Maybe I need to be more explicit... my experience is the following: if you have a large subject (a tennis player for instance) filling a reasonable part of the picture, or a car for instance, the system will work very well, but for fast moving subjects covering only a few sensors, the system is unreliable and unpredictable (to me this is THE issue), I have no clue in the viewfinder about where the system is focusing. I have tons of pictures sharp on the public (tennis, basketball) or the wrong player (indoor basketball), while I first locked focus on the right spot. Now that I don't use 51 points 3D anymore, but AF-ON in combination with the long delay and 9/21 sensors, I have 95% keepers... this is my reality and I am very happy with the AF performance (fast and accurate) with those settings. This works for me and that's the most important to me.
Not doubting your experience, just giving mine. I shot surfing awhile ago and was very happy with the keeper rate. As to knowing where the AF is focusing: once you put the focus point on the subject, it tracks that subject until you click. What I needed to get used to was a red focus point flashing around the VF. The camera is focused on the subject even if the red pt seems to be lagging the subject. This mode is money to many pro shooters.
I've shot static objects with 51pt 3D and not had problems.
gugs wrote:
I have no clue in the viewfinder about where the system is focusing.
Guy
That's not right, in 3D dynamic mode the system should show you what AF rectangle it's using at all times.
The only time the system does not show you what AF points it's using is in automatic AF point selection mode when using continuous servo. Are you using the top (automatic point selection) or the second (dynamic point selection) setting on the three way AF point switch?
I've got to admit I find automatic point selection next to useless for exactly the reason you state, it doesn't show you what points it's using.
tjny wrote:
This appears like a lot of photons from fancy lenses are falling on dead space between ff sensor pixells ... makes sense.
DSLR sensors have an array of 'microlenses' that focus the light towards the sensor sites so as to mitigate the issue of 'dead space between the ...sensor pixels'.
Jammy Straub wrote:
That's not right, in 3D dynamic mode the system should show you what AF rectangle it's using at all times.
The only time the system does not show you what AF points it's using is in automatic AF point selection mode when using continuous servo. Are you using the top (automatic point selection) or the second (dynamic point selection) setting on the three way AF point switch?
I've got to admit I find automatic point selection next to useless for exactly the reason you state, it doesn't show you what points it's using.
Yes indeed, maybe I was not clear enough, I am talking about the full automatic point selection (again this works very well if you are shooting big enough subjects). Dynamic point selection and the other modes (single point) are fast and reliable.
and BTW don't misunderstand me, even if I mention a number of "cons", I still consider the D3 to be one of the best cameras ever, and it does the job for me. Don't forget that as with all other sophisticated tools, it has a serious learning curve to get the best out of it.
James R. I am of the same opinion as gugs with respect to the 51points 3d system for my use. When tracking a horse that is approaching a fence (usually i am at a 30 or 45 degree angle to the horse's direction of movement) the red indicator hops all over the viewfinder and the keeper rate is low. On the other hand using 9points with AF lock on set to medium my keeper rate is pretty much close to 100%, even when the horse is approaching head on. Typically I would be using a 70/200 usually around the 100-150 range and i tend to use a shallow DOF to reduce background clutter. This may be the reason that 51-3D does not work well for me as even a small movement in the lens by the AF system will render the subject oof or at least soft.
It may work well in other situations but an oof shot is lost money and so I have not bothered to try it out - even though I said i would in an early post on this subject.
jordanwd wrote:
ummmm , maybe i didn't make myself clear...2 shots taken from the same field of view (ie ff sensor w/ 600f4 and 1.5 crop w/ a 400f2.8 (say at f4) from the same distance yields a similiar field of view.. when examining both files from its native 14x9.5 inch size (@300ppi) they may look very similiar...but when increasing the size to a 11x17 double page spread size there is a noticable difference especially in finer detailed areas...the denser pixel concentration of the 1.5 sensor yeilds more detail than the ff sensor (this is w/o up rezzing) ... now does that make alittle more sense?...Show more →
If FOV has been compensated for with lens choice, the pixel density and resolution of a 12MP FX or 12MP DX image made at the same print size will be the same. The FX image will show less DOF unless aperture was used to compensate.
Now if you use the same lens and you have to crop the 12MP FX file, then I believe you. But you clearly state that FOV was adjusted for with lens choice. In this case, a 12MP file is a 12MP file and the pixel density of the sensor is NOT a factor as you already compensated for that. Something else in you optical system must be accounting for what you are seeing, but it is NOT pixel density in the scenario you describe.
yes, the 400/2.8 is a sharper lens. look at their MTFs.
Herb...
jordanwd wrote:
ummmm , maybe i didn't make myself clear...2 shots taken from the same field of view (ie ff sensor w/ 600f4 and 1.5 crop w/ a 400f2.8 (say at f4) from the same distance yields a similiar field of view.. when examining both files from its native 14x9.5 inch size (@300ppi) they may look very similiar...but when increasing the size to a 11x17 double page spread size there is a noticable difference especially in finer detailed areas...the denser pixel concentration of the 1.5 sensor yeilds more detail than the ff sensor (this is w/o up rezzing) ... now does that make alittle more sense?...Show more →
Guess what my friend the google came up with