----- Original Message -----
From: "Fernando Gonzalez Gentile"
> Nonono ... either I was a little asleep. or too optimistic, or both.
>
> It's Ok in the shadows boxes, but blows off at the last bright square.
> Next-to-last square is barely distinguishable.
>
> Gamma is set at 2.2, and Adobe Gamma checks it as Ok,
> But I have a 'profiled' Adobe RGB 1998 loaded, instead of sRGB.
> Then, I suppose this could be the reason why at 48% red and green gamma
> lean towards 2.35 - blue is more difficult to asset since it blends
> through a wider range of values. So does gray.
>
> I find it hard to believe that you trust more this web link that the
> hardware calibration tool.
> Is it so, absolutely? I've just bought one ! :-/
At the first day I calibrated my monitors I found the gamma was way off with
all gamma targets I can found in Internet with four computers I have at
home. I have to choose different gamma values for them when I need actual
2.2. Fortunately my wife (she has better eyes) told me the calibrated
monitor still have neutral grey. Besides the EIZO I'm using, the other
computers have cheaper monitors so they do benifit from calibration.
After an extensive search I found this LCD test site looks good, when I
remove the Spyder 2 Pro calibration and just use the monitor's own setting
(gamma 2.2) I found all the gamma at different colors/levels are just much
more accurate and close to 2.2 (with 0.1 max deviation). I also checked
other gamma targets and they were ok so I give my trust to the EIZO and this
web.
>
> Last: yes, I do see what you saw in the small .jpg
> Always, until now, considered such a small 'leaking' a result of a poor
> monitor (it's quite new, nonetheless).
>
You can confirm the problem with RGB values in software, the values increase
slowly when you move the cursor towards the tree.
> So, do you set your USM at NikonScan with large radius and small amount?
> I did exactly the other way round: radius 10% ~15%, intensity 30%,
> threshold as default or 0~1%.
>
The values depends on image size, but I will only use 8-15% intensity. Too
much will affect the contrast of the image, I don't like deviate too much
from original. For your web size image, an Amount of 13% and Radius of 10
pixel looks good.
> Are those your usual settings, or only when high contrast is present?
>
Most my scanned slides require this LCE but with different values.
> What now bothers me most in how difficult it is for the scanner to pick up
> details from the shadows.
> This sometimes makes me to bend upwards the lowest values in the RGB curve
> only, which is very tiresome and time consuming and error prone.
>
I don't have much problem with shadow details even with my Velvia 50 slides.
I think the problem is Nikonscan, other than default factory setting and
sRGB I will not trust any other one (I think I have told you my setting in
details before).
C.H.Ling
--
_________________________________________________________________
Options: http://lists.thomasclausen.net/mailman/listinfo/olympus
Archives: http://lists.thomasclausen.net/mailman/private/olympus/
Themed Olympus Photo Exhibition: http://www.tope.nl/
|