phil_armitage1

<p>I saw bears most days during a trip to the East side of <a href="http://www.philarmitage.net/glacier_hikes.html">Glacier National park</a> in August (the US one, in the Many Glacier and Two Medicine regions). The terrain there is quite favorable for seeing wildlife - open hillsides and avalanche chutes close to the road - but still they were mostly sightings good for binoculars rather than great photos. I did encounter a mother and cubs out hiking... purposefully descending a trail a large group of us were ascending, but that was sheer luck (good or bad, I'm not sure, but we gave way and they ambled by without incident). I've personally seen fewer bears in Banff / Jasper, but again, that may just be the luck of the draw.</p><div></div>

<p>I find it useful for black and white conversions of <strong>very</strong> low contrast scenes. For example the image here (of the Knik river valley in Alaska) was shot at 200mm through maybe 20 or 30 miles of hazy air. Linear conversion was a much better starting point for getting the look I wanted than a standard conversion + a lot of messing with curves. In color it can be interesting too, though not especially to my taste.</p><div></div>

<p>This is not completely up to date (a year ago, in Spring), but back then there were plenty in residence at the El Badi Palace. Not too far away and in quite scenic surroundings.</p>

<p>The manufacturing cost of sensors is most probably determined primarily by their physical size rather than the number of pixels. Intel, for example, migrate their processors to newer process technology as fast as possible - it's cheaper to put the same number of components onto a smaller piece of silicon. There are surely some incremental savings from smaller pixel count devices - a slower processor would suffice etc - but it may not be as much as you think.</p>

<p>That said, it may well be the case that Canon's manufacturing has improved over the years so that the fraction of the 5D MkII's retail cost that is the sensor is smaller than it was for the original 5D... leaving room for a cheaper model. From the fact that such a thing hasn't appeared one can assume that Canon don't (yet) think it would boost their overall profits.</p>

 

<p>The only part of the hike that's at all steep is getting down into the canyon from the rim right at the start. Here's a <a href="http://www.philarmitage.net/zion/zion107x.html">photo looking upstream</a> . It's not at all exposed or troublesome.<br>

Once you reach the creek the informal trail is overgrown and a bit of a scramble in places but not difficult. Once you reach the Subway the rock is pretty slick though.<br>

Remember you need a permit even for a day hike to the bottom.</p>

A classic trip of about that length would be the Paria Canyon, which is comparable to the Zion Narrows but with less water, at least in

Spring / early Summer (a consideration, since the water in the Narrows is cold enough to need a wetsuit in November). It's a beautiful trip

for both hiking and photography. Don't know first hand what the conditions are like in late November though...

Interesting to see the table there with the light collecting efficiency per unit area of

various sensors - on the Canon side there was a big jump between the 10D and 20D

but not much improvement (after allowing for the 1/3 stop difference in true vs

indicated ISO) since - claims about "improved microlenses" etc notwithstanding! Bears

out what is I think common wisdom on these forums that image quality improvements

since the 20D have been incremental at best, but still good to see it stated more

quantitatively.

I think it's true that images are perceived as `sharp' based on the resolution of the luminance component rather than the chroma channel, so I can imagine that some trade off of sensitivity vs color information might make sense. On the other hand, DSLR images at least look basically noise-free at low ISO, and at high ISO what you see most is chroma noise. Can't immediately see how starting with less color information would help there. But maybe this is more relevant to very small sensors anyway.

 

As far as I know, there's no public reference for the end-to-end efficiency of sensors (what fraction of incident photons get focused onto the photosites, how far away from being photon noise limited are the detectors?). So it's impossible to know whether substantial real improvement could be made in sensor design (at fixed sensor size), or whether the future is in this sort of trickery.

I visited the Beartooths for just a few days a couple of years back... near Red Lodge both

the Lake Fork and West Fork of Rock Creek are beautiful valleys. Quinnebaugh Meadows,

Lake Mary and (especially) Black Canyon lake have spectacular settings. Further along the

road (itself very scenic) there are innumerable lakes on the high plateau - the scenery

there is quite unusual though not as rugged.

<p>

Some <a href="http://www.philarmitage.net/beartooth.html">Beartooths pictures</a>.

There's not much danger of damaging the mount or having the tripod tip over without a

ring, but a less than sturdy setup will need to be locked down really tight to prevent the

lens creeping downward. I'm happier with a lighter tripod + ring.

Mine has coped with prolonged drizzle / damp conditions no problem. Subjected to

drenching spray from <a href="http://www.philarmitage.net/yosemite/

yosemite10.html">Yosemite Falls</a> there were glitches (eg LCD review failed a couple

of times), but it was OK when dry. YMMV, Phil

If you're willing to hike a few miles, add to the above sunrise suggestions <a href="http://

www.photo.net/photo/2713016">Chasm Lake</a> (and the waterfall

beneath it). Phil

CCDs that allow image stabilization "on chip" as it were already exist - try a search on

"orthogonal transfer CCD" or see this technical paper:

<p>

<a href="http://adsabs.harvard.edu/cgi-bin/nph-bib_query?

bibcode=1997PASP..109.1154T">The Orthogonal Transfer CCD</a>

<p>

I think this technology will be used for the imagers on the Pan-STARRS telescopes to be

built on Hawaii, so it's not just a theoretical possibility. However I don't know if there's any

technical limitation on the speed of compensation (100Hz is mentioned, which isn't very

much) that would prevent use in a regular camera.

To add to that, the exhibition is at the London Natural History

museum now. I was there at the weekend, and it was crowded but

well worth visiting.

 

Overall I thought that the judges' choices this year weren't

quite as interesting as last year. But there are still many

great images -- Paul Nicklen's polar bear at sunset sticks

in the mind...

From a technical perspective I think that this is quite impressive -

diffractive optics have rarely been used in broadband applications,

let alone consumer products (even rather expensive ones).

 

As others have said, common sense suggests that this lens won't be

marketed unless its flare characteristics are very good. Still,

it will be interesting to see if any differences with conventional

lenses are discernable, since in principle one might expect rather

different problems from this kind of optics.