Lessons Learned
AirVenture Air Show
Oshkosh, WI
August 9, 2003
The recent AirVenture air show
http://www.airventure.org/
and convention at Osh Kosh, WI provided 5 days of demanding photographic
shooting conditions. These included:
Fast moving airplanes some at close distances
Bright sunlight outdoors to near dark conditions inside and at night
Lack of time to do an ideal set up for shots and
Unpredictable rain.
Equipment
AirVenture was a good news bad news story. The good
news was the rich selection of gear and many fantastic shots. The bad
news – carrying
4 camera bodies, between the two of us, was difficult. In the larger
context it was an opportunity to compare Fuji and Nikon lenses directly
against the Canon 10D with Canon lenses.
Camera Set Up
The following was used:
Continuous shooting
Auto WB
AutoFocus – AI Focus with Canon, with some
manual focus on Canon
Exposure was varied to match the lighting and image on the LCD display
Shooting modes: Mostly P with some Aperture preferred to get depth of
focus
Shooting
Approximately 7,000 pictures were taken and on one day over 2,600, a
record. Each evening the images were downloaded, reviewed and adjustments
made in the plan for the next day.
Observations
Snap
In shooting in South America and Africa some images
had a certain quality which I have come to label as “snap.” When one sees such
an image it just leaps out of the screen. This was first noticed when
shooting with the Nikkor 500mm lens which turned out more than any other.
In spite of the fact that images on the Toshiba 6100 notebook would be
at ½ the horizontal resolution one could easily recognize the
images with this quality. Then in ThumbsPlus, using the maximum resolution
mode, i.e., a screen pixel is a sensor pixel, the image quality was
obvious.
It might be said that these images are what professionals
call “technically
perfect” with all the key parameters right on: exposure, white
balance, focus and composition. Only time will tell if my criteria of
snap will meet the excellence criteria of a professional.
Good Images Per Total Images
This is our usual metric for pictures taken and we view that our command
of the camera and lens is good when this ratio is 80% and above.
We have gotten comments that this ratio would mean more if the criteria
for Good Images was much higher such as the ones printed. Yet, it is
my intent for this ratio to illustrate technical accomplishment of an
acceptable image rather than how it gets used.
Lessons Learned
Key Camera and Lens Comparison
In taking air show shots here are my rankings in a 10 point scale.
Fuji S2 500mm Nikkor 8
Fuji S2 80 – 400mm Nikkor 6
Canon 10D 100 – 400mm Canon 4
Canon 10D 400mm Canon 2
Assessment
The 500mm Nikkor images consistently had the snap quality.
Some are fantastic.
The lens is a total pain to carry and shoot with, yet,
the quality is consistent.
The experience at AirVenture confirms what was experienced before but
cemented in side by side shots. What is amazing about this lens is that
autofocus was not an issue in spite of not being a VR (vibration reduction)
lens.
The shots taken were about 50 – 50 on the Wimberley. With fixed
to slow moving planes the Wimberley mount worked well. The major drawback
is that I could not stand due to crowd view blockage – we were
on the first row the day I shot with the 500mm lens. Thus, the camera
and lens would rock about one or more of the tripod legs as I sought
the right vantage and composition. The rest of the time I would hand
hold the lens and camera. In this mode a “focus track” technique
would be used with the object. That is, find the aircraft in the
viewfinder, focus and follow it. This technique was found very effective
when shooting
birds in flight. The problems of shooting here is virtually identical.
The 80 – 400mm Nikkor zoom turned out excellent
results. It was good but did not have the consistent excellent image
quality of the 500mm.
Given the varied shooting conditions due to the focal
length range the VR capabilities appeared to make a difference.
Because this was a much easier lens to carry and its versatility it
was used as much as 10X the 500mm lens. This lens was used every day
while the 500mm lens was used only one day of the five.
The Canon 10D and 100 – 400mm lens (IS) combination
was nearly identical in features as the Fuji Nikkor set above, yet,
the performance
much less so.
A major limitation was that one would think that the image
was in focus when shooting but it was not so. There is no focus tracking
capability
in the same way as the Nikkor. This may in fact be happening but the
image quality was just not there. One does not have the positive lock
on capability. It might appear that the image was in focus but due
to vibration, image quality or focus the end result was much less than
the
Nikkor lens.
Good Images/Total Images – 60%
The Canon 10D and 400mm (non-IS) was a major disappointment.
Luminous Landscape has praised this lens as much superior
to the Canon 100 – 400mm Zoom. There are indications that this is the case.
A few of the shots were at or near the snap quality. Thus, under less
demanding conditions, such as nature or landscape photography, the
400mm lens could do very well. However, due to either motion or focus,
high quality pictures happened infrequently. This may be part due to
technique also. We only spent 2 days with the lens and did not try
and optimize its performance. For example, there may be techniques
to get it to focus more frequently. Given the focus difficulties we
did not spend time seeking to develop techniques for “focus track” described
above.
Good Images/Total Images – 30%
On the first day we attached a 2X teleconverter to make this a 800mm
lens. Given the low light we ended up doing manual focus. This did not
work. There was also not time or hands to track an object, focus and
shoot.
Firing Speed
The Canon 10D is much better in this regard than
the Fuji S2. However, these cameras are still not fast enough
for these conditions. There were
multiple stunts where aircraft would fly within 10’ of
the ground a clip a suspended banner. We just could not get
a picture
with this
happening. Part of this is technique and luck, yet, any hint
of shuttle lag made it much more difficult.
The buffer on the Canon 10D was much appreciated. There were many times
that pictures were taken continuously until the buffer filled.
White Balance
If there is a condition that would not create White Balance problems
this should be it. The Canon worked well here, compared to its poor performance
with varied color temperature illumination.
Exposure
Both the Fuji S2 and the Canon 10D shot hot – over exposed. However,
as we have seen before, the Canon 10D had the greatest variation in off
exposures – some to as high as 2 stops.
Battery Life
Canon rules here. We carried 10+ batteries and each camera used 2. Never
once did the camera run down the battery. We just swapped out each evening.
Storage Media
In spite of carrying 5GB of storage media on our heavy shooting day,
Saturday, we ran out. The new 4GB Microdrives would have been very useful.
Ease of Len Switches being Flipped
We noted how easy it was for the switches on the Canon lenses to get
flipped. On more than one occasion the camera came out of the bag with
AF shifted to MF.
Canon 100mm Macro
Used briefly. Excellent. Should have used more to get
a better sense for the image quality.
Observations
Maximum Pixels on the Object
Rule – Make Every Pixel Count
One of the strong points of digital imaging is the ease with which images
can be rotated and cropped, with the latter being the most important.
Often a poorly positioned image can be made to look very different with
a crop. Yet, this can create a false assurance. The major loss is resolution.
If there is the slightest lowering of image quality from the snap criteria
above the image can significantly degrade based on the pixel density
of the crop.
There are only two means to address this:
1) More sensor resolution (assuming it is adequately
captured by the lens),
This is why the 12mp sensors are so important – they match what
should be the native resolution of the best lenses. Nikon unfortunately
is going in the wrong direction with the DX lenses. They are making lenses
with poorer performance to match the size of the sensor. Given what we
have learned at 6mp and the experience of those with the Canon 1Ds Nikon’s
argument is vacuous.
2) Improve the composition to have the primary object
occupy more of the frame or all of it.
Here is where our rule above enters. However, as we
saw at AirVenture this is hard to do with moving aircraft – just
like birds in flight. One is forced to take what one gets.
Zoom lenses play a large role in this latter approach. Note that another
trade comes into play. If the zoom lens lacks in quality the gain in
object coverage may be offset by the inherent lower quality of the lens.
In the case of the 500mm lens above one can only frame based on the projection
of the image area onto the object.
This is all a part of a continual compromise between:
Lens performance;
Shooting conditions including object movement;
Ability to cover the object when shooting with maximum resolution.
Look Forward
The length of time we typically own a camera is less
than 12 months. If there is a better solution, such as a Fuji S3 and
highly unlikely
full frame Nikon, we would leap at the opportunity. The decision to go
with Canon was an important one in that the 10D is a transient camera – here
today and gone tomorrow. The significant part of this decision was the
investment in lenses. As harsh as this report is we believe there is
nothing fundamentally wrong with the lenses but the autofocus system
Canon uses in the camera. As this report outlines, a more fair test of
this premise would be with the 400mm lens in less demanding conditions.
Thus, as many of our experiences as reported in these lessons learned,
is just that – learning. A photographer with more experience with
Canon could well do better.
Canon has come under harsh criticism in the on-line
forums for the 10D’s
focus. We cannot say if the AirVenture experience supports this. We do
know that the 10D focus performance is much less that one would expect
from the camera and lenses of this caliber.
If there is a near term bottom line it is:
Pick the lens which best matches the shooting conditions.
Rather obvious but not always easy to do. |
