Re: Nikon LS-1000

David Mosberger-Tang (David.Mosberger@acm.org)
Fri, 29 Aug 1997 10:47:39 -0700

>>>>> On Fri, 29 Aug 1997 15:52:10 +0200 (METDST), Per Harald Myrvang <Per.Harald.Myrvang@cc.uit.no> said:

Per> Hello! I've got my hands onto a Nikon LS-1000 dias-scanner
Per> (with the slide feeder SF-100), and [probably] being stupid,
Per> I'm going to try to write a backend for it. I've got it hooked
Per> up to a HP Apollo 735 running HPUX 9.05 (which also has a
Per> ScanJet IIc attached, which works xscanimage), which I'm going
Per> to use as a development platform.

Huh, you're saying the HP-UX scsi support works? That's news to
me---I wrote that code without being able to test it. If it really
works, please confirm in a private mail so I can update the SANE home
page. Are you using sane-0.62?

Per> Since I really don't have a clue, I'm probably going to need
Per> some help. I've found a Linux driver for Nikon LS-20 at
Per> <ftp://ftp.nvg.ntnu.no/pub/linux/sunsite/apps/graphics/capture/coolscan-0.2.tgz>,
Per> but this one specifies that it won't work on the LS-1000 (how
Per> typical!). I've contacted the Norwegian Nikon representative,
Per> in order to get technical manuals, but they've been slow to
Per> respond.

Is the LS-* series completely different from the Coolscan line of
products? Didier Carlier <carlierd@acm.org> expressed interest in a
driver for the latter. If the two are related at all, maybe a unified
backend could be written?

Per> I need to know if there are any "standard" commands for
Per> scanning, or if this differs for each scanner.

SCSI-2 defines the basic scanner commands, but there are lots of
"vendor-specific" bits. The UMAX scanners appear to be (mostly?) in
conformance with the SCSI-2 specs, so it may provide a good starting
point. If you want to start simple, I'd say that the HP backend is
about as simple as a (real) scanner backend gets. However, HP uses
its own Scanner Control Language (SCL) which is nothing like the
SCSI-2 scanner protocol.

Per> I need to know
Per> if there are anything I can do that would damage the
Per> scanner. In fact, any and all input from you guys out there
Per> would be appreciated!

It really depends. Mustek scanners can be "crashed" quite easily by
specifying too long a scan area. But I'd think that negative scanners
do not suffer from this problem. Also, given the price of negative
scanners, I'd hope they're more robust than lowest-cost flatbed
scanners.

--david

--
Source code, list archive, and docs: http://www.azstarnet.com/~axplinux/sane/
To unsubscribe: mail -s unsubscribe sane-devel-request@listserv.azstarnet.com