 
> Could someone give some sort of a status report?
> 
> That would be very interesting.
> 
> Keep up the good work!
> 
> Albert
> 
> 
> 
> 
> 
>
-------------------------------------------------------
> This sf.net email is sponsored by:ThinkGeek
> Welcome to geek heaven.
> http://thinkgeek.com/sf
> _______________________________________________
> Canonscanner-devel mailing list
> Can...@li...
>
https://lists.sourceforge.net/lists/listinfo/canonscanner-devel


__________________________________________________
Do you Yahoo!?
New DSL Internet Access from SBC & Yahoo!
http://sbc.yahoo.com

Re: [Canonscanner-devel] CanoScan N670U - x,y position of scanned pic

[Johann.Philipp <joh...@we...>]

Okay,

I made a mistake when calculating the tl_x_pixels variable: I rounded up to 
the next inch, but should have rounded to the next pixel. Fix this simply by 
moving the ")" of the round function behind the 
c_h->model_parms[PAR_MAX_HZ_RES].i factor.

Some more to the snowy-picture-problem: I have this effect, too. It appears 
also, when un- and replugging the scanner and reloading the kernel module. 
The effect seems to be of random nature, so the initializing procedure is not 
totally OK. Maybe the ordering. I'll try some more.

Johann

> Am Montag, 16. September 2002 22:29 schrieb Johann.Philipp:
> > Hi once more,
> >
> > I think, I should test some more...
> > Currently I have no idea, what's wrong. I'll try to get a closer look on
> > this soon, but I could also be happy with more detailed
> > what-happens-when's.
>
> Well, before I had posted this message, I had made some tests to figure out
> some rules the error followed but it is very difficult to explain.
>
> For example:
>
> Think of a picture, where you can see a humans head on it. You put a frame
> around it and you scan it. The result shows the head but mostly it is cut
> off at one site. Sometimes the left, sometimes the right site. The
> dimensions of the pic are okay.
>
> I can not say more to it. Do you have these problems too on your N1240U or
> is it just on the N670U?
>
> Regards
>
> Christian
>
> P.S.: I am reading the SANE-API at the time. Maybe my bugreports will go
> better then ;-) I am on page 21 for now.
>
>
> -------------------------------------------------------
> Sponsored by: AMD - Your access to the experts on Hammer Technology!
> Open Source & Linux Developers, register now for the AMD Developer
> Symposium. Code: EX8664 http://www.developwithamd.com/developerlab
> _______________________________________________
> Canonscanner-devel mailing list
> Can...@li...
> https://lists.sourceforge.net/lists/listinfo/canonscanner-devel

[Canonscanner-devel] Status report

[Albert P. <a....@wa...>]

After following very hopefull messages on this list I was wondering
what the actual status is. Maybe that would be handy. Also
which scanners are actually supported (or working) or should
be working.

There seems to be differences between the different models:

42 bits (14 bits per color):

N650U/N656U (600x1200 dpi)
N1220U (1200x2400 dpi)

48 bits (16 bits per color):

N670U/N676U (600x1200 dpi)
N1240U (1200x2400 dpi)

How are all these supported?

The 656 and 676 seem to be identical to the 650 and 670 model, resp.
Only difference is the metallic finish (although it's plastic).

The software of Jason Millward works very well for my N650U scanner,
although the scanner wants to go a bit further than it can (which makes
an uneasy noise, but it's just a little bit).

Could someone give some sort of a status report?

That would be very interesting.

Keep up the good work!

Albert

Re: [Canonscanner-devel] CanoScan N670U - x,y position of scanned pic

[Christian <c.r...@ro...>]

Am Montag, 16. September 2002 22:29 schrieb Johann.Philipp:
> Hi once more,
>
> I think, I should test some more...
> Currently I have no idea, what's wrong. I'll try to get a closer look o=
n
> this soon, but I could also be happy with more detailed
> what-happens-when's.
>

Well, before I had posted this message, I had made some tests to figure o=
ut=20
some rules the error followed but it is very difficult to explain.

For example:

Think of a picture, where you can see a humans head on it. You put a fram=
e=20
around it and you scan it. The result shows the head but mostly it is cut=
 off=20
at one site. Sometimes the left, sometimes the right site. The dimensions=
 of=20
the pic are okay.

I can not say more to it. Do you have these problems too on your N1240U o=
r is=20
it just on the N670U?

Regards

Christian

P.S.: I am reading the SANE-API at the time. Maybe my bugreports will go=20
better then ;-) I am on page 21 for now.

Re: [Canonscanner-devel] CanoScan N670U - x,y position of scanned pic

[Johann.Philipp <joh...@we...>]

Hi once more,

I think, I should test some more...
Currently I have no idea, what's wrong. I'll try to get a closer look on this 
soon, but I could also be happy with more detailed what-happens-when's.

Johann

> Hello,
>
> Although I wrote about just two remaining errors, I have found one more.
> Sorry.
>
> After selecting a frame to scan, everthing seems to be alright except the
> fact that the scanned picture does not fit the selected area.
>
> Just the y position is okay.The x position is wrong. This bug is new. It
> occured since the last updates to CVS.
>
> Happy coding
>
> Regards
>
> Christian
>
>
> -------------------------------------------------------
> This sf.net email is sponsored by:ThinkGeek
> Welcome to geek heaven.
> http://thinkgeek.com/sf
> _______________________________________________
> Canonscanner-devel mailing list
> Can...@li...
> https://lists.sourceforge.net/lists/listinfo/canonscanner-devel

Re: [Canonscanner-devel] canon_usb - only few bugs left

[Johann.Philipp <joh...@we...>]

Hello again,

> Hi,
>
> the new code works very, very well. As far as I can see, there are only 2
> bugs left:
>
> 1.)
> Scanner initialization seems not to be correct:
>
> When starting xscanimage, xsane, ... for the first time right after booting
> and you make a previewscan, the picture is snowy.
>
> After this, the problem is gone until next boot of the computer. (Did you
> mean this with your Bug-comment in canon_usb_xyz.c ?)
>

It might be the same source of problem, but my scanner just doesn't do 
anything then. Another thing is, that for Win98 scanning, I need to replug 
the scanner. I'm not sure at all, but maybe it is not correct to just read / 
write to the devicefile, but use some library / kernel functions. But I need 
some time to study this phenomena further. Currently it is not a too big 
hassle!

> 2.)
> Right, the scanned picturesize is fixed now. But the previewscanarea still
> lacks the y-range ;-)
> I changed 10inches to 11.80 and therefor a DIN A4 should be seen
> completely.
>

OK, OK, I put it on the todo list... But it is not a bug, isn't it? ;-)

> That's all.
>
> Thank you very much Johann Philip! Great work! Really
>
> Regards
>
> Christian
>
>
>
> -------------------------------------------------------
> This sf.net email is sponsored by:ThinkGeek
> Welcome to geek heaven.
> http://thinkgeek.com/sf
> _______________________________________________
> Canonscanner-devel mailing list
> Can...@li...
> https://lists.sourceforge.net/lists/listinfo/canonscanner-devel

[Canonscanner-devel] CanoScan N670U - x,y position of scanned pic

[Christian <c.r...@ro...>]

Hello,

Although I wrote about just two remaining errors, I have found one more.=20
Sorry.

After selecting a frame to scan, everthing seems to be alright except the=
 fact=20
that the scanned picture does not fit the selected area.

Just the y position is okay.The x position is wrong. This bug is new. It=20
occured since the last updates to CVS.

Happy coding

Regards

Christian

[Canonscanner-devel] canon_usb - only few bugs left

[Christian <c.r...@ro...>]

Hi,

the new code works very, very well. As far as I can see, there are only 2=
 bugs=20
left:

1.)
Scanner initialization seems not to be correct:

When starting xscanimage, xsane, ... for the first time right after booti=
ng=20
and you make a previewscan, the picture is snowy.

After this, the problem is gone until next boot of the computer. (Did you=
 mean=20
this with your Bug-comment in canon_usb_xyz.c ?)

2.)
Right, the scanned picturesize is fixed now. But the previewscanarea stil=
l=20
lacks the y-range ;-)
I changed 10inches to 11.80 and therefor a DIN A4 should be seen complete=
ly.

That's all.

Thank you very much Johann Philip! Great work! Really

Regards

Christian

Re: [Canonscanner-devel] CanoScan N670U - minor problems

[Christian <c.r...@ro...>]

Am Sonntag, 15. September 2002 16:49 schrieben Sie:
> > Hi,
> >
> > after some discussions with Johann Philipp, I got his SANE-API workin=
g
> > for the CanoScan N670U.
> >
> > I sent him a diff file (canon_usb.c.diff), which contains a different
> > _IOR call and a expanded switch/case construct in function
> > sane_get_devices(...)
> >
> > After his last email to me, I changed every explicit dpi resolution f=
rom
> > 1200 into 600 and changed all HZ_RES_DIVER_x to half of its values
> > (HZ_RES_DIVIDER_4 =3D> HZ_RES_DIVIDER_2). These changes were made in
> > canon_usb_mid.c
> >
> > There are still some problems:
> >
> > a)
> > The previewscan only scans DIN B5 in y direction and about 20cm in x
> > direction, although the scanarea was chosen "full area" in xsane-0.87=
=2E
>
> This is because I restricted the scan area in "const SANE_Range x_range=
".
> You can try some other values. Hint: use xscanimage for testing. It is
> closer to the backend, and not providing fancy and irritating options w=
hich
> are not given by the backend.

xscanimage does not work very good. The produced outputfile out.pnm is no=
t=20
useable. Gimp is unable to load the picture because of a corrupted file e=
nd.

xsane is working very fine.

>
> > b)
> > If I choose an area from the previewed picture, the scanned picture o=
ften
> > does not show that part after scanning.
>
> I didn't test all this very much. Try to uncomment some of the printf's=
 I
> put at various places in the source code, maybe they can give some hint=
s.
>
> > c)
> > Sometimes the scanned picture looks phasing-shifted.
> >
> > I think, these problems occur, because I took the canon_usb_mid.c fro=
m
> > N1240U and have not changed everthing to the N670U model, yet.
>
> I think this is a problem not connected to specific models. I believe, =
that
> the parameter pixels_per_line doesn't correspond always to the values
> calculated in config_pixel_ranges(...), due to rounding errors. Try to
> hotfix by adding simply 1 to the value for REG_DATA_PIXELS_END_HI. The
> lm9833 manual says, supernumerous sensor pixels, that cannot form a
> complete output pixel, will be discarded. (e.g. 4 sensor pixels are nee=
ded
> on a 600dpi scanner to form 1 output pixel for a 150 dpi image). In fut=
ure
> I should use the pixels_per_line value to calculate the
> REG_DATA_PIXELS_END_HI.

I made about 50 tests with pixels_per_line so I can say: Evertime the val=
ue=20
was odd, the picture was phasing-shifted. So I think there is no problem =
with=20
rounding. (?? The thing with odd or even came from my cousine. He had the=
=20
same problem on an other scanner driver he has coded. So maybe it is a no=
rmal=20
behaviour, isn't it?)

I tried to write a hotfix but the frontend crashes with segmentation faul=
t. I=20
wrote a if-function in void config_pixel_ranges(...):

sane_get_parameters(c_h, 0);
if (c_h->parms.pixels_per_line & 0x01) {
=09// Code taken from MAXX + (int) round(...) + 1
else
=09// same without adding 1

This does not work. Any ideas?
Maybe it is something like pixels_per_line <-> byzes_per_line which diffe=
rs if=20
I manipulate the function with the code shown above?

What do you think about it?

Regards

Christian

[Canonscanner-devel] update for sane backend

[Johann.Philipp <joh...@we...>]

Hi everyone,

I updated the backend some, look at the README.canon_usb. I think, I fixed 
some bugs, but one thing is still unsolved:
I haven't got a universial ioctl for determining the vendor_id and the 
product_id. Any ideas?

Johann

Re: [Canonscanner-devel] sane backend

[R N d. <ran...@ya...>]

Hi,
I'm sorry but i've some problem to send e-mails to
mailing list.... :-(

I'd like to compile ct_N1240U.c to try it on my
N656U. Before this last release it worked, but now
it needs "dev/usb/usb.h" into some files.
I had only usb.h (under usr/include) but it dos not
work. What version of libusb are you using?

Thanks
Angelo


--- "Johann.Philipp" <joh...@we...> wrote:
> Hi guys,
> 
> good news: I was able to knit a sane backend for my
> N1240U. It works fairly 
> well and is basically my testprogram ct_N1240U.c
> with the SANE-API. I hope 
> this will be a good starting point for merging our
> knowledge on the different 
> models. Also I hope for some feedback... Just have a
> look at the cvs!
> 
> Have a good week!
> Johann
> 
> 
>
-------------------------------------------------------
> This sf.net email is sponsored by: OSDN - Tired of
> that same old
> cell phone?  Get a new here for FREE!
>
https://www.inphonic.com/r.asp?r=sourceforge1&refcode1=vs3390
> _______________________________________________
> Canonscanner-devel mailing list
> Can...@li...
> https://lists.sourceforge.net/lists/listinfo/canonscanner-devel

__________________________________________________
Do you Yahoo!?
Yahoo! News - Today's headlines
http://news.yahoo.com