primax: Thread: Primax Colorado D600 Support ?

Hi there,

are there any news regarding the support of the Primax Colorado D600
Parallel Scanner ? Can we expect a new version of the Primax/Linux
Driver in the next time ? (I have so many Fotos to scan...)

Greetings
  Christoph

On 09-Jun-99 Christoph Adomeit wrote:
> Hi there,
> 
> are there any news regarding the support of the Primax Colorado D600
> Parallel Scanner ? Can we expect a new version of the Primax/Linux
> Driver in the next time ? (I have so many Fotos to scan...)
> 
Hi,

we have signed an NDA and will get the Specs for the Primax Colorado Direct
Scanner. Unfortunatly the Primax Colorado is not compatible to the D600.
Perhaps we will get the specs for the D600 later.

cu.



---
Christian Ordig             | The Primax Scanner Driver Project for Linux/UN*X |
Germany                     |   http://thor.prohosting.com/~chrordig/Primax/   |
   __   _                   |                                                  |
  / /  (_)__  __ ____  __   | Why Linux? Because it is free, stable, and       |
 / /__/ / _ \/ // /\ \/ /   | bugs can be fixed in source opposed to waiting   |
/____/_/_//_/\_,_/ /_/\_\   | for a stable WinTendo from Micro$oft.            |

Christian Ordig wrote:
> 
> On 09-Jun-99 Christoph Adomeit wrote:
> > Hi there,
> >
> > are there any news regarding the support of the Primax Colorado D600
> > Parallel Scanner ? Can we expect a new version of the Primax/Linux
> > Driver in the next time ? (I have so many Fotos to scan...)
> >
> Hi,
> 
> we have signed an NDA and will get the Specs for the Primax Colorado Direct
> Scanner. Unfortunatly the Primax Colorado is not compatible to the D600.
> Perhaps we will get the specs for the D600 later.
> 
> cu.
> 
> ---
> Christian Ordig             | The Primax Scanner Driver Project for Linux/UN*X |
Hi,

It looks like it is the time for me to expose my tests.

I have a Storm Totalscan scanner. From the FCCID, and after taking a
careful look on the Primax nederlands page, it is very similar to a
Colorado D600 (600x1200ppp, 30bits). In fact, the Primax Colorado D600
driver does work completely with the Totalscan (on Windows NT).

So I have tried the linux driver, and even if it is not working
completely (scan results give black images), hacking just a little bit
in the code allows to switch the lamp on/off and to drive the CCD at the
right positions...

Of course, I would be happy to get the final information to make it
work... but would be sad if this final information is not available ;-|

In any case, I am ready to test (and even hack if necessary) any piece
of code that have a chance to work...

Any reactions ?

-- 

Nicolas ####@####.####

Marco forgot to send it to the list, too...

-----FW: ####@####.####

Date: Sat, 12 Jun 1999 11:21:49 +0200
From: Marco Foglia ####@####.####
To: Nicolas Delisle ####@####.####
Subject: Re: Primax Colorado D600 Support ?
Cc: ####@####.####

Nicolas Delisle wrote:
> It looks like it is the time for me to expose my tests.
> 
> I have a Storm Totalscan scanner. From the FCCID, and after taking a
> careful look on the Primax nederlands page, it is very similar to a
> Colorado D600 (600x1200ppp, 30bits). In fact, the Primax Colorado D600
> driver does work completely with the Totalscan (on Windows NT).
> 
> So I have tried the linux driver, and even if it is not working
> completely (scan results give black images), hacking just a little bit
> in the code allows to switch the lamp on/off and to drive the CCD at the
> right positions...
>
> In any case, I am ready to test (and even hack if necessary) any piece
> of code that have a chance to work...


Hi,

maybe some more technical details will help you a little bit.

If you can turn on/off the lamp the wakeup sequence was successful.
This means the win driver is using vpmxd.vxd and your scanner is 
quite compatible to the D300. Resolutions are different, so try to
adjust x_/y_table and calibration_table. The setup of the ADC is also
of importance. Try different values. 

The part numbers of the ICs in your scanner would be a good 
information for me. Especially the ADC and the central ASIC.
It would be great if you could look them up.


Marco

--------------End of forwarded message-------------------------

Marco wrote:
> If you can turn on/off the lamp the wakeup sequence was successful. 
> This means the win driver is using vpmxd.vxd and your scanner is 
> quite compatible to the D300. Resolutions are different, so try to 
> adjust x_/y_table and calibration_table. The setup of the ADC is also 
> of importance. Try different values. 

I'm new to this list, and picked up the above from the archive on the
Web - it was in answer to a post by Nicolas Deslisle.

This is just to confirm that changing the relevant three lines in the
code to these:

static const int x_table_size = 0x13DC;
static const int y_table_size = 0x17BA;
static const int calibration_table_size = 0x3BC4;

... (just doubling each number) makes my Primax Colorado D600 work on my
system (Kernel 2.0.36-0.7, libtiff 3.4-4, parallel port set to SPP)

I have to give the command-line dimensions as twice what they really are
(-d6x6 if I want -d3x3 inches - which makes sense given the above
changes, although I haven't yet looked for where in the source one could
fix that).

The scan has a green cast (a known bug I gather) which is trivial to get
rid of using the gimp's Auto Levels). Otherwise, the scan looks pretty
good.

> The part numbers of the ICs in your scanner would be a good 
> information for me. Especially the ADC and the central ASIC. 
> It would be great if you could look them up. 

I'll see if I can get these for you - don't yet know where to look.

One detail : it doesn't work if I pass the cable through my HP7200 CD
writer - only with the cable going direct to the scanner - no big sweat.

Thanks for all the good work on this!

Cheers,
Colin

-- 
Colin Muller
http://www.durbanet.co.za
####@####.####

On 15-Jun-99 Colin Muller wrote:
> Marco wrote:
>> If you can turn on/off the lamp the wakeup sequence was successful. 
>> This means the win driver is using vpmxd.vxd and your scanner is 
>> quite compatible to the D300. Resolutions are different, so try to 
>> adjust x_/y_table and calibration_table. The setup of the ADC is also 
>> of importance. Try different values. 
> 
> I'm new to this list, and picked up the above from the archive on the
> Web - it was in answer to a post by Nicolas Deslisle.
> 
> This is just to confirm that changing the relevant three lines in the
> code to these:
> 
> static const int x_table_size = 0x13DC;
> static const int y_table_size = 0x17BA;
> static const int calibration_table_size = 0x3BC4;
> 
> ... (just doubling each number) makes my Primax Colorado D600 work on my
> system (Kernel 2.0.36-0.7, libtiff 3.4-4, parallel port set to SPP)
> 
> I have to give the command-line dimensions as twice what they really are
> (-d6x6 if I want -d3x3 inches - which makes sense given the above
> changes, although I haven't yet looked for where in the source one could
> fix that).
> 
This sounds really good. Which version of Windows95 driver do you have? vpmd.vxd
or vpmd2.vxd? If you have the first one your Scanner might not really be a
D600, because all the D600 we´ve heard of use the second one.... I´m not really
sure about this, but reporting it to us, would be really nice.

> The scan has a green cast (a known bug I gather) which is trivial to get
> rid of using the gimp's Auto Levels). Otherwise, the scan looks pretty
> good.
> 
Yes, that´s because there´s no calibration, yet. The code for the
software-calibration should be in version 0.2, but it´s not intended for use,
since the scanner has its own ability for calibrating. The scanner is able to
use its full 30 bit for calibration, software has only 24 Bits...
 
>> The part numbers of the ICs in your scanner would be a good 
>> information for me. Especially the ADC and the central ASIC. 
>> It would be great if you could look them up. 
> 
> I'll see if I can get these for you - don't yet know where to look.
> 
Well, open the scanner´s cover.  Near the hinge you´ll find 2 skrews. Simply
remove them and you´ll be able to open the scanner. On the side where all the
connectors are, you´ll find a circuit board with 6 ICs on it (well in the
Colorado Direct there are 6, maybe you´ll find more or less...). Well write
down the things written on these ICs.

> One detail : it doesn't work if I pass the cable through my HP7200 CD
> writer - only with the cable going direct to the scanner - no big sweat.
> 
Well, this seems to be a problem of the CD-Writer... Furthermore we don´t know
how the sent material changes when there´s another device on the parallel
port... it also doesn´t work if I switch on my HP Printer...

Thank you for supporting us.


---
Christian Ordig             | The Primax Scanner Driver Project for Linux/UN*X |
Germany                     |   http://thor.prohosting.com/~chrordig/Primax/   |
   __   _                   |                                                  |
  / /  (_)__  __ ____  __   | Why Linux? Because it is free, stable, and       |
 / /__/ / _ \/ // /\ \/ /   | bugs can be fixed in source opposed to waiting   |
/____/_/_//_/\_,_/ /_/\_\   | for a stable WinTendo from Micro$oft.            |

Christian Ordig wrote:
> This sounds really good. Which version of Windows95 driver do you 
> have? vpmd.vxd
> or vpmd2.vxd? If you have the first one your Scanner might not really 
> be a
> D600, because all the D600 we´ve heard of use the second one.... I´m 
> not really
> sure about this, but reporting it to us, would be really nice.

vpmxd.vmd, dated 17-01-98
The scanner lid has the label D600, but of course that may mean nothing
(marketing tactics ...)

> Well, open the scanner´s cover.  Near the hinge you´ll find 2 skrews. > Simply
> remove them and you´ll be able to open the scanner. On the side where > all the
> connectors are, you´ll find a circuit board with 6 ICs on it (well in > the
> Colorado Direct there are 6, maybe you´ll find more or less...). Well > write
> down the things written on these ICs.

Okay, I should be able to do that later today.

A couple of other things which are probably predictable to people who
are familiar with the code:

- When I specify, say, 6 x 8 inches at 100dpi, the image area scanned is
(as I mentioned) really 3x4 inches, BUT the resolution is actually 200
dpi and I end up with a 600x800dpi image - I suppose that's another
side-effect of the changes I made to the constants.
- The scan seems to start a little bit before the clear glass area of
the flatbed - and I think it then also ends a bit before the end of the
specified area. Some sort of offset may be needed there (I haven't
looked in the code to see whether there is one I could have adjusted -
apologies if I'm telling you what you already know).

But really, I'm amazed at how well you guys got this working before
getting any support from Primax.

-- 
Colin Muller
http://www.durbanet.co.za
####@####.####

Colin Muller wrote:

> - When I specify, say, 6 x 8 inches at 100dpi, the image area scanned is
> (as I mentioned) really 3x4 inches, BUT the resolution is actually 200
> dpi and I end up with a 600x800dpi image - I suppose that's another
> side-effect of the changes I made to the constants.

Did you change X_/Y_RESOLUTION to 600? This might solve the problem.

> - The scan seems to start a little bit before the clear glass area of
> the flatbed - and I think it then also ends a bit before the end of the
> specified area. Some sort of offset may be needed there.

If Y_RESOLUTION is set to the correct value the lamp should move out
of the calibration area (move_carrige_forward(0.42) inside scan()).

You will see the calibration area if you open your scanner. It is 
white with a black line. This to-and-fro of the win driver at the 
beginning of every scan is the calibration procedure (3 scans). 

Marco

> This is just to confirm that changing the relevant three lines in the
> code to these:
> 
> static const int x_table_size = 0x13DC;
> static const int y_table_size = 0x17BA;
> static const int calibration_table_size = 0x3BC4;
> 
Hmm,

my D600 doesn't say anything, even after changing the above
lines. The Problem might even be with my kernel/ or hardware
setup. I enabled SPP/EPP in my Bios, but when I run
primax_scan -lon I receive the following message:

parse_options:
  lamp:                 on
  dimension:          0x  0 pixel
parse_options:       switching lamp
Probing port 0378
    SPP port present
    PS/2 bidirectional port present
    EPP 1.9 with hardware direction protocol

:-(
Scanner did not switch to EPP mode.....not compatible?

Any ideas what I can try ? What modules besides parport
should be loaded ? BTW: I still run a 2.1.132 kernel on
this machine....

Thanks
  Christoph

On 15-Jun-99 Christoph Adomeit wrote:
> Hmm,
> 
> my D600 doesn't say anything, even after changing the above
> lines. The Problem might even be with my kernel/ or hardware
> setup. I enabled SPP/EPP in my Bios, but when I run
> primax_scan -lon I receive the following message:
> 
> parse_options:
>   lamp:                 on
>   dimension:          0x  0 pixel
> parse_options:       switching lamp
> Probing port 0378
>     SPP port present
>     PS/2 bidirectional port present
>     EPP 1.9 with hardware direction protocol
> 
>:-(
> Scanner did not switch to EPP mode.....not compatible?
> 
> Any ideas what I can try ? What modules besides parport
> should be loaded ? BTW: I still run a 2.1.132 kernel on
> this machine....
> 
Well,

I had some problems with my printer, when I compiled the parport stuff as
module.... (Kernel 2.2.x). I didn´t try the scanner with this config, but when
I compiled the parport stuff directly into the kernel the printer works (and
the scanner also... ;-)

Not sure if this helps...

But I think the really significant difference between you and Colin is the
Windows95 driver... (I think your D600 driver is named vpmxd2.vxd, isn´t it?)

Marco found out that there are quite large differencies between the vpmxd and
the vpmxd2 driver... Perhaps Colin´s scanner is named D600, but is not an
original one.

Knowing the IC types would help us very much!

Thank you.


---
Christian Ordig             | The Primax Scanner Driver Project for Linux/UN*X |
Germany                     |   http://thor.prohosting.com/~chrordig/Primax/   |
   __   _                   |                                                  |
  / /  (_)__  __ ____  __   | Why Linux? Because it is free, stable, and       |
 / /__/ / _ \/ // /\ \/ /   | bugs can be fixed in source opposed to waiting   |
/____/_/_//_/\_,_/ /_/\_\   | for a stable WinTendo from Micro$oft.            |