It's been a while since I originally brought this up, but I've pulled out the ol' scanner for another go. <br>to recap: color calibration isn't quite right on my canon lide35--the dark colors drop out to black. I turned on SCAN_FLAG_DISABLE_SHADING as you instructed, and as you predicted this recovered my lost dark colors. Where should I go from here? Should I crank up SANE_DEBUG_GENESYS and send you the scanimage output and *.pnm files?
<br><br>Thanks, Luke<br><br><br><div><span class="gmail_quote">On 3/7/06, <b class="gmail_sendername">Pierre Willenbrock</b> <<a href="mailto:pierre@pirsoft.dnsalias.org">pierre@pirsoft.dnsalias.org</a>> wrote:</span>
<blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">> - Color reproduction is better with the sane driver than the windows driver,<br>> which tends to boost the reds way too much (yaay)
<br>> - Brightness reproduction is quite poor with the sane driver. It seems that<br>> darker colors just drop off to black very quickly.<br><br>A better tool to determine if the bright and dark colors are correctly
<br>reproduced is a histogram tool. You can clearly see, if there is still<br>room at the bottom and top of the histogram(not too much, as you still<br>want some useable color values ;-)).<br><br>This reveals that dark colors are cut off for the sane driver, but it
<br>also shows that the windows driver cuts off bright colors.<br><br>Anyway, it shows a incorrect calibration on the sane side.</blockquote><div><snip> <br></div><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
To get an unshaded image, you can disable shading correction in<br>gl841_init_regs_for_scan by changing the flags(last parameter to<br>gl841_init_scan_regs, currently 0) to include SCAN_FLAG_DISABLE_SHADING.<br><br>If the image then still shows dark colors cut off, the offset/exposure
<br>calibration is to blame. Otherwise the shading calibration calculates<br>incorrect data(which i suspect).<br><br>I should put this lengthy description online somewhere..<br><br>Regards,<br> Pierre<br></blockquote></div>
<br>