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 1:29 PM, David J. Littleboy wrote:
> It don't work that way<g>.
>
> The 5D user shoots at ISO 400 with the same image quality (photon shot
> noise) and same shutter speed and sees the same DOF (and same
> background
> blurring effects) at f/4.0 as the 4/3 user does at f/2.0.
>
> It is seriously cool how digital cameras with the same pixel count
> scale
> across formats.
>
> (At ISO 100, the 5D should have a two stop dynamic range advantage,
> except
> that the A/D converters don't have enough bits.)
So you have an unrealized two-stop advantage at low ISO. I can see
how important that unrealized potential could be. ;-)
> The bottom line is that if you think a smaller format buys you
> anything
> other than lighter weight/smaller size/lower price, you've done
> your math,
> physics, and/or optics wrong.
Theoretically. Funny how things don't always work that way practically.
> You are already shooting two stops smaller with the 5D for the same
> DOF. And
> for portrait work, you don't shoot at f/4.0 with FF, you shoot at f/
> 2.0 and
> wider. For a DOF effect that simply isn't available from the 4/3
> format.
> (Although I wish Canon had an 75 or 85/1.4. The f/1.2 is overmuch.)
If you shoot portraits exclusively then the selective focus issue is
always going to be your overriding priority. The larger the "film"
the shallower the DOF. Large format is your friend in the studio. Of
course, Olympus doesn't actually have a single decent portrait lens
in their lineup. If that's the kind of work you do then the 4/3 line
of cameras and optics isn't something to be considered.
> That's the difference with digital: you can get a reasonable 10MP
> image from
> the 4/3 camera at ISO 100. You really can't get a reasonable film
> image from
> 1/4 the area of 35mm.
Well, it kind of depends. With cinema cameras you used to always be
fighting against generation loss. I think I can get better IQ from a
16mm scanned negative than we used to get from a 35mm negative that
had gone through four or five generation losses. This would make 16mm
an ideal format for television if those productions were still shot
like they were 20 years ago, but with faster film stocks the
evolution of the medium has favored using less lighting for heat/
power cost savings as well as the need for less crew. 35mm using ISO
500 stocks (pretty much the standard now) doesn't translate down to
16mm because the apparent grain signature will be more dominant.
> Again, if you are using a 10MP 4/3 camera, then the comparison is
> with the
> 70-200/4.0 (IS).
I know you like that f/4 comparison, but like you said earlier, with
the A/D converters as they are you aren't seeing a dynamic range
advantage at low ISO, so the comparison doesn't hold. Unless you're
still dwelling on DOF. Any excuse to erect a straw man? :-)
> The "telecentric" bit strikes me as nothing other than lying snake
> oil.
Heh...makes you feel better about that CMOS dust-magnet you bought? ;-)
> At the end of the day, one shoots a camera that meets one's needs.
> If the
> 4/3 meets your needs, there's no reason to move to a larger format
> (just
> don't try to tell me that it's better; it ain't).
It's better at some things, certainly. If, for example, you're doing
forensic work you have additional DOF and since you can use lower
stops you extend the range of your strobes.
> Just as 645 meets my needs
> but not the needs of someone making larger landscape prints.
I prefer my 6x7. ;-)
> Foveon doesn't buy you anything the human eye can actually see. And
> not
> using a low-pass filter reduces real resolution by it's snap-to-
> grid effect
> which puts features in the wrong place; it's an artificial
> sharpening trick
> at best.
Foveon, and actually any capture medium that delivers 4:4:4 color,
should really shine when you start manipulating the image in post.
The more color timing you do the quicker a Bayer image will fall
apart when compared to, say, the image from a scanning back. I assume
Foveon will hold up the same way, but the implementation of the
technology seems shaky at best right now.
----------------------------------------------------------------------------------------
Unsubscribe by mail to listserver@halftone.co.uk, with 'unsubscribe
filmscanners'
or 'unsubscribe filmscanners_digest' (as appropriate) in the message title or
body
|