I had promised to take a good look at your tests when I got back home
from my trip and had a better monitor and image viewing tools. But now
that I'm back and have taken a closer look at some of the images and the
test conditions I'm afraid I have to disavow my offer.
The first thing I noted was very significant exposure differences
between some of the images. I couldn't imagine how that could be unless
some of the lenses weren't actually closed to the specified aperture.
But then I noticed you were shooting in aperture priority mode. That's
a decision I just don't understand when you supposedly have a fixed
scene with fixed lighting. To me that demands a fixed, manually
controlled exposure. I suppose it might be part of your criteria to see
if a lens delivers proper exposure on your E-P1 in that mode but I don't
consider it a test of the imaging capability of the lens.
The next problem was resizing for the web. I had assumed that all of
the images would be available in full size or nearly so. I'm not at all
comfortable that a 450x600 image (where each pixel is an average of 45
of its neighbors) begins to tell the whole story. I suppose the small
images are OK for a gross comparison but if two images appear similar
then a larger and perhaps full size image might be required to break the
tie.
Finally, I had hoped to do side-by-side comparisons using comparison
viewing software but, with only one image per page, loading the images
to my own machine for comparison would be just too tedious.
I thought you had a very interesting idea and liked the test rig. But I
don't think I can reliably compare anything without a great deal of work
and, given the image size and exposure problems, I'm not sure whether my
judgments would be valid.
Chuck Norcutt
On 4/23/2011 9:09 AM, Wayne Harridge wrote:
> A couple of weeks ago I assembled a stack of "normal" lenses, adapters and
> an E-P1 to shoot a comparison to see how each lens rendered the scene I had
> set up. The lenses ranged in focal length from 45mm to 58mm and included a
> few zooms set at 50mm. This link:
> http://lrh.structuregraphs.com/test/lens_02_f22.htm shows the setup which I
> hoped would help to differentiate the rendering of the different lenses, in
> particular bokeh. Note the elements which normally show up bad bokeh:
> sticks, specular reflections (aluminium foil on a stick, candelabra against
> the wall), striped wallpaper. I used a 500W photoflood for lighting,
> tungsten WB on the E-P1, ISO 400, aperture priority mode. The images were
> only resized to 450x600 for the web with no other post processing. Note
> that the E-P1 doesn't maintain consistent exposure when the lens is stopped
> down.
>
> The comparison page is here:
>
> http://lrh.structuregraphs.com/test/50mm_comparison.htm
>
> Any comments welcome, in particular I'd like to know which lens you think
> has the best rendition of the subject.
>
> Also let me know if you find broken links or any other problems.
>
> ...Wayne
>
> Wayne Harridge
> http://lrh.structuregraphs.com/
>
>
--
_________________________________________________________________
Options: http://lists.thomasclausen.net/mailman/listinfo/olympus
Archives: http://lists.thomasclausen.net/mailman/private/olympus/
Themed Olympus Photo Exhibition: http://www.tope.nl/
|