kopia lustrzana https://gitlab.com/sane-project/website
127 wiersze
4.9 KiB
HTML
127 wiersze
4.9 KiB
HTML
<!-- received="Mon May 12 15:25:19 1997 MST" -->
|
|
<!-- sent="Mon, 12 May 1997 23:59:08 +0200" -->
|
|
<!-- name="Thomas Blume" -->
|
|
<!-- email="mblum1@nightshift.uni-paderborn.de" -->
|
|
<!-- subject="Mustek - Speed - Buffer" -->
|
|
<!-- id="199705122159.XAA06163@nightshift.uni-paderborn.de" -->
|
|
<!-- inreplyto="" -->
|
|
<title>sane-devel: Mustek - Speed - Buffer</title>
|
|
<h1>Mustek - Speed - Buffer</h1>
|
|
<b>Thomas Blume</b> (<a href="mailto:mblum1@nightshift.uni-paderborn.de"><i>mblum1@nightshift.uni-paderborn.de</i></a>)<br>
|
|
<i>Mon, 12 May 1997 23:59:08 +0200</i>
|
|
<p>
|
|
<ul>
|
|
<li> <b>Messages sorted by:</b> <a href="date.html#142">[ date ]</a><a href="index.html#142">[ thread ]</a><a href="subject.html#142">[ subject ]</a><a href="author.html#142">[ author ]</a>
|
|
<!-- next="start" -->
|
|
<li> <b>Next message:</b> <a href="0143.html">becka@sunserver1.rz.uni-duesseldorf.de: "Re: Mustek Scanner distorted image lines"</a>
|
|
<li> <b>Previous message:</b> <a href="0141.html">Stefan Luettringhaus-Kappel: "Mustek Scanner distorted image lines"</a>
|
|
<!-- nextthread="start" -->
|
|
<!-- reply="end" -->
|
|
</ul>
|
|
<!-- body="start" -->
|
|
Hello David,<br>
|
|
<p>
|
|
<i>> Thomas> what is a reasonable size for SG_BIG_BUF? Is it possible to</i><br>
|
|
<i>> Thomas> prepare configure for profiling ?</i><br>
|
|
<i>> </i><br>
|
|
<i>> I'm using 128KB which works nicely.</i><br>
|
|
<p>
|
|
Well, in the meantime I tried larger buffers and I noticed, that 127.5K<br>
|
|
is the upper limit one could use. From SCSI-Programming-HOWTO:<br>
|
|
<p>
|
|
13.1. Transfer Lengths<br>
|
|
<p>
|
|
[...]<br>
|
|
generic header as well as the command block on input. SG_BIG_BUFF can<br>
|
|
be safely increased upto (131072 - 512). To take advantage of this, a<br>
|
|
new kernel has to be compiled and booted, of course.<br>
|
|
<p>
|
|
Perhaps this should be included in the sane-mustek man page to avoid<br>
|
|
users getting messages like this: <br>
|
|
kernel: kmalloc of too large a block (161792 bytes).<br>
|
|
<p>
|
|
At least for me the restriction in kernel mem had been unknown before<br>
|
|
this :-)<br>
|
|
<p>
|
|
<p>
|
|
<i>> However, are you saying that the Mustek backend is slower than other</i><br>
|
|
<i>> scan programs? It shouldn't be. I know there is a problem with</i><br>
|
|
<i>> 3-pass scanners, but AFAIK, 1-pass scanners perform well (with a large</i><br>
|
|
<i>> enough scsi buffer).</i><br>
|
|
^^^^^^^^^^^^^^^^^^ <br>
|
|
<p>
|
|
I just had the time to make a small comparison in speed for different<br>
|
|
resolutions. NT 4.0 data was obtained with Picture-Publisher 6.0<br>
|
|
(bundled with MFS-1200SP), Mustek Twain module, Color scanning, times<br>
|
|
stopped "by hand".<br>
|
|
Linux data with sane-0.57, Linux 2.0.29, used the coordinates<br>
|
|
I took from NT.<br>
|
|
<p>
|
|
time scanimage -d mustek --speed Normal --mode Color \<br>
|
|
-x 102.1 -y 146.5 --resolution $res > out.pnm 2>> scantimes.log<br>
|
|
<p>
|
|
Pentium 133, 48MB, Adaptec 1542C<br>
|
|
<br>
|
|
dpi sec Linux sec NT<br>
|
|
<br>
|
|
20.00 26.81 <br>
|
|
40.00 26.87 <br>
|
|
60.00 27.43 <br>
|
|
80.00 28.45 30.00 <br>
|
|
100.00 29.24 31.00 <br>
|
|
120.00 30.31 31.00 <br>
|
|
140.00 31.72 31.00 <br>
|
|
160.00 46.01 46.00 <br>
|
|
180.00 47.68 47.00 <br>
|
|
200.00 49.91 48.00 <br>
|
|
220.00 52.55 48.00 <br>
|
|
240.00 55.06 48.00 <br>
|
|
260.00 55.57 49.00 <br>
|
|
280.00 61.10 49.00 <br>
|
|
300.00 64.50 49.00 <br>
|
|
320.00 272.36 81.00 <br>
|
|
340.00 282.31 81.00 <br>
|
|
360.00 287.75 82.00 <br>
|
|
380.00 300.88 82.00 <br>
|
|
400.00 309.66 83.00 <br>
|
|
560.00 415.46 88.00 <br>
|
|
580.00 435.00 89.00 <br>
|
|
600.00 438.76 90.00 <br>
|
|
620.00 296.01 157.00<br>
|
|
640.00 886.09 158.00<br>
|
|
<p>
|
|
Ok, scanning with high resolutions is not what I do normally, but <br>
|
|
maybe some others do so. The difference is quite significant.<br>
|
|
The "low" number at 620 dpi under Linux is reproduceable.<br>
|
|
<i>>From the number of "stops" under NT I estimate the buffer used to</i><br>
|
|
about 2MB. <br>
|
|
Each start/stop of the scanner sums up to a significant contribution<br>
|
|
to scan times. <br>
|
|
<p>
|
|
<p>
|
|
But how to obtain a larger buffer??<br>
|
|
<p>
|
|
<p>
|
|
Thomas<br>
|
|
<p>
|
|
<pre>
|
|
--
|
|
Thomas Blume <<a href="mailto:mblum1@hrz.uni-paderborn.de">mblum1@hrz.uni-paderborn.de</a>>
|
|
Technische Chemie und Chemische Verfahrenstechnik, UNI-GH-Paderborn
|
|
Tel.: +49 5251 602598 Fax: +49 5251 603244
|
|
<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="0143.html">becka@sunserver1.rz.uni-duesseldorf.de: "Re: Mustek Scanner distorted image lines"</a>
|
|
<li> <b>Previous message:</b> <a href="0141.html">Stefan Luettringhaus-Kappel: "Mustek Scanner distorted image lines"</a>
|
|
<!-- nextthread="start" -->
|
|
<!-- reply="end" -->
|
|
</ul>
|