Re: Mustek 1200SP: Missing Lines

Joachim Woll (woll@physik.uni-kassel.de)
Thu, 30 Oct 1997 23:37:27 +0100

Hi there,
first of all IŽd like to thank for the response to my posting last week.
Unfortunately I've to say that my Scanner still is not working properly,
but I want to resume...

-->The DTC436E SCSI-Card is not recognized by the Kernel.

becka@rz.uni-duesseldorf.de wrote:
>Known problem ... the card is ... *#$?%
...yes!!
>However there have recently been some reports that it can be made working
>somehow ...
Would be great. Do you remember where you read these reports?

Michael K. Johnson wrote:
>If I recall correctly, that card is just an adaptec 152x clone
>without the bios. You just have to give arguments to the 152x
>driver to tell it where to find the card. DTC tested this for
>us with Red Hat 4.2, with 2.0.30, and were successful.
No! I guess you mean the DTCT 3x80,329x.These Cards are
adaptec 1540 compatible. And I read in the meantime in the
german magazine iX that the DTC436E is definitely *not*
supported.

-->MUSTEK 1200SP hooked to the adaptec 2940

David.Mosberger@acm.org wrote:
>Not quite, unfortunately. What I see are shifts, not missing lines.
>Still, the problems might be related.
Could be. In my case there are missing lines -sometimes up to 10
at a time.

becka@rz.uni-duesseldorf.de wrote:
>> These problems are described in sane-mustek, but applying the fixes
>> through mustek.conf did not change anything.
>Setting the strip-height should affect the frequency of the stripes.
Yes it does. A smaller value leads to a higher frequency.

>> So I guess that Problem is due to a corrupt data-transfer because of a
>> timing problem or so.
>I do not think so, as it always seems to be at the beginning/end of a
>transferred buffer. So I think the last alignment between the current and the
>next buffer is not quite right. I don't know, if this is a HW problem, or if
>we are doing something wrong.
You are right. This problem occurs at every buffer boundary.

>However the main developer of the Mustek drivers semms to be able to reproduce
>this problems recently, so maybe there is hope ...
:-)

Joachim.

--
Source code, list archive, and docs: http://www.mostang.com/sane/
To unsubscribe: echo unsubscribe sane-devel | mail majordomo@mostang.com