How To Detect Snapshooters from Photographers (was: Reason for so many focus errors we see today?)

whisky-dave wrote:

So you have to decide what you want in focus before taking the shot.

Yes, that has been a problem since the invention of lenses ...

For me that's another thing to add to shutter delay.

Use a pinhole camera then, and everything's in focus.

If you're taking a shot and you have a cobweb in the corner of the
shot a potrait of a person wearign a hat in the centre and a UFO
flying above thier head how would the camra know where to focus.

By you telling the camera.

I'd prefer to focus after the pictures been taken, perhaps not
possible yet, but in the future I believe that it will be possible,
and I don't mean by artificially sharpening via contrast control.

http://www.refocusimaging.com/

Your shiny new zillion megapixel camera might be capable of
4x6 inch prints that way, after tons of computations.

-Wolfgang

"Jürgen Exner" wrote in message
...
"whisky-dave" wrote:
"Savageduck" wrote in message

As far as the D300 and some other Nikon DSLRs go, up to 51 focus points
including 51 point 3D tracking, manually selectable using the multi
selector, single point AF, dynamic area AF, predictive focus tracking
(51
pnt 3D tracking), Auto-area AF.

So you have to decide what you want in focus before taking the shot.

Well, yeah, usually it is a good thing to decide what you want in focus
before taking the shot. For me that's standard practice, don't know
about others, thou.

I feel the same for exposure, by these things change on the fly.


For me that's another thing to add to shutter delay.

If you're taking a shot and you have a cobweb in the corner of the shot
a potrait of a person wearign a hat in the centre and a UFO flying above
thier head
how would the camra know where to focus.

At least my camera will cycle through different likely objects if you
release the half-pressed shutter and and then half-press it again.

That sounds like an addition to shutter delay time.
But if it works it must be worth it.


Easy if you have spot AF, but that makes framing difficult.

Why? That's the other standard way I am using: if the camera doesn't get
the focus object right by the let's say the third try then move the
center to the desired object, half-press the shutter, re-frame, and
click.

Sounds OK if you're taking pictures of a grave yard but at an air show or
sporting event or as a bird photographer.

In article , John Navas
wrote:

Easy if you have spot AF, but that makes framing difficult.

Set single area focus
Put the subject in the focus area
Half press the shutter to lock the focus
Recompose and shoot
takes way less time to do than to explain.

that sometimes works, sometimes does not. with shallow depth of field,
the difference in distance can be a problem, plus the metering bases
the exposure on when the focus was locked, not after it was recomposed,
unless those functions are split. that's possible on some cameras but
can be awkward depending on how well implemented it is.

http://visual-vacations.com/Photography/focus-recompose_sucks.htm
http://www.bobatkins.com/photography/technical/focus_recompose.html

It'd be nice to be able to get home and them decide which of the 3 (in this
case)
images should be the sharp, one, two or all 3.

Sounds like you want automatic focus bracketing.
Might be an interesting feature.

it exists.

"Wolfgang Weisselberg" wrote in message
...
whisky-dave wrote:

So you have to decide what you want in focus before taking the shot.

Yes, that has been a problem since the invention of lenses ...

and before, of course but as eyesight couldn;t be corrected there was
little point.


For me that's another thing to add to shutter delay.

Use a pinhole camera then, and everything's in focus.
That's not what most want i.e everything in focus.


If you're taking a shot and you have a cobweb in the corner of the
shot a potrait of a person wearign a hat in the centre and a UFO
flying above thier head how would the camra know where to focus.

By you telling the camera.
The same way as I tell the camera what exposure to set ???????//


I'd prefer to focus after the pictures been taken, perhaps not
possible yet, but in the future I believe that it will be possible,
and I don't mean by artificially sharpening via contrast control.

http://www.refocusimaging.com/

Your shiny new zillion megapixel camera might be capable of
4x6 inch prints that way, after tons of computations.

yes exactly but I doubt you'd need a zillion megapixel camera.
And even if you did would that be far off ?
I've a picture of IBM's first 5MB massive storage (portable) from 1956
being unloaded with being unloaded by a fork lift truck from an airplane,
I think is was a $1M worth, byut dio you think anyone at the time thought
that
a P&S camera costing less than a days wage could fill it up with just one
shot.

"John Navas" wrote in message
...
On Mon, 29 Jun 2009 17:02:44 +0100, "whisky-dave"
wrote in h2aood$bdl$1@qmul:


"Savageduck" wrote in message
news:2009062906520711272-savageduck@REMOVESPAMmecom...

As far as the D300 and some other Nikon DSLRs go, up to 51 focus points
including 51 point 3D tracking, manually selectable using the multi
selector, single point AF, dynamic area AF, predictive focus tracking
(51
pnt 3D tracking), Auto-area AF.

So you have to decide what you want in focus before taking the shot.
For me that's another thing to add to shutter delay.

If you're taking a shot and you have a cobweb in the corner of the shot
a potrait of a person wearign a hat in the centre and a UFO flying above
thier head
how would the camra know where to focus.

Better current cameras do a remarkably good job of intelligent
multi-area focusing.

True, as they do a remarkable job of accessing exposure.


Easy if you have spot AF, but that makes framing difficult.

Set single area focus
Put the subject in the focus area
Half press the shutter to lock the focus
Recompose and shoot
takes way less time to do than to explain.

true, but I've rarely found it practical.
especailyl on subjects that actually move such as sporting events.



It'd be nice to be able to get home and them decide which of the 3 (in
this
case)
images should be the sharp, one, two or all 3.

Sounds like you want automatic focus bracketing.
Might be an interesting feature.

My G10 has that, tried it once, must give it another try something.

http://www.dpreview.com/forums/read....ssage=31467614
"I learned today that Photoshop CS4 is capable of blending layers of varying
focus ("focus stacking"). And I thought boy! my G10 will do focus
bracketing, something my Nikon D300 hasn't even heard of. So let's give it a
go! "

-------------------------
Ideally in the future you'd get your image up and then just click on the
area(s)
you want in focus, perhaps shift-click anything you want out of focus.
of cause this will take a lot of processing power so you'll need a computer,
but it won;t be that long before a camera could process that info itself.
afterall if you think about it the chipNpin on credit cards contain about
1000
time more processing power than the Apollo mission to the Moon used

So you have to decide what you want in focus before taking the shot.

Basic physics theory says this is impossible.

PDM

On Wed, 1 Jul 2009 14:19:23 +0100, "whisky-dave"
wrote:


My G10 has that, tried it once, must give it another try something.

http://www.dpreview.com/forums/read....ssage=31467614
"I learned today that Photoshop CS4 is capable of blending layers of varying
focus ("focus stacking"). And I thought boy! my G10 will do focus
bracketing, something my Nikon D300 hasn't even heard of. So let's give it a
go! "


Any Powershot's "focus bracketing" is remedial at best. Mostly useless. The
step size is far too small, even when set to the max setting to try to make
it useful enough. Been there done that. It's why this was such a wanted and
often voted on addition to CHDK for Powershot cameras. Use CHDK's much more
powerful focus-bracketing. Any number of shots at any step-size that you
want, from 1mm to many meters between shots, done in high-speed continuous
burst mode. In 3 flavors of step direction, +, -, and alternating +/-
steps. There are some exceptional focus-bracketing scripts too that surpass
even what CHDK can do alone (without scripts). My favorite being the one
where you focus on the subject's near focus point, then again on the
subject's far focus point. The script and camera automatically divvies-up
how many frames you need at what focusing step based on the aperture and
focal-length in use at the time, to ensure that you get all parts of your
subject in focus in the least amount of frames needed. Extremely valuable
for macro-photography buffs. No need to do the complex calculations
required. THEN use your focus-stacking software. I prefer Picolay
(freeware), much quicker, more options, less hassle, results as good as or
better than any costly bloat-ware.

PDM wrote:
So you have to decide what you want in focus before taking the shot.

Basic physics theory says this is impossible.

How does "basic physics theory" say that it is impossible for someone to
decide what he wants?

whisky-dave wrote:
"Wolfgang Weisselberg" wrote in message
whisky-dave wrote:

Use a pinhole camera then, and everything's in focus.
That's not what most want i.e everything in focus.

Blurring after the fact is easier than the opposite.

If you're taking a shot and you have a cobweb in the corner of the
shot a potrait of a person wearign a hat in the centre and a UFO
flying above thier head how would the camra know where to focus.

By you telling the camera.

The same way as I tell the camera what exposure to set ???????//

Of course not. Unless your 'exposure setting dial' also
changes the focus.

I'd prefer to focus after the pictures been taken, perhaps not
possible yet, but in the future I believe that it will be possible,
and I don't mean by artificially sharpening via contrast control.

http://www.refocusimaging.com/

Your shiny new zillion megapixel camera might be capable of
4x6 inch prints that way, after tons of computations.

yes exactly but I doubt you'd need a zillion megapixel camera.

Ok, just be aware that almost all of the pixels will be used
to record information for a focus you'll not use in the end.

And even if you did would that be far off ?

No, you can buy a 9216 × 9216 pixel (85 MPix) sensor today.
However, you probably will have to buy lenses that fill a 8 x 8
cm² image circle, as the pixel size of 8.75²µm² is reasonable.
Of course, you'll probably have to sell your Canon 1200mm f/5.6
to finance the sensor ...

What will stay a problem is the fact that photons are limited
in number. The obvious way to alleviate that problem is to loose
less photons --- any ideas how to do that in the real world,
so we get, say, twice the photons (one stop) captured?

-Wolfgang

In article , PDM
writes
So you have to decide what you want in focus before taking the shot.

Basic physics theory says this is impossible.

No it doesn't!

Basic physics theory says that if you record both intensity and phase
with adequate precision then you can reconstruct any focus, and viewing
position, that you want after the event.

Current cameras record intensity at very high spatial resolution, but
not phase at all. Holograms record phase and intensity, but at limited
spatial resolution and spectral waveband.

This is a technology limitation, not a physics limitation.
--
Kennedy
Yes, Socrates himself is particularly missed;
A lovely little thinker, but a bugger when he's ****ed.
Python Philosophers (replace 'nospam' with 'kennedym' when replying)