Autofocus is often taken for granted, but
how do you ensure that the system is paying ball? We put the spyderlenscal
reference system to the test
25 YEARS photographers have enjoyed the
benefits of autofocus systems in their SLRs. For spontaneous shooting or when
taking your time is not an option, autofocus is something that most of us could
not do without. But have you ever considered the accuracy of this system that
we rely so heavily on? Every camera is slightly different within manufacturing
tolerances, add to that the number of different lenses available and that no
two lenses of the same model are exactly the same and you have a lot of
variables that can affect the accuracy of your camera's autofocus system.
Spyderlenscal
Well constructed and simple to set up, the SpyderLensCal provides a clear
target and reference system to test and adjust your autofocus system.
The solution used to be that you sent your
camera and lenses off for calibration, often at considerable cost and the
inconvenience of being out of action whilst this was done. Fortunately, many
DSLRs now come with autofocus microadjustment options built into the camera's
menu system, allowing you to fine-tune your camera for optimal performance with
each one of your lenses, and therefore tailor its accuracy for the glass you
are using.
It's worth mentioning that we are looking
at microadjustments rather than large corrections; if your camera is
consistently selecting focus incorrectly by a sizeable margin then the
adjustments available in-camera will most likely not correct this and we would
take this as an indication that there is a more serious problem with your
camera. To a similar end, if you find that all of your lenses mis-focus by a
similar margin, it could indicate a focusing problem with your camera and it
might be time to treat it to a service.
It has always been relatively simple to
tell whether focus is accurate or not, but less so to tell whether the camera
is front or back focusing, and by how much. This is where the Datacolor
SpyderLensCal comes in. The SpyderLensCal offers a reference test system,
allowing you to check your camera's autofocus accuracy and then make
microadjustments accordingly before retesting and comparing the results.
The SpyderLensCal is a relatively
straightforward device and comes packaged with a brief quick-start guide. The
advice within instructs you on howto set up and use the test target, but offers
no advice on how to make the microadjustments. The presumption is that you are
either aware of how to find and adjust your camera's microadjustment function,
or that you study the camera's manual to find out. The device comes fitted with
a tripod mount and built-in bubble level so that you can set it up level for
accurate readings. It's worth noting that your camera also needs to be mounted
on a tripod for consistency between adjustments. Most people do not have two
tripods but placing the SpyderLensCal on a flat surface, on a level plane to
your camera, will also achieve accurate enough results. The quick-start guide
also offers no advice on the distance between the device and camera, only that
you should shoot in aperture-priority with the lens at maximum aperture.
Independent opinion is that you should set a shooting distance of 25x the focal
length, or set a working distance you are most likely to use the lens at.
The
Datacolor SpyderLensCal with Canon DSLR and lenses for scale.
Accuracy
Although not mentioned in the
SpyderLensCal guide, there are additional steps you can take to ensure the
test shots taken offer the best chance of making accurate microadjustments.
By using a remote release and enabling your camera's mirror lock-up mode, you
will reduce vibration during the exposure - important when analysing focal
accuracy at high magnification. Furthermore, I also chose to shoot using the
monochrome Picture Style offered in the camera's menu system. This may be an
over-precautionary measure, but in doing so you eliminate the chances of
chromatic aberration affecting your judgement of the test results. Finally, I
reviewed images on a computer monitor rather than the camera's LCD, as I
found the camera's review mode simply did not zoom in far enough to make an
accurate assessment.
