kopia lustrzana https://gitlab.com/sane-project/website
93 wiersze
4.6 KiB
HTML
93 wiersze
4.6 KiB
HTML
<!-- received="Tue Sep 2 14:18:20 1997 MST" -->
|
|
<!-- sent="Tue, 2 Sep 1997 18:07:43 +0200 (MET DST)" -->
|
|
<!-- name="Rogier Wolff" -->
|
|
<!-- email="R.E.Wolff@BitWizard.nl" -->
|
|
<!-- subject="Re: Backtracking in color scanning" -->
|
|
<!-- id="199709021607.SAA02473@cave.BitWizard.nl" -->
|
|
<!-- inreplyto="9709021448.AA22170@gypsy.atlantic.infi.net" -->
|
|
<title>sane-devel: Re: Backtracking in color scanning</title>
|
|
<h1>Re: Backtracking in color scanning</h1>
|
|
<b>Rogier Wolff</b> (<a href="mailto:R.E.Wolff@BitWizard.nl"><i>R.E.Wolff@BitWizard.nl</i></a>)<br>
|
|
<i>Tue, 2 Sep 1997 18:07:43 +0200 (MET DST)</i>
|
|
<p>
|
|
<ul>
|
|
<li> <b>Messages sorted by:</b> <a href="date.html#127">[ date ]</a><a href="index.html#127">[ thread ]</a><a href="subject.html#127">[ subject ]</a><a href="author.html#127">[ author ]</a>
|
|
<!-- next="start" -->
|
|
<li> <b>Next message:</b> <a href="0128.html">Allan Nielsen: "Mustek problems."</a>
|
|
<li> <b>Previous message:</b> <a href="0126.html">Michael K. Johnson: "Re: UMAX Scanner ?"</a>
|
|
<li> <b>In reply to:</b> <a href="0125.html">Henning Sauer: "UMAX Scanner ?"</a>
|
|
<!-- nextthread="start" -->
|
|
<!-- reply="end" -->
|
|
</ul>
|
|
<!-- body="start" -->
|
|
<a href="mailto:shendrix@gypsy.atlantic.infi.net">shendrix@gypsy.atlantic.infi.net</a> wrote:<br>
|
|
<i>> </i><br>
|
|
<i>> </i><br>
|
|
<i>> In message <19970901104750.53420@drawbridge.in-berlin.de> </i><br>
|
|
<i>> <a href="mailto:woju@keep.in-berlin.de">woju@keep.in-berlin.de</a> (> ) writes:</i><br>
|
|
<i>> </i><br>
|
|
<i>> > Just calcualte the Data the scanner needs to transfer.</i><br>
|
|
<i>> > As Far as I could find the Mustek Scanner justs uses async transfer</i><br>
|
|
<i>> > to the HOST which could be as slow as 1.5MB/S (even if you Adapter</i><br>
|
|
<i>> > could get up to 7MB/s )</i><br>
|
|
<i>> > An tiny Area of 1" times 1" brings you with 24 Bit and 1200 dpi Color</i><br>
|
|
<i>> > into 4320000 Bytes of Imagedata. if the scanner is able to scan that within</i><br>
|
|
<i>> > 1 Second the scanner needs to stop to get rid of the</i><br>
|
|
<i>> > Data.</i><br>
|
|
<p>
|
|
I did some calcualations, but my scanner never hits the data transfer<br>
|
|
rate problem. It seems to get slower at higher resolutions all by<br>
|
|
itself.<br>
|
|
<p>
|
|
<i>> </i><br>
|
|
<i>> Well, under Windows95 it usually never did backtracking. When it did,</i><br>
|
|
<i>> it only backed up half the number of times it does with sane. This</i><br>
|
|
<i>> was with the little cheap controller it came with too.</i><br>
|
|
<p>
|
|
So, they are using a buffer that is twice as large. <br>
|
|
<p>
|
|
<i>> There had got to be a way to get it as fast as under Windows. My scanner</i><br>
|
|
<i>> is backtracking every half inch sometimes. Under Win95 it always moved at</i><br>
|
|
<i>> least a couple of inches before backtracking.</i><br>
|
|
<p>
|
|
I'm suspecting that a Windows 95 driver may be able to do some tricks<br>
|
|
like scatter-gather that are not fully implemented under Linux yet. <br>
|
|
<p>
|
|
If W95 allows scatter-gather, the application allocates a megabyte of<br>
|
|
memory, requests a read from the SCSI device, and the scatter/gather<br>
|
|
will get the data into the user-pages in the right order.<br>
|
|
<p>
|
|
Under Linux, there is a sg_buffer in kernel space, which is contiguous<br>
|
|
so that devices may DMA into that area, and then the data is copied<br>
|
|
over to userspace, using the MMU for the pseudo-scatter.<br>
|
|
<p>
|
|
Microsoft operating systems have a vague distinction between<br>
|
|
user-space and operating system. Because of this, there are also<br>
|
|
some barriers that don't need overcoming under MS software. <br>
|
|
<p>
|
|
<p>
|
|
Roger.<br>
|
|
<p>
|
|
<pre>
|
|
--
|
|
** <a href="mailto:R.E.Wolff@BitWizard.nl">R.E.Wolff@BitWizard.nl</a> ** +31-15-2137555 ** <a href="http://www.BitWizard.nl/">http://www.BitWizard.nl/</a> **
|
|
Florida -- A 39 year old construction worker woke this morning when a
|
|
109-car freigt train drove over him. According to the police the man was
|
|
drunk. The man himself claims he slipped while walking the dog. 080897
|
|
<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="0128.html">Allan Nielsen: "Mustek problems."</a>
|
|
<li> <b>Previous message:</b> <a href="0126.html">Michael K. Johnson: "Re: UMAX Scanner ?"</a>
|
|
<li> <b>In reply to:</b> <a href="0125.html">Henning Sauer: "UMAX Scanner ?"</a>
|
|
<!-- nextthread="start" -->
|
|
<!-- reply="end" -->
|
|
</ul>
|