kopia lustrzana https://gitlab.com/sane-project/website
111 wiersze
5.6 KiB
HTML
111 wiersze
5.6 KiB
HTML
<!-- received="Tue Mar 16 08:54:20 1999 PST" -->
|
||
<!-- sent="Tue, 16 Mar 1999 10:53:12 -0600" -->
|
||
<!-- name="john allen pitney" -->
|
||
<!-- email="pitney@students.uiuc.edu" -->
|
||
<!-- subject="Re: HP Photosmart R032 problem" -->
|
||
<!-- id="" -->
|
||
<!-- inreplyto="19990316172239.B3025@bitterling.LeidenUniv.nl" -->
|
||
<title>sane-devel: Re: HP Photosmart R032 problem</title>
|
||
<h1>Re: HP Photosmart R032 problem</h1>
|
||
<b>john allen pitney</b> (<a href="mailto:pitney@students.uiuc.edu"><i>pitney@students.uiuc.edu</i></a>)<br>
|
||
<i>Tue, 16 Mar 1999 10:53:12 -0600</i>
|
||
<p>
|
||
<ul>
|
||
<li> <b>Messages sorted by:</b> <a href="date.html#165">[ date ]</a><a href="index.html#165">[ thread ]</a><a href="subject.html#165">[ subject ]</a><a href="author.html#165">[ author ]</a>
|
||
<!-- next="start" -->
|
||
<li> <b>Next message:</b> <a href="0166.html">Ewald R. de Wit: "Re: HP Photosmart R032 problem"</a>
|
||
<li> <b>Previous message:</b> <a href="0164.html">J. Milgram: "Understanding resolution."</a>
|
||
<!-- nextthread="start" -->
|
||
<!-- reply="end" -->
|
||
</ul>
|
||
<!-- body="start" -->
|
||
On Tue, Mar 16, 1999 at 05:22:39PM +0100, Ewald R. de Wit wrote:<br>
|
||
<i>> john allen pitney (<a href="mailto:pitney@students.uiuc.edu">pitney@students.uiuc.edu</a>) wrote:</i><br>
|
||
<i>> </i><br>
|
||
<i>> > This slide isn't particularly dark, but the image I get with a gamma of </i><br>
|
||
<i>> > 1.0 is simply unusable. A histogram shows that it has almost no </i><br>
|
||
<i>> > pixels with a value above about 150 (out of 255). This is with the </i><br>
|
||
<i>> > recommended 150% exposure. Applying a gamma and white point correction </i><br>
|
||
<i>> > to the scanned image does fix it up, but, as I pointed out, it leaves </i><br>
|
||
<i>> > an image with very many large gaps in its histogram and markedly reduced </i><br>
|
||
<i>> > color resolution. </i><br>
|
||
<i>> </i><br>
|
||
<i>> 150% is not by any means a recommended value (it's just the HP default</i><br>
|
||
<i>> value for slides). Try a higher value and you'll get a better picture</i><br>
|
||
<i>> with a somewhat smoother histogram. You mustn't set it too high</i><br>
|
||
<i>> though otherwise you loose all detail in the highlights.</i><br>
|
||
<p>
|
||
Hmm, it seems like the tooltip recommended 150% for normal slides. I'll <br>
|
||
experiment with it.<br>
|
||
<p>
|
||
<i>> > Is it the right thing to do, to acquire an 8-bit/channel image with a </i><br>
|
||
<i>> > flat 1.0 gamma curve, then apply a gamma of 2.2 to the result? </i><br>
|
||
<i>> </i><br>
|
||
<i>> No it's not the right thing to do but it's the best that can be done</i><br>
|
||
<i>> right now...</i><br>
|
||
<p>
|
||
Hmm, OK. Maybe I should put together some example scans. I think we can <br>
|
||
do better than what I'm seeing. Maybe there's a config problem on my end.<br>
|
||
<p>
|
||
<i>> > I agree there--the densitometric data should be in the frontend. My </i><br>
|
||
<i>> > concern is that possible the HP PS R032 can't output a usable 24-bit </i><br>
|
||
<i>> > image unless there is a (10x8?) tone map downloaded to it (which seems to </i><br>
|
||
<i>> > be happening in the VueSmart logs). </i><br>
|
||
<i>> </i><br>
|
||
<i>> As far as I understood it Vuesmart does the tonemapping in software.</i><br>
|
||
<i>> You might have mistaken the calibration data for the tonemap.</i><br>
|
||
<i>> </i><br>
|
||
<i>> Revision 032 of the photosmart doesn't seem to allow any tonemap up or</i><br>
|
||
<i>> downloading so we will actually be forced to do it in software.</i><br>
|
||
<p>
|
||
Well, from the log I put up <br>
|
||
(<a href="http://www.students.uiuc.edu/~pitney/vuesmart1.log">http://www.students.uiuc.edu/~pitney/vuesmart1.log</a>)<br>
|
||
<p>
|
||
I interpreted this as a tone map download:<br>
|
||
<p>
|
||
time 0 command 0a flags 54 status 01 length 5 *a5T<br>
|
||
time 0 command 0a flags 54 status 01 length 5 *a6D<br>
|
||
time 0 command 0a flags 54 status 01 length 8 *a1024W<br>
|
||
time 0 command 0a flags 54 status 01 length 1024 <br>
|
||
<p>
|
||
According to <a href="http://www.dmo.hp.com/devexchange/nam/scanjet/sclv9.htm">http://www.dmo.hp.com/devexchange/nam/scanjet/sclv9.htm</a> the <br>
|
||
*a6D means a 10x8 tone map is about to be downloaded. Am I reading this <br>
|
||
right? I take that last line above to be the downloading of 10-bits <br>
|
||
worth of gamma table. <br>
|
||
<p>
|
||
<i>> > I just wanted to see how VueSmart was doing such a good job! So far, I </i><br>
|
||
<i>> > haven't managed to get a comparable image from xscanimage/hp backend. Doing </i><br>
|
||
<i>> > so will require either manipulation of 10-bit data in software or </i><br>
|
||
<i>> > downloading a (10x8?) tone map to the scanner. </i><br>
|
||
<i>> </i><br>
|
||
<i>> Yes, the current state of affairs does suck somewhat. I'm writing a</i><br>
|
||
<i>> new frontend for the KDE project that will address these and other</i><br>
|
||
<i>> concerns but it will be a few months till a public release though.</i><br>
|
||
<p>
|
||
KDE!?! :-) Anyway, enjoy your project. I think that xscanimage/hp backend <br>
|
||
is pretty close right now, it just needs a few more tweaks. <br>
|
||
<p>
|
||
<i>> > By the way, what does the 10x8 notation mean? Map 10-bit data to 8-bit? </i><br>
|
||
<i>> </i><br>
|
||
<i>> Yep!</i><br>
|
||
<p>
|
||
Thanks!<br>
|
||
<p>
|
||
John<br>
|
||
<p>
|
||
<p>
|
||
<pre>
|
||
--
|
||
Source code, list archive, and docs: <a href="http://www.mostang.com/sane/">http://www.mostang.com/sane/</a>
|
||
To unsubscribe: echo unsubscribe sane-devel | mail <a href="mailto:majordomo@mostang.com">majordomo@mostang.com</a>
|
||
</pre>
|
||
<!-- body="end" -->
|
||
<p>
|
||
<ul>
|
||
<!-- next="start" -->
|
||
<li> <b>Next message:</b> <a href="0166.html">Ewald R. de Wit: "Re: HP Photosmart R032 problem"</a>
|
||
<li> <b>Previous message:</b> <a href="0164.html">J. Milgram: "Understanding resolution."</a>
|
||
<!-- nextthread="start" -->
|
||
<!-- reply="end" -->
|
||
</ul>
|