AG Schnozz wrote:
> I think Bob's premise is actually pretty good.
I disagree on simple grounds of logic.
Assume two things, systems, whatever.
Under ideal circumstances, A performs at level 120 and B performs at
level 100.
A test is devised to determine which is the best. For whatever reasons,
the test design does not allow either alternative to reach maximum
potential.
Each alternative has different technology and processes. No comparison
may be made except at final results
There is no absolute or relative scale for measuring the degree by which
the performance of either alternative is degraded by test design.
Logically, it is impossible to conclude from the test which alternative
is better under ideal circumstances.
It is also impossible to know except by rerunning the test how
variations in test details might effect rankings.
One can easily imagine a test where the following occurs:
Alternative A B
Best case performance 120 100
% degradation in performance 30% 10%
Net performance 84 90
The kicker is - if we knew the above numbers, we could make a sensible
judgment. But we don't.
So, the results of the test are valid only for the specific test
conditions, and not for any other use of the alternatives. And their
future validity depends entirely on the consistency future process in
matching the test parameters.
In the case at hand, I suspect the big variable is the film scans to CD.
My own experience has been that they may vary wildly from processor to
processor, as well as between visits to the same processor.
> Not everybody is looking for the "ultimate" image 100% of the time.
Absolutely agreed. Otherwise, I'd buy some huge, expensive L glass and
get rid of my small cameras. Of course, I'd then probably take very few
pictures.
> Even I don't bother. Take last weekend as an example. I photographed an
> event and actually used my E-1 in JPEG HQ mode. Why? Because the end use
> was nothing over 8x10 and the users of the images (clients?) wanted JPEGs in
> that size. Sweet--I just shot the JPEGs in-camera and came up with 1.5MB
> image files that I could quickly throw onto a memory stick and and to them.
>
> When shooting family stuff, birthday parties of the girls, etc., I'm almost
> always in JPEG mode. Tastes great, less filling. These are snapshots and
> photoalbum things, not works of art. But once in a while I'll be shooting
> something that screams "enlargement". Ok, so I quickly change storage modes
> to RAW or RAW+JPEG, shoot and then go back to JPEG only. No problem,
> whatsoever.
>
This is meaningless to my circumstances. The vast majority of the time I
take pics, the client is me and I am seldom to know ahead of time which
will be the potential display quality image. I've been fooled before;
the shot I thought was the winner wasn't and another I took because it
was there is the winner. I'll be fooled again.
The situation you use as example is valid for two reasons that don't
apply to the test. First, you have throughly wrung out your equipment
and techinque. You know which alternative and what settings are suited
various purposes from empirical experience.
As a result, you know exactly what is required to meet the specific
requirements of the job/client. If one doesn't have those two factors,
it's risky to make such a decision.
> My only issue with Bob's testing parameters is that I think he's going a bit
> low-quality on his file settings and sizes.
For both alternatives.
> I'd up those settings a level or size, but otherwise it's quite alright.
>
> Just to put things into perspective, in the case of the Olympus E-1, you have
> to REALLY try to see any difference IN PRINT between RAW, SHQ and HQ modes.
> There will be certain textures or subjects which will cause troubles, but
> only when pushed to the limits. This, however, is not for critical work, but
> for the 99% rule, it does very very well.
>
> Which is better? Digital or film? Depends. When comparing "tangibles",
> digital stomps film. However, there are still intangible aspects of film
> which make it not only viable but desirable for certain applications. My
> OM-4T has really opened up new photographic opportunities for me.
Sorry, this was the tipping point. I can't resist. I've been quietly
chuckling ever since you first report on using the OM-4. All those years
touting the OM-2S, how it was all one could want or need, how there is
no justification for an OM-4 for your uses.......
You might consider this next time you are tempted to dis something based
on your experience with other things and analysis without actually using
it. In the mean time, I'm enjoying my chuckle - and thinking how lucky
I've been not to get caught out in a similar way - yet. :-)
I am really, really glad you are getting so much pleasure out of your
new toy/tool.
> Could I do the same thing with digital? Sure, but maybe not in the same way
> or with the same result.
>
I agree overall. Digital has huge advantages for most of my use. Yet,
after many months, I shot a roll of film in April and am 2/3 of the way
through another. I certainly got some lovely images from the April roll.
Better than digital? I don't think so. But for most practical uses, I'm
not convinced there is a real difference.
I'm not nearly as far along as a printer as you are. There may be
differences there I'm not experiencing.
> I've been working hard, lately, at some high-quality scans of 35mm images
> shot over the past 20 years. Man, I'm a good photographer! (ahem)
Oh good! I remember too well the low point a while ago, where you were
beating yourself up and some of us were saying what you just said for
yourself. You are a damn fine photographer.
> Anyway, I've been scanning some 'chromes with the V-ED using Vuescan. I'm
> doing up to FIVE passes. Is that necessary for all images? Hardly. Get
> real. I'm only doing this to this extent because these select images are
> "portfolio grade" AND I am intending on reissuing prints up to 20x30" or so.
> Yes, I'm pulling out the stops on these.
Nice.
> But nearly all images get single-pass scanning with some automatic
> adjustments going on.
But at that, they are light years ahead of consumer film processor scans
on CD.
> I'd slice my wrists if I had to do five pass scanning on every image.
>
> (BTW, why five-pass? Because the extra passes help define the shape of the
> grain.)
>
Why bother, when NR is going to get rid of it anyway? (Sound of Moose
ducking and running. Hard to do with hooves over ones ears.)
Moose
==============================================
List usage info: http://www.zuikoholic.com
List nannies: olympusadmin@xxxxxxxxxx
==============================================
|