Panasonic's contrast focusing puts them miles ahead of any competition

nospam wrote:
John Navas wrote:

The reality is that both systems have pros and cons, that good
implementations of both systems involve some hunting, that both systems
can be made to work very well, and that claiming one is "better" than
the other is silly and pointless trolling.

nonsense. it's not trolling. phase detection is faster.

currently, the fastest contrast detect autofocus is about as good as
the slowest phase detect autofocus.

until contrast detection is as fast or faster than the fastest phase
detection, phase detection is very clearly better.

Only for speed, not for accuracy.

whether someone wants fast autofocus is another story. some people
prefer manual focus, which can be 'better' in certain scenarios, such
as macro.

"Wilba" writes:
The accuracy of the PD system is directly dependent on careful
calibration of all components, whereas the accuracy of the CD system
is not.

Hmm, it's certainly dependent on the alternate optical path (to the PD
sensors) having the same length as the actual image path, but that's
true of all traditional focusing systems (SLR focusing screens,
rangefinder optics, etc), so it's obviously an issue that's been well
handled for a very long time.

For that reason, though, a CD system may be a better choice for very
cheap cameras, as it's less affected by poor tolerances in construction.

-Miles

--
Rational, adj. Devoid of all delusions save those of observation, experience
and reflection.

Miles Bader wrote:
Wilba writes:

The accuracy of the PD system is directly dependent on careful
calibration of all components, whereas the accuracy of the CD
system is not.

Hmm, it's certainly dependent on the alternate optical path (to the PD
sensors) having the same length as the actual image path, but that's
true of all traditional focusing systems (SLR focusing screens,
rangefinder optics, etc), so it's obviously an issue that's been well
handled for a very long time.

Not always well handled. A significant number of 450Ds (including mine) need
to be returned to Canon for calibration before their PD AF works within
Canon's specification. I don't know if the defect rate is very different for
that model compared to others, or it's just that the 450D has become
renowned for it.

For that reason, though, a CD system may be a better choice for very
cheap cameras, as it's less affected by poor tolerances in construction.

And for the same reason CD AF is the choice for ultimate accuracy when speed
isn't a requirement.

"John Navas" wrote in message
...


"The single most important component of a camera
is the twelve inches behind it." -Ansel Adams

A friend of mine who used a Rollei must have agreed with that. Boy, did he
eat a lot.

Miles Bader wrote:

Hmm, it's certainly dependent on the alternate optical path (to the PD
sensors) having the same length as the actual image path,

Actually, not for AF. A PDAF depends on knowing the correct
offset. If the alternate optical path is 1mm longer and the
lens gives the best *overall* results for all wavelengths when
defocussed a tiny amount from the wavelenghts the AF is most
sensitive for --- no problem, just add up the 2 offsets, don't
drive the focus motor to where the AF sees most clearly but to
the combined offset from that position and all is well.

but that's
true of all traditional focusing systems (SLR focusing screens,
rangefinder optics, etc), so it's obviously an issue that's been well
handled for a very long time.

That's true.

For that reason, though, a CD system may be a better choice for very
cheap cameras, as it's less affected by poor tolerances in construction.

It's certainly cheaper. Large format uses a form of CD focussing,
it's called ground glass and is then replaced by the sensor
plate/film box/whatever. :-) That works well, but is not really
suited to action photography.

-Wolfgang

In rec.photo.digital Wilba wrote:
Miles Bader wrote:
Wilba writes:

The accuracy of the PD system is directly dependent on careful
calibration of all components, whereas the accuracy of the CD
system is not.

Hmm, it's certainly dependent on the alternate optical path (to the PD
sensors) having the same length as the actual image path, but that's
true of all traditional focusing systems (SLR focusing screens,
rangefinder optics, etc), so it's obviously an issue that's been well
handled for a very long time.

Not always well handled. A significant number of 450Ds (including mine) need
to be returned to Canon for calibration before their PD AF works within
Canon's specification. I don't know if the defect rate is very different for
that model compared to others, or it's just that the 450D has become
renowned for it.

Some of Sony's cheaper DSLRs also seem often to show AF calibration
problems when used with very critical lenses with very shallow depths
of field. It could be argued that Sony expect people buying such
critical and expensive lenses to be using them with more expensive
camera bodies. The kind of small AF errors which show up with very
critical lenses at their extremes won't be noticeable with kit zoom
lenses. The depth of field will simply always swallow them.

It wouldn't be surprising if other camera makers spend a little less
care in calibrating the AF in their less expensive models for the same
reasons. Fortunately on the Sony alpha models the trimming screws
which adjust the distance and orientation of the AF sensor plane are
externally accessible and can be trimmed by owners with the technical
skills. Some of those who have done so claim to be able with care to
trim to greater AF precision than you often get by sending the
camera back to the maker for calibration.

That's why "have you got accessible AF trimming screws?" is one of
the questions I ask of a new DSLR which wants me to buy it :-)

--
Chris Malcolm

"Chris Malcolm" wrote:

That's why "have you got accessible AF trimming screws?" is one of
the questions I ask of a new DSLR which wants me to buy it :-)

Even better is adjustment on a per-lens basis...

--
David J. Littleboy
Tokyo, Japan

Chris Malcolm wrote:
Wilba wrote:
Miles Bader wrote:
Wilba writes:

The accuracy of the PD system is directly dependent on careful
calibration of all components, whereas the accuracy of the CD
system is not.

Hmm, it's certainly dependent on the alternate optical path (to the PD
sensors) having the same length as the actual image path, but that's
true of all traditional focusing systems (SLR focusing screens,
rangefinder optics, etc), so it's obviously an issue that's been well
handled for a very long time.

Not always well handled. A significant number of 450Ds (including mine)
need to be returned to Canon for calibration before their PD AF works
within Canon's specification. I don't know if the defect rate is very
different for that model compared to others, or it's just that the 450D
has
become renowned for it.

Some of Sony's cheaper DSLRs also seem often to show AF calibration
problems when used with very critical lenses with very shallow depths
of field. It could be argued that Sony expect people buying such
critical and expensive lenses to be using them with more expensive
camera bodies. The kind of small AF errors which show up with very
critical lenses at their extremes won't be noticeable with kit zoom
lenses. The depth of field will simply always swallow them.

It wouldn't be surprising if other camera makers spend a little less
care in calibrating the AF in their less expensive models for the same
reasons. Fortunately on the Sony alpha models the trimming screws
which adjust the distance and orientation of the AF sensor plane are
externally accessible and can be trimmed by owners with the technical
skills. Some of those who have done so claim to be able with care to
trim to greater AF precision than you often get by sending the
camera back to the maker for calibration.

That's why "have you got accessible AF trimming screws?" is one of
the questions I ask of a new DSLR which wants me to buy it :-)

Yeah, mine is, " do you have focus micro-adjustment?" :- )

David J. Littleboy wrote:
"Chris Malcolm" wrote:
That's why "have you got accessible AF trimming screws?" is one of
the questions I ask of a new DSLR which wants me to buy it :-)

Even better is adjustment on a per-lens basis...

One to make sure the image to focus screen/AF is correct. Lens
correction for fine tuning (Sony a900 has this, but goes by FL. So if
you have two different lenses of the same FL, it can't discriminate. No
real issue for most shooters, but I could see having a 135 f/1.8 and a
135 f/2.8 [T4.5] STF in my kit. OTOH the STF lens is MF only, so
shouldn't matter).

In article , John Navas
wrote:

That's all speculation, and it doesn't take into consideration such
critical factors as the effectiveness of the focusing algorithms, the
data they have on the systems*, the speed and power of the focusing
motors, the masses they need to move, and the distance they need to move
it. As we say in computers, GIGO.

* An advantage of a non-interchangeable lens camera is that the focusing
system can know exactly how the entire focusing system behaves. An
interchangeable lens system (SLR) inevitably has _less_ information
available.

that's wrong. interchangeable lenses contain a rom chip that holds all
of the parameters of the given lens so that the focusing system knows
how to properly control it.

in addition to the obvious data such as focal length and f/stop are the
ballistics of the motor and/or gearing, so the camera can accurately
determine how much to turn the motor.

either system can have all the necessary information. there is no
advantage to a fixed lens camera.

but it's very
surprising if it works better than the phase-based system used in DSLRs.

Both contrast and phase based systems can be made to work very well in
practice, and it's silly and pointless to make sweeping claims about one
being "better" than the other.

although both can work well, real world cameras that exist today show
that the best phase detect systems outperform the best contrast detect
systems. maybe one day that will change, but *today* that's how it is.