Re: COOLSCAN II backend - resend
Rogier Wolff (R.E.Wolff@BitWizard.nl)
Tue, 5 Aug 1997 22:29:21 +0200 (MET DST)
Michael K. Johnson wrote:
>
>
> carlierd@acm.org writes:
> >Sorry if you did receive the first message, but I don't think it went
> >through ...
>
> I don't think so either. Leastways, I don't remember getting it.
>
> >Is there anybody working on this already, and, if not, what would be
> >the best starting point among the already available backends for SCSI
> >scanners ?
>
> I'm not aware of any.
>
> You might consider looking at the umax driver. I don't know that you
> would want to start from it exactly, but umax scanners do a pretty good
> job of following the scsi 2 scanner standard (part 14, if memory serves
> me right).
>
> However, that standard allows a LOT of potential manufacturer-specific
> additions. I see only two ways to write a driver for a scsi scanner:
> either get protocol documentation from the manufacturer, or use a scsi
> sniffer to watch the protocol. With the aid of scsi 2 part 14, you can
> get a long way, then modify one thing at a time and figure out enough
> of the rest. I used both; I got a working driver from the standards,
> and someone with a sniffer and the same scanner helped me figure out
> a few confusing parts of the documentation faster. The problem is that
> scsi sniffers are expensive. The cheapest one I can find is nearly
> $2000, if I recall correctly. Definitely go with the protocol docs
> if you can get them. I hope Nikon is cool about protocol docs...
He-he... I traced the Windows software talking to the Coolscan II
using a $50,000 logic analyser..... :-)
Anyways, we attempted to adapt the tamarack (which doesn't follow
the SCSI standard too closely) program that I had, but that didn't
lead to quick success. We could try using the umax backend.....
Roger.
--
Source code, list archive, and docs: http://www.azstarnet.com/~axplinux/sane/
To unsubscribe: mail -s unsubscribe sane-devel-request@listserv.azstarnet.com