«

Bruce <> writes:

> David Dyer-Bennet <dd-> wrote:
>>We're not to the point of having fully solved the focus problem yet, but
>>we're certainly making progress. We'll get there. In pretty much every
>>other way, it's better without mirrors .
>
>
> The Panasonic G3, GH2 and GX1 and Olympus E-P3, E-PL3 and E-PM1 all
> have the fastest and most accurate AF available for static and
> slow-moving subjects. The E-M5 should have it too. If you need fast
> AF on moving subjects, the Nikon V1 has it, although its low light
> focusing (using contrast detect AF) is on the slow side.

And I'm shooting a D700 normally.

Possibly the Olympus has improved, but the EPL-2 most certainly is not
even in the ballpark. And the V1 gets good reviews but largely from
people comparing it to the *entry-level* DSLRs, and I don't think it can
focus as many places around the frame.

So, I think we're coming along, but we're not anywhere near *there* yet
so far as I can see.

(The complexities from routing light to the phase AF sensors are a
source of expense and of error, and I'm sure in the long run it will go
away. And I'm sure that for many people, a bit slower AF that's always
*accurate* is a better tradeoff.)

> If these cameras haven't solved the focus problem, they are getting
> very close indeed. They offer faster and more accurate AF than any
> entry-level DSLR, even on moving subjects. They only fall down when
> compared with mid to high end DSLRs, and only on fast moving subjects.

Yes, exactly. But I'm using a D700 to track the jammer making her way
through the pack in roller derby.
--
David Dyer-Bennet, dd-; http://dd-b.net/
Snapshots: http://dd-b.net/dd-b/SnapshotAlbum/data/
Photos: http://dd-b.net/photography/gallery/
Dragaera: http://dragaera.info

In article <>, David J.
Littleboy <> wrote:

> >> Exactly. But to get back to my now clipped point: if you actually look at
> >> the images cameras like the D700, 5D, and 5D2 produce, they are superb.
> >> Worlds better than we ever got with 35mm film. Medium format quality with
> >> 35mm convenience. For anyone who actually takes photos and actually cares
> >> about print quality, life is really really good.
> >
> > Yes, life is extremely good that way. And in another direction, I can
> > get much cleaner pictures at high ISO than I could with film also
> > *especially* in color.
> >
> > But "good" isn't the same as "entirely satisfactory". I'm still
> > interested in more. I admit I'm less hungry than I used to be.
>
> The problem is that if you ask for more than is mathematically possible or
> engineeringly reasonable, you end up with worse than you started with.

foveon fans not only ask for what is mathematically impossible, but
they fully believe they are getting it.

> You can't correctly resolve high contrast lines just under the Nyquist
> frequency without an AA filter, so removing the AA filter reduces resolution
> (and leaves you with jaggies on every edge in every image). You do get a
> strong bogus sharpening effect, though. And if you ask for a full-color
> sensor, you get worse color and noise. And that's doubly stupid because
> Bayer sensors already have a higher ratio of color to luminance resolution
> than the human eye, which is, essentially, just an inferior Bayer sensor
> since the random arrangement of color sensors doesn't do as well as Bayer.

all true.

Alfred Molon <> wrote:
> nospam says...

>> it's not a mess at all. the output of bayer cameras is very, very good,
>> about as good as it can get.
>
>Then why do Bayer images look so soft when watched at 100%?

That says everything you need to know about sensor-technology cultists.

--
Ray Fischer | None are more hopelessly enslaved than those who falsely believe they are free.
| Goethe

Me wrote:
> IIRC he got a 5d - then a 5dII.

Seems you recall correctly. And offer less than no help on my jealousy
issues.

> Data on raw performance of the 5dIII seems to be showing that it's
> damned near exactly the same as the 5dII, which is good and not so good.
> It's good, because if you're happy with the 5dII focus system, (and
> shoot raw) then there's absolutely no point spending $3500 on a 5dIII.

Because Amazon is still selling the 5dII for $2,169.00. How much it
must suck for those early adopters, to see their tech toy lose all of
14% of it's before-inflation dollar value, with only a couple years of
photographs to show for it. Boo hoo for them.

> It's not so good, because what the heck have Canon been doing with
> sensor development over the past 4 years? There seems to have been major
> effort in improving in-camera jpeg NR, but almost no advance in reducing
> read noise.

I had a bunch of reasons for not buying a 5D, nor a 5D II.
Insufficient megapixels and excessive read noise were not among them.

"David J. Littleboy" <> writes:

> "David Dyer-Bennet" <dd-> wrote in message
> news:...
>> "David J. Littleboy" <> writes:
>>
>>> Exactly. But to get back to my now clipped point: if you actually look at
>>> the images cameras like the D700, 5D, and 5D2 produce, they are superb.
>>> Worlds better than we ever got with 35mm film. Medium format quality with
>>> 35mm convenience. For anyone who actually takes photos and actually cares
>>> about print quality, life is really really good.
>>
>> Yes, life is extremely good that way. And in another direction, I can
>> get much cleaner pictures at high ISO than I could with film also
>> *especially* in color.
>>
>> But "good" isn't the same as "entirely satisfactory". I'm still
>> interested in more. I admit I'm less hungry than I used to be.
>
> The problem is that if you ask for more than is mathematically possible or
> engineeringly reasonable, you end up with worse than you started with.

Not if the engineers are any good; good ones will smile sadly and say
"Sorry about that!"

> You can't correctly resolve high contrast lines just under the Nyquist
> frequency without an AA filter, so removing the AA filter reduces resolution
> (and leaves you with jaggies on every edge in every image). You do get a
> strong bogus sharpening effect, though. And if you ask for a full-color
> sensor, you get worse color and noise. And that's doubly stupid because
> Bayer sensors already have a higher ratio of color to luminance resolution
> than the human eye, which is, essentially, just an inferior Bayer sensor
> since the random arrangement of color sensors doesn't do as well as Bayer.

I don't have any special hankering for Foveon; I understand my own eyes
have much lower spatial resolution in color than in monochrome, and in
fact match Bayer quite well. So a Bayer image is adequate for me to
look at. No big complaints there from me.

Look, people producing professional photography in a number of areas are
enthusiastically working without AA filters (on very high-res medium
format systems). The problems you describe aren't practical imaging
problems at least in that environment.

I am kind of interested in the playing around with filter layouts Fuji
is doing in the new X<whatever>Pro model. I'm not yet convinced Bayer
is completely optimal.
--
David Dyer-Bennet, dd-; http://dd-b.net/
Snapshots: http://dd-b.net/dd-b/SnapshotAlbum/data/
Photos: http://dd-b.net/photography/gallery/
Dragaera: http://dragaera.info

"David J. Littleboy" <> writes:

> "Me" <> wrote:
>>>>
>>>> Could be even better. Get rid of the mirrors and the huge bodies and
>>>> you
>>>> have medium format quality with P&S convenience.
>>>
>>> We're not to the point of having fully solved the focus problem yet, but
>>> we're certainly making progress. We'll get there. In pretty much every
>>> other way, it's better without mirrors .
>> >
>> It is about time Nikon/Canon made some mirrorless dx/fx models with EVF.
>> As soon as this gets mentioned, dslr users seem to get defensive, but
>> there's no reason mirrorless models can't exist alongside slrs.
>
> Well, mirrorless would be nice if the electronic viewfinders and AF didn't
> blow dead camels. But they do.
>
> Next time you wish for an EVF, look through a 5D or D700 viewfinder and
> report back.

I chose to use Live View rather than looking through my D700 viewfinder
last night. Does that count? (It was more convenient for the
particular setup, and it was a completely static subject.)

> And in terms of practical photography, SLRs mirrors simply aren't the
> problem: subject motion and hand-induced camera shake are. Using a larger
> sensor with higher ISO and IS are the solutions to those problems.

The system directing light to the AF sensors is largely the cause of
systematic AF system inaccuracies in most sample cameras, though. And
the mirror slap IS quite significant around the lower shutter speeds
that would otherwise be hand-holdable. (I shoot in places where subject
motion is the limiting factor quite a lot, but I still have to worry
about other things too.)

> Now, I might be interested in a FF rangefinder camera of some sort. But
> that's a different beast altogether.

Not that different, to my mind.
--
David Dyer-Bennet, dd-; http://dd-b.net/
Snapshots: http://dd-b.net/dd-b/SnapshotAlbum/data/
Photos: http://dd-b.net/photography/gallery/
Dragaera: http://dragaera.info

On Tue, 13 Mar 2012 08:12:57 +0900, "David J. Littleboy"
<> wrote:
> Next time you wish for an EVF, look through a 5D or D700 viewfinder
and
> report back.

Well, for me, wearing glasses, I never get to see the entire frame
through any of todays dslr viewfinders. Let alone seeing the exposure
data at the same time. On top of that I find myself composing more
consciously and with better results when I use digital screens like
with P&S cameras. That does not speak in favor of evfs, but somewhat
neutralizes the advantage of large bright dslr viewfinders.

Tom

David J. Littleboy <> wrote:

> Exactly. But to get back to my now clipped point: if you actually look at
> the images cameras like the D700, 5D, and 5D2 produce, they are superb.
> Worlds better than we ever got with 35mm film. Medium format quality with
> 35mm convenience. For anyone who actually takes photos and actually cares
> about print quality, life is really really good.

> Folks who don't take photos play counting angels on heads of pins games.

If they were taking photos, would a loupe lens like the MP-E
suffice or would you need a microscope adapter? And (given
that they probably have a halo) is ISO 800 enough to freeze
them with enough DOF? Or do we need flash (and wouldn't they
go up in smoke from too much light)?

Finally, were all the angles that appeared behind powerful
enlarger lenses, or were they ... ah, somewhat taller,
preventing them from dancing on heads of pins in first place?

And what loupe lens would one use if one used a D800E?

-Wolfgang