What I find interesting is that the camera sets itself at infinity
if it cannot find focus in non flash mode. I had not observed
that, perhaps because when I first see that it doesn't get it on 2
tries, I instinctively shift the focus point toward more contrast
in the image( a habit made necessary with my 990). May be I
should be more patient before hitting the M button. I will try
that.
I have some strong doubt that it will be successful though, because
IF camera movement essentially starts the procees over at 200mm,
then the only time you will get the camera setting to infinity is
when you have been fortunate enough to hold the focus point really
steady--very hard to do with such a light camera if you are using
spot focus, but somewhat easier if you are in multi seg mode where
the camera will do some searching for you, although the focus box
is quite small and as Bryan has suggested can be moved off the
original area enough to start the process over. Doesn't this
suggest that the AF success rate will be lower in spot or flex
focus when the contrast is lower in the scene at long focal lengths
where camera shake is accentuated?
I have noticed in using my Oly B300 tele on the D7 that focusing is
appreciably better, faster and very assured. I discoverd that when
using it on my 990+2X+B300 combo (c 400mm) the focus lock
dramatically improved, and I long ago attributed that to the vastly
better contrast of the B300 over the Nikon lenses. Now, I suspect
that the same thing is happening with the D7+B300 based on your
information. I should add that I have only used this combo for
40-50 shots and have done no testing to confirm the accuracy of my
perception.
At any rate, pragmatically, aren't we better off setting to
manual/infinity if we know we will be operating at
distances/aperatures that will put us in an infinity focus context.
For one thing, the LCD shows a steady state fully focused image
from the get go, not having to be taken in and out of focus
searching for a lock that may or may not occur. (Annoying to say
the least) In many of my shooting contexts, I have found it to be
much less frustrating and a lot quicker.
Thanks again for the hard work.
dh
Then
after all 4 passes, it seems to indicate focus/no focus, it does
not guess or try to use a hyperfocal setting. I don’t know if it
uses information from all 4 passes in “guessing” or not, but it
usually seems to indicate a clean focus/no focus that is accurate.
I mostly agree with Brian but let me add a few extra points:
(Note that I am not talking about these few occasions where focus
is incorrectly established)
If focus can not be established then 2 things may happen
Without flash, the focus will always be set to infinity
With flash, the focus will be set between 1,8 and 4m or infinity
depending on scene brightness and focal length setting.
(The manual incorrectly states that with the use of flash and no
focus achieved that the distance is set between 3 and 3,8 m, but
maybe they mean this as a DOF range)
These are not random things, the camera consistently takes the same
decisions in the same circumstances
Other observations I made:
For a successful focus The camera makes either "one pass" attempt
consisting 3 in-out-in movements or either a "two pass" attempt
consisting of in-out-in-out-in
in both cases the last movement is very very short.
For an unsuccssful focus it makes a "two pass" in-out-in-out
movements.
The difference is here I think:
If the contrast is good then only one pass is used and the stepping
speed is higher.
If the contrast is poor, then two passes are used and the second
pass is executed slower to obtain better accuracy.
If no sharpness improvement whatsovever could be noticed during the
first pass, then the second pass will be shorter (only between
infinity and 1.5 meter or so) to save time.
Every time where the camera is being is significantly shaking
during focusing, a wrong focus distance will be achieved and
acknowledged as correct focus, but only for far bright objects, for
nearby or low contrast objects, a no focus indication is given.
This is not the case in contineous mode but then again, in this
mode you can never be sure that the picture you are taking is sharp
either since the red focus indicator can light up a fraction before
you press the shutter.
Anyone has different observations than mine ?
