----- Original Message -----
From: "Carlos J. Santisteban"
> Hi C.H., Ken and all,
>
> From: "C.H.Ling" <ch_photo@xxxxxxxxxxxxx>
>>I think you have mixed up the capture capability (range) and display
>>capability, 4000ED has d-range of 0 to 4.2, it is enough to capture the
>>brightness and the darkest range of RVP50.
>
> I'm referring to capture range. But the figures alone are meaningless --
> OK,
> the 4000ED can "see" up to 4.2D but how clean is the captured image at
> such
> densities? I have never owned a dedicated film scanner, so my experience
> has
> been always noisy shadows -- now I have a CAnoScan 9950F, but the previous
> Agfa StudioStar was _much_ worse.
As I memtioned in my previous email, I see the limit is film grain not the
scanner, at least my 4000ED. No experience with the scanners you have but my
Epson 4870 is not bad either.
>
>>I don't see why you need to brighten the shadows, brighten the shadows
> means
>>you are reducing the display d-range.
>
> Sometimes I have to increase the contrast of an underexposed slide. Or
> simply want to reduce _global_ D-range (in order to fit inside limited
> display capabilities) without sacrificing _local_ contrast -- just like I
> would do in the darkroom thru dondging and burning-in.
>
If I have to increase the shadow of a slide to an extend that objectionable
noise/grain is visible that means the slide has very serious exposure error,
I haven't encounter this before. I also don't like HDR type processing,
shadow is shadow, it won't be an important object, just let it look darker,
for darker area noise is not that critical for visual even it is available.
Underexposed slides usually has high contrast, low contrast problem is from
your scanner when you try to boost up the brightness.
>>Here is a direct scan of RVP50 slide with high contrast scene, from the
>>histrogram you can see 4000ED has captured almost every details of the
>>brightest and darkest range.
> <snip>
>><www.accura.com.hk/temp/RVP-direct.jpg>
>
> Seems well captured. However, histograms are meaningless to me -- they
> wouldn't tell noise from actual shadow detail...
>
>>Here is the adjusted version, which look more close to the original slide
>>view through lightbox:
>>
>><www.accura.com.hk/temp/RVP-fit-vision.jpg>
>
> It looks much stronger, for sure. But I still prefer the first version,
> which has much more shadow detail.
It is not which one is more preferable just the higher contrast one look
more like the original slide.
>
>>Also, I don't see limtation on any negative even T-Max 100 since the
> d-range
>>of negative is much more narrow than slide.
>
> T-Max 100 doesn't have a big density range, IME... a strongly developed
> Pan-F or the defunct Technical Pan can be as dark as a slide!
>
T-Max 100 has the strongest contrast I will accept, I don't see I want
higher contrast B/W film. If so, I will just do with post processing.
>>At the mean time how many
>>bit/channel doesn't indicate the capture d-range capability of the
>>scanner,
>>too less bit only create potential bending problem for post editing.
>
> Theoretically, each 0.3D equals 1 bit (linear PCM scale), so 4.2 would
> need
> no less tha 14 bits/channel... but again, the A/D conversion may not be
> linear and/or noise could invalidate the data from the least significant
> bits. Just like digital multimeters, adding another digit does increase
> the
> price steeply, not just for the extra digit itself (almost free) but for
> the
> x10 required precision.
As mentioned before, banding is the only problem with lower bit number AND
if you do processing. If the image is ready for print, 8 bit/channel is what
most people can use. BTW, it is already a dream if you can see 256 different
shade with your monitor (I mean if you can identify the different between
222 and 223, 224 and 225...etc), just don't know what one can see with
print.
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/
|