I feel a bit awkward here. I'm not really pushing digital. I"ve just
spent the evening happily playing with a roll of FILM, gasp. :-) On
the other hand, I like to keep things straight.
I don't know about Winsor, but I am regularly up in the wee hours. I
have no schedule to keep most days and have always been a night owl when
external requirements (like earning a living, school etc.) have been absent.
Chris Barker wrote:
>Hmm, interesting and persuasive, but not conclusive (for me). The
>camera is a full-frame sensor and so defeats my arguments about
>relative size, fine, but the camera is likely to be processing the
>image once the shot is taken, so removing the noise in the sky (which I
>admit I get a lot of in my very best Velvia shots).
>
The camera certainly corrects for any known bad pixels. However, those
digital sensors in DSLRs are essentially noise free at 100 iso, which is
what Michael used in his comparison. Quite correct too, as the film was
also 100 iso. If you look at the sampler I posted, all the image pieces
except the far right are as they came out of RAW files, i.e. without in
camera procesing, with no noise reduction in the RAW conversion or after
that <http://www.moosemystic.net/Gallery/Compar2FM1600sa.jpg>. Except,
of course, the far right samples, which have had noise reduction applied
to the immediately adjacent 1600 iso samples. I just can't see any noise
in the 100 iso sample and have to kind of talk myself into seeing any in
the 200. I specifically included the smooth oof areas to see the kind of
noise that shows up in featureless sky. And it is there compared to the
busier samples at higher isos. Here is a sample at full pixel size of a
sky shot at iso 200 on the 300D, no noise reduction in RAW processing
<http://www.moosemystic.net/Gallery/skysample.jpg>. Since the 1Ds is top
of the line and my little (thank goodness!) 300D is bottom of the line,
I think you may be assured that the lack of sky noise in his test is for
real.
There is a flip side. The low iso samples look sort of soft and blah.
Although there is noticeable noise in the top samples starting at 400,
the bottom samples start looking better at that point, more interesting,
more apparently detailed. Of course, one can use sharpening, local
contrast enhancement, and other tools to perk up the low iso samples,
but the point is that different isos have different inherent subjective
'looks' and choice of iso, like choice of film, can be part of the
creative process. Noise isn't inherently good or bad, it just is what it
is, and may be used to further one's ends.
>My counter-argument would be that I can sharpen my slide scans to make it
>look as good as that.
>
And yet, the digital image can be sharpened as well. The point is that
he tried to do a fair and comparable comparison and with the same lens,
within moments of the same time, the digital full frame imager produced
more resolution than the film. It also produced less noise/grain.
>Finally, the price of this camera makes it an unfeasible choice for most
>people who carry out a simple "investment
>appraisal".
>
True, true. Yet I seem to recall statements about what film could do
that digital of the same sensor size couldn't. And as I recall, they
were made without consideration, at least not in writing, of relative
prices.
Again, I see film and digital as media that are supeficially similar,
but differ enough in the details to make a difference in what they are
best suited for. And the higher reaches of both cost money.
Moose
==============================================
List usage info: http://www.zuikoholic.com
List nannies: olympusadmin@xxxxxxxxxx
==============================================
|
