kopia lustrzana https://gitlab.com/sane-project/website
93 wiersze
4.7 KiB
HTML
93 wiersze
4.7 KiB
HTML
<!-- received="Fri Jul 25 17:49:38 1997 MST" -->
|
|
<!-- sent="Fri, 25 Jul 1997 17:30:10 -0700" -->
|
|
<!-- name="David Mosberger-Tang" -->
|
|
<!-- email="davidm@azstarnet.com" -->
|
|
<!-- subject="Re: More on Mustek 8000SP adventures..." -->
|
|
<!-- id="199707260030.RAA01739@panda.mosberger" -->
|
|
<!-- inreplyto="199707241547.RAA03244@cave.BitWizard.nl" -->
|
|
<title>sane-devel: Re: More on Mustek 8000SP adventures...</title>
|
|
<h1>Re: More on Mustek 8000SP adventures...</h1>
|
|
<b>David Mosberger-Tang</b> (<a href="mailto:davidm@azstarnet.com"><i>davidm@azstarnet.com</i></a>)<br>
|
|
<i>Fri, 25 Jul 1997 17:30:10 -0700</i>
|
|
<p>
|
|
<ul>
|
|
<li> <b>Messages sorted by:</b> <a href="date.html#137">[ date ]</a><a href="index.html#137">[ thread ]</a><a href="subject.html#137">[ subject ]</a><a href="author.html#137">[ author ]</a>
|
|
<!-- next="start" -->
|
|
<li> <b>Next message:</b> <a href="0138.html">David Mosberger-Tang: "Re: HP ScanJet 5p in SANE 0.6"</a>
|
|
<li> <b>Previous message:</b> <a href="0136.html">Rogier Wolff: "Re: More on Mustek 8000SP adventures..."</a>
|
|
<li> <b>In reply to:</b> <a href="0134.html">Rogier Wolff: "Re: More on Mustek 8000SP adventures..."</a>
|
|
<!-- nextthread="start" -->
|
|
<li> <b>Next in thread:</b> <a href="0135.html">becka@sunserver1.rz.uni-duesseldorf.de: "Re: More on Mustek 8000SP adventures..."</a>
|
|
<!-- reply="end" -->
|
|
</ul>
|
|
<!-- body="start" -->
|
|
<i>>>>>> On Thu, 24 Jul 1997 17:47:42 +0200 (MET DST), <a href="mailto:R.E.Wolff@BitWizard.nl">R.E.Wolff@BitWizard.nl</a> (Rogier Wolff) said:</i><br>
|
|
<p>
|
|
<i> >> Hmmh, I don't think I would want to _force_ this behavior, since</i><br>
|
|
<i> >> it would slow down scanning by quite a bit. However, maybe we</i><br>
|
|
<i> >> could put an option into mustek.conf that lets you limit the</i><br>
|
|
<i> >> maximum distance scanned in one chunk. Then we can distribute</i><br>
|
|
<i> >> SANE with a mustek.conf that is safe for owners of an unpatched</i><br>
|
|
<i> >> Linux ncr810 driver. This leaves the option of getting back the</i><br>
|
|
<i> >> old behavior when it's known to be safe.</i><br>
|
|
<p>
|
|
Rogier> The Tamarack looses 2 seconds on each "read". At 32k/read<br>
|
|
Rogier> and 25Mb per scan, that comes to 2000 seconds. In this case<br>
|
|
Rogier> the buffer size is the limiting factor.<br>
|
|
<p>
|
|
OK, I ended up adding an option to the mustek.conf that is called<br>
|
|
`strip-height'. It lets us limit the maximum strip height scanned with<br>
|
|
a single SCSI command. The syntax is:<br>
|
|
<p>
|
|
option strip-height HEIGHT<br>
|
|
<p>
|
|
where HEIGHT is a floating point number specifying the height in<br>
|
|
inches.<br>
|
|
<p>
|
|
My plan is for all future SANE distributions will ship with a default<br>
|
|
strip-height of 1 inch. This should be safe even for the ncr810<br>
|
|
driver.<br>
|
|
<p>
|
|
Limiting the strip-height does not seem to affect scan performance in<br>
|
|
lineart and grayscale mode on my scanner (Paragon 600 II SP). The<br>
|
|
story is quite different for color mode though:<br>
|
|
<p>
|
|
strip-height: scan time (@ 18 dpi):<br>
|
|
------------- ---------------------<br>
|
|
1" 78sec<br>
|
|
2" 56sec<br>
|
|
unlimited 40sec<br>
|
|
<p>
|
|
A strip-height of 2" seems to take about 7 seconds to scan, which is<br>
|
|
about as close as I would want to get to the 10sec timeout in the<br>
|
|
ncr810 driver. As you can see, this limit causes a slow down by 16<br>
|
|
seconds, which certainly exceeds my level of patience. Thus, I'd<br>
|
|
still recommend to patch the ncr810 driver where applicable and remove<br>
|
|
the `strip-height' option. That will give you the old level of<br>
|
|
performance. Note that you don't have to do anything if you have SANE<br>
|
|
already installed and working---the "make install" step does not<br>
|
|
overwrite existing configuration files, hence you don't have to worry<br>
|
|
about getting slow scans when upgrading in the future.<br>
|
|
<p>
|
|
BTW: Did anyone try out the pre0.62 patch? I haven't gotten any<br>
|
|
reports so far.<br>
|
|
<p>
|
|
--david<br>
|
|
<p>
|
|
<pre>
|
|
--
|
|
Source code, list archive, and docs: <a href="http://www.azstarnet.com/~axplinux/sane/">http://www.azstarnet.com/~axplinux/sane/</a>
|
|
To unsubscribe: mail -s unsubscribe <a href="mailto:sane-devel-request@listserv.azstarnet.com">sane-devel-request@listserv.azstarnet.com</a>
|
|
</pre>
|
|
<!-- body="end" -->
|
|
<p>
|
|
<ul>
|
|
<!-- next="start" -->
|
|
<li> <b>Next message:</b> <a href="0138.html">David Mosberger-Tang: "Re: HP ScanJet 5p in SANE 0.6"</a>
|
|
<li> <b>Previous message:</b> <a href="0136.html">Rogier Wolff: "Re: More on Mustek 8000SP adventures..."</a>
|
|
<li> <b>In reply to:</b> <a href="0134.html">Rogier Wolff: "Re: More on Mustek 8000SP adventures..."</a>
|
|
<!-- nextthread="start" -->
|
|
<li> <b>Next in thread:</b> <a href="0135.html">becka@sunserver1.rz.uni-duesseldorf.de: "Re: More on Mustek 8000SP adventures..."</a>
|
|
<!-- reply="end" -->
|
|
</ul>
|