Apache-Talk @lexa.ru 

Inet-Admins @info.east.ru 

Filmscanners @halftone.co.uk 

Security-alerts @yandex-team.ru 

nginx-ru @sysoev.ru 

   


   


   















      :: Filmscanners
Filmscanners mailing list archive (filmscanners@halftone.co.uk)

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[filmscanners] Re: film and scanning vs digital photography



On Jul 7, 2007, at 7:34 AM, David J. Littleboy wrote:

> But you are forgetting to take the other aspects of the format
> difference
> into account.

This seems like an assumption. ;-)

> For the same pixel count (to a rough first approximation, 10 is
> about the
> same as 12.7), a 4/3 camera's pixels are 1/4 the area, and thus are
> two
> stops less sensitive.

Natch.

> And DOF scales with the format size, so you "gain" two stops of
> DOF. (Only
> at the wide end, at smaller apertures, diffraction kicks in two stops
> sooner, so while f/16 on FF results in sharp images, apertures
> smaller than
> f/8 on 4/3 will show diffraction effects.

But since DOF is two stops shallower you don't need to stop the lens
down as much to get the same effective DOF.

> So that sexy-sounding f/2.0 lens will be functionally
> indistinguishable from
> an f/4.0 28-70mm lens on FF (with the FF at four times the ISO for
> identical
> noise/dynamic range).

That's assuming a linear comparison of sensitivity where the 4/3
sensor is functionally two stops less sensitive than the FF sensor
across its entire ISO range, which in a technical sense it may well
be. However, 100 ISO is 100 ISO on both a FF and a 4/3 sensor. From
my experience shooting with 4/3 the images from my E-1 looked
wonderful at ISO 100-200. The combination of the lovely color
rendition of the Kodak CCD used in the camera and the microcontrast
qualities of the Zuiko glass conspired to create a beautiful capture
device. Where you started losing IQ with the E-1 was at 400 and
above. Not terrible at 400. Mostly a luminance noise pattern that
looked almost like film grain at 400. At 800 it was starting to
contain enough color speckling from the rising curve of the
chrominance noise to look more electronic.  Which comes back to that
issue of high ISO on the 4/3 chips being problematic. That doesn't
mean that you're going to suffer at low ISO, though.

So a birder, for example, will have a two-stop DOF advantage over a
FF guy right out of the gate just because of his format of choice.
Add in the faster Zuiko f/2.0 lens at ISO 100 and he can use a higher
shutter speed at a lower aperture all day long.

You're right, though, when you get to the end of the day and the
light starts to fall the extra speed of the lens becomes a crutch
that attempts to overcome the limits of the sensor. Still, the high-
end Oly glass tends to be very sharp wide open and you don't have to
stop them down much at all to hit their sweet spot.

> Note that to actually be equivalent, the 4/3 lens has to provide
> _twice_ the
> resolution (twice the lp/mm at any given MTF, or an MTF curve
> shifted up by
> a factor of two due to the finer pixel pitch) at f/2.0 than the FF
> 28-70mm
> lens does at f/4.0. (Interestingly, MTF performance does scale up with
> decreasing format sizes, so this point may not be a problem; but
> the need
> for twice the resolution at a much wider f stop may be problematic.)

This is the biggest problem with the format, IMO. You're always going
to be fighting that battle. It's the same thing with shooting 16mm
instead of 35mm cine stuff. The 16mm gear is lighter, has greater DOF
for run-and-gun work and is obviously a lot less expensive to work
with. But the frame is roughly a quarter the size of the 35mm frame,
so the glass always has to be much better than glass would have to be
on a comparable 35mm rig and obviously the grain is going to be
magnified on top of that. A grain pattern that looks subtle and
wonderful in 35mm may look really bad in 16mm, so you can't even use
the same standards of judging what stock to use because 5263 is not
the same at the end of the day as 7263 when you take the format into
consideration.

So that's the rub when you have to decide on buying glass from
Olympus now. The 35-100mm f/2 is a really nice lens. Effectively a
70-200mm f/2 lens, but it carries a price tag of $2200. Is it equal
to a Nikon 70-200mm f/2.8 on APS? Or a Canon 70-200mm f/2.8 on a FF
camera? Hard to say. More than the MTF numbers of the lens play into
it, of course. Those Canon FF cameras have a sensor with a diagonal
nearly as wide as their lens mount where the 4/3 sensor is tiny in
comparison to the 4/3 mount. That allows a lot of advantageous
geometry when it comes to lens design and how the light strikes their
sensor it a big part of the "4/3 advantage" (to quote the nauseating
Olympus PR machine).

At the end of the day I think it's about what camera you enjoy using
as much as almost anything else, unless you have some particular
application that draws you to one camera over another. I prefer CCD
sensors and my E-1 and now my D200 both have CCDs. I don't know what
options will be available to me in the future, though. I'd love to
see the Foveon chips get it together. I'd take full color information
over just about any other consideration, but so far I'm unconvinced
that they've got that format ironed-out. I really like the highlight
and color characteristics of the Fuji Super CCD SR Pro. If Olympus
could shove one of those in their new "pro" camera it would probably
override any other concerns I had about the format. Not likely, though.

-Rob

----------------------------------------------------------------------------------------
Unsubscribe by mail to listserver@halftone.co.uk, with 'unsubscribe 
filmscanners'
or 'unsubscribe filmscanners_digest' (as appropriate) in the message title or 
body



 




Copyright © Lexa Software, 1996-2009.