plustek: Thread: epson 1250 perfection usb scanner

OK, I'm real confused here. I just bought a Epson 1250 Perfection USB
Scanner as a gift for my nephew, naively figuring that since my Epson
1650 scanner works both USB and SCSI, the USB-only should work. Yeah.

I tried installing it on his machine (Red Hat 7.2), and have yet to
get USB working properly. So I backed off and tried installing it on
the machine that I had previously working with the 1650 and USB. I
think the USB part is OK. When I cat /proc/bus/usb/devices, I get
(along with other stuff):

  T:  Bus=01 Lev=01 Prnt=01 Port=00 Cnt=01 Dev#=  3 Spd=12  MxCh= 0
  D:  Ver= 1.10 Cls=ff(vend.) Sub=00 Prot=ff MxPS= 8 #Cfgs=  1
  P:  Vendor=04b8 ProdID=010f Rev= 1.00
  S:  Manufacturer=EPSON
  S:  Product=EPSON Scanner 010F
  C:* #Ifs= 1 Cfg#= 1 Atr=c0 MxPwr=  2mA
  I:  If#= 0 Alt= 0 #EPs= 3 Cls=ff(vend.) Sub=00 Prot=ff Driver=(none)
  E:  Ad=81(I) Atr=03(Int.) MxPS=   1 Ivl= 16ms
  E:  Ad=82(I) Atr=02(Bulk) MxPS=  64 Ivl=  0ms
  E:  Ad=03(O) Atr=02(Bulk) MxPS=  64 Ivl=  0ms

I don't know what all this means, but at least it matches the vendor
and product codes in your Plustek-USB.txt file. scanimage -L and
sane-find-scanner return nothing (but messages).

This is how I got this far:

  Red Hat Linux release 7.1 (Seawolf)
  Kernel 2.4.2-2 on an i686

/etc/modules.conf had this line:

  alias usb-controller usb-uhci

I did rpm -e on the old sane modules, then downloaded and built:

  sane-backends-1.0.7
  sane-frontends-1.0.7
  xsane-0.86

For the plustek module, I downloaded plustek-module-0_42_9.tar.gz, then:

  cd sane-backends-1.0.7/backend
  tar zxvf ../../plustek-module-0_42_9.tar.gz
  cd plustek_driver
  make all
  make install

(Actually, I did the tar after untarring sane-backends and running
.configure on it, but before make.)

The INSTALL file in the plustek_driver directory says to add the
following lines to /etc/modules.conf, then run depmod:

  alias char-major-40	pt_drv
  pre-install pt_drv modprobe -k parport
  options pt_drv lampoff=180 warmup=15 port=0x378 lOffonEnd=0 mov=0 slowIO=1

I did, but I don't understand what this has to do with a USB scanner. The
doc says that a /dev/pt_drv file is needed, but I don't have such a file,
and can't find how to mknod it. Looking at the plustek.conf file, it looks
like /dev/pt_drv is used for parport scanners, and /dev/usbscanner is used
for USB scanners. I don't have a /dev/usbscanner either, but presumably
that should be a link to whichever /dev/usb/scanner* is correct (0-11).
Back in Plustek-USB.txt it also says to do:

  modprobe scanner

or

  modprobe scanner vendor=0x4b8 product=0x10f

I did both (in that order); lsmod now says (among other things):

  scanner   6240 0 (unused)
  usb-uhci 20720 0 (unused)
  usbcore  49664 1 [scanner usb-uhci]

I also don't understand how the plustek module gets used for an Epson
scanner (although in this case it very probably isn't).

I scanned back through the last 3-4 months of mail archive; evidently
some people have gotten the thing working (somewhat), so I must be doing
something wrong. What?

Also, is there any reason not to think that it will eventually work
pretty well? (While I can still afford to take it back.)

-- 
/*
 *  Tom Hull * thull at kscable.com * http://www.tomhull.com/
 */

Hi Tom,

On Freitag, 24. Mai 2002 06:34, Tom Hull wrote:
[SNIP]
> OK, I'm real confused here. I just bought a Epson 1250 Perfection USB
> Scanner as a gift for my nephew, naively figuring that since my Epson
> 1650 scanner works both USB and SCSI, the USB-only should work. Yeah.

The 1650 and the 1250 are qutie different!!!!

>
> This is how I got this far:
>
>   Red Hat Linux release 7.1 (Seawolf)
>   Kernel 2.4.2-2 on an i686

Here's the problem!!! The 2.4.2 does not support the automatic detection
of scanners by a backend... see below!

>
> /etc/modules.conf had this line:
>
>   alias usb-controller usb-uhci
>
> I did rpm -e on the old sane modules, then downloaded and built:
>
>   sane-backends-1.0.7
>   sane-frontends-1.0.7
>   xsane-0.86

Okay so far!

>
> For the plustek module, I downloaded plustek-module-0_42_9.tar.gz, then:
>
>   cd sane-backends-1.0.7/backend
>   tar zxvf ../../plustek-module-0_42_9.tar.gz
>   cd plustek_driver
>   make all
>   make install

NO NEED TO DO THIS FOR USB DEVICES!!!! This driver is only good
for the Plustek parallel port scanner.

[SNIPSNAP]
>   modprobe scanner vendor=0x4b8 product=0x10f

For Kernel 2.4.2, this one is correct!

>
> I did both (in that order); lsmod now says (among other things):
>
>   scanner   6240 0 (unused)
>   usb-uhci 20720 0 (unused)
>   usbcore  49664 1 [scanner usb-uhci]
>
> I also don't understand how the plustek module gets used for an Epson
> scanner (although in this case it very probably isn't).

As said before, the Plustek module is not needed, only the Plustek-backend,
as this controls LM983x based scanners. The LM983x is an ASIC, developed
by National Semiconductors, to build cheap scanners. And such a beast is
inside the 1250.

>
> I scanned back through the last 3-4 months of mail archive; evidently
> some people have gotten the thing working (somewhat), so I must be doing
> something wrong. What?
>
> Also, is there any reason not to think that it will eventually work
> pretty well? (While I can still afford to take it back.)

Well, the main problem currently is the calibration. But do a test on your
own before bringing it back...

You only have to alter the plustek.conf file in your sane.d directory to
enable the EPSON, the following entry should be enough to enable it:

[usb] 0x04B8 0x010F
device /dev/usbscanner


Hope this helps
  Gerhard

 -------------------------
|gerhard AT gjaeger DOT de|
 -------------------------

I have the Epson 1250 Perfection USB scanner "working" -- i.e., the
scanner runs and sends data which I can see sort of resembles the
image I'm scanning. However, the results strike me as much more than
a "calibration" problem. Please take a look at:

  http://www.tomhull.com/scan1.jpg

This is a 40k jpeg reduction of the scan of a book cover which is
mostly black text on white background. As you can see, overall there
is a lot of red, and there is a lot of banding. When I fiddle with
the xsane brightness controls, it looks to me like the red channel
is coherent but oversaturated; the blue and green channels, OTOH,
show the banding but no coherent image.

My plustek.conf file matches the one posted by Gene Heskett, including:

  option lampOff 600
  option warmup  180
  option lOffOnEnd 1

However, once the lamp comes on it stays on indefinitely. (Turns off
when rebooting.)

"Jaeger, Gerhard" wrote:
> 
> Hi Tom,
> 
> On Freitag, 24. Mai 2002 06:34, Tom Hull wrote:
> [SNIP]
> > Also, is there any reason not to think that it will eventually work
> > pretty well? (While I can still afford to take it back.)
> 
> Well, the main problem currently is the calibration. But do a test on your
> own before bringing it back...

-- 
/*
 *  Tom Hull * thull at kscable.com * http://www.tomhull.com/
 */

Hi Tom,

this picture reminds me of some earlier tries with the driver...
I thought - Gene correct me - that the color problems are gone so far.
The only thing that is still missing is some sort of dark-calibration but
the colors should not be that bad...
Can please try the latest 1.0.8 SANE stuff?

Gerhard

On Thursday, 30. May 2002 17:05, Tom Hull wrote:
> I have the Epson 1250 Perfection USB scanner "working" -- i.e., the
> scanner runs and sends data which I can see sort of resembles the
> image I'm scanning. However, the results strike me as much more than
> a "calibration" problem. Please take a look at:
>
>   http://www.tomhull.com/scan1.jpg
>
> This is a 40k jpeg reduction of the scan of a book cover which is
> mostly black text on white background. As you can see, overall there
> is a lot of red, and there is a lot of banding. When I fiddle with
> the xsane brightness controls, it looks to me like the red channel
> is coherent but oversaturated; the blue and green channels, OTOH,
> show the banding but no coherent image.
>
> My plustek.conf file matches the one posted by Gene Heskett, including:
>
>   option lampOff 600
>   option warmup  180
>   option lOffOnEnd 1
>
> However, once the lamp comes on it stays on indefinitely. (Turns off
> when rebooting.)
>
> "Jaeger, Gerhard" wrote:
> > Hi Tom,
> >
> > On Freitag, 24. Mai 2002 06:34, Tom Hull wrote:
> > [SNIP]
> >
> > > Also, is there any reason not to think that it will eventually work
> > > pretty well? (While I can still afford to take it back.)
> >
> > Well, the main problem currently is the calibration. But do a test on
> > your own before bringing it back...

-- 
-------------------------
gerhard AT gjaeger DOT de

On Saturday 01 June 2002 06:30 am, Jaeger, Gerhard wrote:
>Hi Tom,
>
>this picture reminds me of some earlier tries with the driver...
>I thought - Gene correct me - that the color problems are gone so
> far. The only thing that is still missing is some sort of
> dark-calibration but the colors should not be that bad...
>Can please try the latest 1.0.8 SANE stuff?
>
>Gerhard

Its useable here with the last 1.07-tar.gz build, Gerhard.  Not 
perfect, but useable, main complaint is a yellow to magenta 
highlight shift along the edges, but the white clipping in actually 
putting it back on paper with gimp-print from gimp usually reduces 
that to near vanishing.  And one other effect I've noted several 
times but which seems rather ephemeral is that if it sits idle for 
a day or more, the first scan is absolutely beautifull, but the 
next and all subsequent scans, even if shut down during the workday 
for several hours, show this high-light  shading that makes it look 
as if the lamp isn't warmed up and uniform.  I have mine set for a 
near one minute warmup as you can see below.

If one looks really close to a scan of a chemical photograph, a 
multicolored vertical pattern of pixels can be seen in the grey 
areas below that brightness that would represent about half the 
brightness of a kodak 18% grey card, and going on down to black 
from there.

At risk of making it set stakes and call a surveyer slow, it would 
be nice if a black sample were taken (in color at the same 
resolution currently in use) and that were then subtracted from 
each line of data that came in, which should subtract a large 
amount of that.  That wouldn't be too slow, but the next step is to 
save the white sample, derive a difference from the ideal gain on a 
pixel by pixel basis for all 3 colors, and then multiply each color 
pixel in each line received by this 'per pixel' gain coefficient 
for this pixel in the white balance sample.  That would slow it 
down some even if the cpu's math unit doesn't have the infamous 
pentium bug.  I of course have NDI how to go about that!

OTOH, we are trying to make a 100 dollar scanner look like a 2000 
dollar scanner here, which may not be possible in the grand scheme 
of things due to entropy of the universe or some such limiting 
effect. :-)

>On Thursday, 30. May 2002 17:05, Tom Hull wrote:
>> I have the Epson 1250 Perfection USB scanner "working" -- i.e.,
>> the scanner runs and sends data which I can see sort of
>> resembles the image I'm scanning. However, the results strike me
>> as much more than a "calibration" problem. Please take a look
>> at:
>>
>>   http://www.tomhull.com/scan1.jpg
>>
>> This is a 40k jpeg reduction of the scan of a book cover which
>> is mostly black text on white background. As you can see,
>> overall there is a lot of red, and there is a lot of banding.
>> When I fiddle with the xsane brightness controls, it looks to me
>> like the red channel is coherent but oversaturated; the blue and
>> green channels, OTOH, show the banding but no coherent image.
>>
>> My plustek.conf file matches the one posted by Gene Heskett,
>> including:
>>
>>   option lampOff 600
>>   option warmup  180
>>   option lOffOnEnd 1
>>
>> However, once the lamp comes on it stays on indefinitely. (Turns
>> off when rebooting.)

That sounds like he has a very early 1.07 cvs version.  Mine does 
indeed shut down when I quit the program xsane.  And at one point, 
it was returning a patterned but non-sensical raster of data.

>> "Jaeger, Gerhard" wrote:
>> > Hi Tom,
>> >
>> > On Freitag, 24. Mai 2002 06:34, Tom Hull wrote:
>> > [SNIP]
>> >
>> > > Also, is there any reason not to think that it will
>> > > eventually work pretty well? (While I can still afford to
>> > > take it back.)
>> >
>> > Well, the main problem currently is the calibration. But do a
>> > test on your own before bringing it back...

-- 
Cheers, Gene
AMD K6-III@500mhz 320M
Athlon1600XP@1400mhz  512M
98.96+% setiathome rank, not too shabby for a hillbilly