Follow along with the video below to see how to install our site as a web app on your home screen.
Note: This feature may not be available in some browsers.
Generally, this is true, but there are some (rare) exceptions: the Leica 70-180/2.8 APO zoom lens is rumored to be as good as any prime lens in that range; see:So if one wants the best possible quality, one must use primes.
--flawed.........this is why.
recent post about 10D being softer is in my opinion based on
downloaded chart pics and their included EXIF info show that 2
different lenses were used.
the D60 test pic was done with the 100mm lens set at f/6.3 and shot
at ISO100.
the 10D pic was done with a 28-70mmL set at f/5 and shot at ISO200.
here is the topic i'm refferrring too, and i'm posting the EXIF
data in another post to this one.
http://forums.dpreview.com/forums/read.asp?forum=1019&message=4510746
'He who works his land will have abundant food, but he who chases
fantasies will have his fill of poverty.' -- King Solomon
----flawed.........this is why.
recent post about 10D being softer is in my opinion based on
downloaded chart pics and their included EXIF info show that 2
different lenses were used.
the D60 test pic was done with the 100mm lens set at f/6.3 and shot
at ISO100.
the 10D pic was done with a 28-70mmL set at f/5 and shot at ISO200.
here is the topic i'm refferrring too, and i'm posting the EXIF
data in another post to this one.
http://forums.dpreview.com/forums/read.asp?forum=1019&message=4510746
'He who works his land will have abundant food, but he who chases
fantasies will have his fill of poverty.' -- King Solomon
Actually, that was with the 28-70mm. I changed it to read 1400 lines horizontally with the 105mm f/2.8.I just wonder about these numbers:
About 10D:
"I found "strong detail" out to at least 1,300 lines horizontally
and 1200 vertically, although there was still meaningful detail
beyond that point. "Extinction" of the target patterns didn't occur
until about 1,550 lines."
Well, yes, as much as they were in the first place, although I'm now a little more conservative in what I consider to be "extinction."About D60:
"It didn't start showing artifacts in the test patterns until
resolutions as high as 850 lines per picture height vertically and
horizontally. I found "strong detail" out to at least 1,200 lines,
and "extinction" of the target patterns occurred at about 1,700
lines."
Any comments? Are these results still valid?
Actually, that was with the 28-70mm. I changed it to read 1400I just wonder about these numbers:
About 10D:
"I found "strong detail" out to at least 1,300 lines horizontally
and 1200 vertically, although there was still meaningful detail
beyond that point. "Extinction" of the target patterns didn't occur
until about 1,550 lines."
lines horizontally with the 105mm f/2.8.
Well, yes, as much as they were in the first place, although I'mAbout D60:
"It didn't start showing artifacts in the test patterns until
resolutions as high as 850 lines per picture height vertically and
horizontally. I found "strong detail" out to at least 1,200 lines,
and "extinction" of the target patterns occurred at about 1,700
lines."
Any comments? Are these results still valid?
now a little more conservative in what I consider to be
"extinction."
test target though. There's an awful lot going on in a test target
- This points up the problem with picking numbers off the ISO-12233
like this, and a single number really does a very poor job of
representing the final result. I called the "strong detail" point
on the D60 at 1200 lines because that was the point at which it
showed some pretty blatant artifacts/aliasing. The 10D is much
cleaner at that point, not running into serious trouble until about
1400 lines horizontally. - But after its bad bobble at 1200 lines,
the D60 actually continued to show a fair bit of detail a goodly
ways beyond that point, really coming in pretty close to the 10D.
Actually, the resolution of the two cameras is pretty near to
identical, but the 10D looks to have a bit more sharpening added in
than did the D60, producing images that are crisper-looking
(slightly), even though I'm not sure there's much more detail
actually present.
Which really illustrates my point - All of us reviewing cameras
really need to post a huge disclaimer that these "lines of
resolution" numbers are only approximate indicators. - Anybody who
claims they can discriminate + - 50 lines of resolution is kidding
themselves. Or, they may be able to call one camera as50 lines
sharper than another all else being equal, but what do they propose
to do about cameras that apparently resolve more detail but with
horrible artifacts and aliasing along the way?
There's a much more informative set of data that can be extracted
from the ISO-12233 target, that shows "spatial frequency response,"
something akin to the frequency response curves you've probably
seen for audio gear, only dealing with spatial frequencies across
the image plane instead. The problem with these though, is that
you'll find all sorts of wild bumps and lumps in the curves, caused
by the camera mfrs' choice of sharpening algorithms, making it
impossible to reduce them to some universally meaningful number,
akin to the minus 3db point on an audio response curve. Given how
bizarre some of these curves can end up looking, I decided not to
promulgate them on my site, feeling that they'd only add to the
general confusion over resolution and sharpness. OTOH though, they
do show a lot of what a mfr is doing behind the scenes with their
sharpening algorithms. If I could find time, I'd perhaps reconsider
adding these into my reviews of high-end cameras. - But time's
definitely an issue.
I'll bow out of this discussion now though - I don't want to tax
Phil's good graces too much, and have at this point dealt with the
quality/accuracy issue with my 10D res target image. (Phil recently
told me that he didn't have a problem with my posting notes here to
respond to issues or flames about my reviews, but I understand his
desire for me not to be any sort of an active part of the
discussion.)
