RE: Microtek ScanMaker E3 Plus - Parallel Port - working!

From: Adam Benjamin (aebenjam@opentext.com)
Date: Thu Apr 05 2001 - 07:53:52 PDT

  • Next message: scc: "Re: pers.: Re: Xsane build problems"

    > > Step 1: Get your scanner working/visible as a SCSI device under
    > > Linux.
    >
    > This is also written in the Sane-Faq:
    > http://www.xs4all.nl/~ljm/SANE-faq.html
    > I guess you didn't find it?

    Actually, I did find various pieces I posted in various places. I
    simply wanted to put it all in one document specifically for the
    E3Plus. (I forget where all I searched but I went through a few
    different mailing lists including this one and never found evidence
    that someone had success.) My own research took me to the
    scsi-over-parallel patch pretty quickly, but it was a little ambiguous
    as to which the most recent one was - hence my inclusion of the
    information.

    > If I remember correctly, some time ago someone used also a scanner with
    > modelcode 90, but after patching the backend, it still didn't work.
    > Maybe that's the reason it was left out of the backend. Because there
    > were other issues with this backend.

    Not sure. My assumption was that they had already coded the
    instructions for the E3+ and I knew mine was an E3+ - hence it seemed
    safe to kludge mine to being recognized as model code 92. I would
    also guess that this is not necessarily safe in the general case as
    perhaps there are other model code 90 Microtek2 scanners that are not
    E3+.

    > > + case 0x90:
    > > + mi->model_code = 0x92;
    > > + *model_string = "E3+ / Vobis HighScan";
    > > + break;
    >
    > I'm no programmer, so maybe this is a stupid question, but the line:
    > > + mi->model_code = 0x92;
    > What does it do?

    As above - I said, "since you think this is model code 0x90 (the case
    statement) then pretend it's really model code 0x92 (the line you
    asked about) and set the model_string ..." Hence the code continued
    on thinking it had really found model code 0x92 - which the code was
    commented as thinking was an E3+.

    > In the source file this line is not set at the other models. Could it be
    > that this makes a difference in functionality?

    I'm not sure I understand your question. If you mean "Was there
    anything assigned to model code 0x90?" the answer was No. That value
    was not recognized - hence the I/O error reported by the backend. (It
    didn't know how to talk to any scanner listed.)

    Hope that helps,

    Adam Benjamin
    Systems Administrator
    Open Text Corporation

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



    This archive was generated by hypermail 2b29 : Thu Apr 05 2001 - 07:38:31 PDT