Digital BW, The Print

> Zeiss lenses for Canon.

It's always good to see top manufacturers compete.  I don't think 
Zeiss has any offerings in the range I'm now targetting.  Zeiss is 
clearly a favorite of mine, however.  I'll keep an eye on their 
offerings.

For now and for mild wide angles, I'm inclined to use Leica-M as the 
standard against which to compare MTF curves -- in part because they 
exist.  I'm just staying away from the extreme wide angles until I 
see acceptable mild wide results.

>... Note that based on sensor pixel pitch the 5D2 is 
> probably diffraction limited from about F5.6 ...

> See Lloyd Chambers discussion of the 21 mpixel full 
> frame diffraction issues
> (based on 1Ds MK3, but still 100% relevant).
> http://diglloyd.com/diglloyd/free/Diffraction/example-1DsM3.html

As with most good lenses, the Leica lens in the above test was best 
in the center at f 5.6.  After that, diffraction becomes a limiting 
factor -- in the center with the 21 mp sensor as well as with most 
other systems I've used.  I'm impressed the sensor can see the 
difference between f5.6 and f8. On the other hand, I was amazed my 
little Rebel 8 mp can detect as much as it can. 

I think these things are all somewhat additive factors.  Diffraction 
limit is not a solid wall.  Diffraction losses increasingly and 
continuously lower the contrast of moderate resolution test targets 
as lens openings become smaller.  But that is just one factor.  
Similarly, I expect the resolution increases of the sensors will 
continue to improve the overall image quality, even when the lens is 
stopped down somewhat and is contributing quite a bit of the contrast 
losses of the overall system.

Then again, at the edge it's usually more that diffraction limit that 
is the problem with the lens (and sensor).  For me, it's just a 
question of whether I can squeeze enough out of the system to be, 
overall, worth using.


> ... we also have to deal with processing of the image ...
> There are many ways to improve or screw up an image.

That's definitely the case, and I've experienced both directions.  
Hopefully, my average is becoming more positive.

> Is any one aware of a good comprehensive test or series 
> of test that have been done to take the myth and marketing 
> out of it and get down to the brass tacks?

I don't know of any single test that gives me the bottom line I'm 
looking for.  So, I just explore each factor I'm aware of and do the 
best I can.  

With lens sharpness, in my own testing I use resolution test targets 
for some purposes, as well as a sharp, clear distant mountain range 
(or similar detail rich distant target) to look for de-centering of 
an element, which shows up as one side being sharper than the other.  
This is a manufacturing defect for which I do send lenses back to the 
sellers when I detect the problem.  It's sadly common, but if you buy 
from NY, state law, I believe, requires the sellers to give a 14 day 
return.  Check for decentering and return if its significant.

In terms of outside testing, the Photodo tests are about the best 
I've seen.  (I think the person may have been a Hasselblad employee 
who was able to use their equipment.)  I'm not sure how much I trust 
the Canon charts, but they're better than nothing. In these MTF 
tests, as I indicated before, the edge performane of the highest 
resolution target seems to be most telling.

European magazines and groups seem to do better testing that we get 
in the US magazines.  

I followed Modern Photo for years, but I could detect only a small 
correlation between what they said and what I saw in my own testing 
and shooting.   

The newer Popular Photo tests don't show center v. edge, which is a 
rather critical omission, in my view.  In one lens I was exploring, 
the Popular Photo test was amazing, but a person with both the 
supposedly great independent lens and the Canon L comparable one 
showed that there was a huge difference in off-center performance 
that was lost in the rather un-critical Pop Photo test.  I've seen 
more than one example of this.  (Have you noticed magazines are 
reluctant to be very critical of major advertisers?)

In addition to lens sharpness, the vibration of SLRs has been a real 
problem for me.  I was very impressed, however, with the little Canon 
Rebel.  Even with my light travel tripod, there are almost no losses 
with the 90 TS due to camera vibration -- even with the mirror 
flapping.  (I do wish they'd make an auto bracketing that left the 
mirror up between shots, however.)

Paul
www.PaulRoark.com