Hi--

>One puzzling thing I found in some of my tests is that I get a fair
>variation in sharpness if I replicate a test several times, even though
>I'm pretty careful checking the focus visually (and relying at times on
>the AF, or using the focus confirmation signal on my AF Nikon bodies when
>manually focusing). I'm beginning to wonder if part of the problem is
>lack of flatness in the film itself, either because the pressure plate
>isn't holding the film absolutely flat, or because of minor flaws in the
>film. The variations won't normally show up on projection or at normal
>enlargement sizes, but they do show up under 50X with a microscope. Have
>you experienced this? It seems to happen independently of which camera
>body I use, so I don't think it's a problem with the focusing. The
>variations I get are beyond the 10% error margin you need to allow anyway
>given the subjectivity often involved in "close calls" in examining the
>images of the test patterns, which have about a 10% increment built into
>them (in the case of my Edmund Scientific chart).

I find AF focus quite unreliable, and the AF indicators laughable in their
innacuracy. Even with careful manual focus with a good finder/screen and
an easy target, I find that 4 attempts will get me two good focuses, one less good, one a little sharper (on average). I don't think this is due to
variable film curvature, but to being able to see slight focus errors on film when viewed with sufficient magnification (makes testing, ah, interesting..., and is why I recommend familiar subjects with a bit
of depth, rather than flat test targets [when you miss the focus on
charts, it is hard to tell poor focus from unsharpness in the lens -
a subject with depth will almost always have something in focus to
show the error level]).

>Next time I see a used Nikkor 500/8 at my favourite local store I'll rent
>it for a weekend (the store generally lets me do this, given the amount of
>cash I've dropped their way over the years) and compare it with my 300/4 +
>TC and let you know what I find.

I think you will be surprised - the 500mm Nikkor is quite sharp, if you
can get it focused, and hold it still......

>Good idea about the diagonal line to double check the film plane really is
>flat to the chart-- I'll to that next time. I've been relying on bubble
>levels on the chart (I've laminated it on particle board) and on the
>tripod head (I use the Manfrotto camera leveller gizmo, though it adds a
>lot of heft to the tripod if you carry it any distance), but I'm a little
>nervous about whether everything is as square as I think it is, especially
>when I'm shooting long lenses wide open.

One method I use is to forget flatness and move the lens/camera to place
the same target subject in each corner (or edge) of the frame (shooting position constant, lens not refocused).

>Another phenomenon I noticed with push-pull zooms, especially the longer
>ones but even the 70-210 AF Nikkor pulled out to 210, is the damn things
>actually droop. I've seen this on every long zoom I've tested -- I assume
>(hope) the manufacturers correct for this in the lens design. It becomes
>especially noticeable with lenses that don't have a tripod collar. You
>can often actually see the droop if you stand to the side of the rig when >it's on the tripod; it also becomes evident when you've carefully centred
>the lens mount on the chart close up, then pull the tripod back on a dolly
>across a reasonably level floor, attach the lens, pull it out full length,
>and discover that the centre of the chart is no longer centred in the
>viewfinder but has shifted up (indicating the lens has drooped). At first
>I thought it was looseness in the tripod head or in the quick-release
>plate's attachment to the head, but careful examination of the lens with a
>straight-edge shows the thing is drooping, the axis curves. Scarey, and
>another reason why I prefer to use prime lenses rather than zoom lenses
>when I can afford to carry them around. (I've also found consistently
>better sharpness with primes over zooms, at the same focal length and
>aperature). (The "drooping" problem doesn't seem to occur with shorter >zooms like the 35-105).

The newer 35-105 is IF. You think zooms are bad - grab the front of most
AF lenses, and see how much tilt can be present with different lens orientations! Bleah! Yuck! I stick with MF whenever possible for
both zooms and primes! It is amazing that any AF lens (except the IF
ones) could check out well-aligned on film!

>The worst example I've seen is the Tamron 200-400 AF a
>friend brought to my workshop -- made me glad I spent the extra money and
>got the 300/4 instead (the 300 is a lot sharper, as well as being a stop
>faster).
>
>Ed Overstreet

The 300 is a good lens, though maybe not with converters, alas....
David Ruether