On 3/6/2013 2:08 PM, usher99@xxxxxxx wrote:
> ...
>> If you want to see the difference, open an ORF in RawTherapee, or any other
>> converter that ignores, or may be set to ignore, the lens data. Here's a
>> very clear example of how ACR (as well as Viewer2) uses the lens data to
>> correct linear
>> distortion and vignetting.
>> <http://www.moosemystic.net/Gallery/tech/Lenses/Misc/_1153848_lindistort.htm>
> Yes, less misinformed than old data--didn't take too many months for
> Adobe to use the metadata it seems.
I think it's been there long before the E-M5. The same lenses and correction
metadata have been around on the Pen series
> I now recall that you said that ACR corrects to about 80 % which is about the
> same as for Canyon lenses ...
I think you are misunderstanding what Adobe has done. They are simply applying
the data the Oly lenses provide, as a
part of the Raw conversion, and before any application of ACR or user profiles.
Any decisions about how fully to correct are left to the manufacturer of the
lens. This makes perfect sense to me, as
Adobe would be working with one example of the lens, whereas Oly knows the
range of sample variation. (BTW, I don't
recall saying anything about an 80% solution, although I do know that slight
under correction is better than slight over
correction.)
If you look at the old MF lens tests in Modern. They allow ±5% for focal
length; for example, a 'long' 18 mm may be
18.90, and a 'short' 21 mm may be 19.95. Like magic, a 3 mm difference becomes
1 mm, a horizontal AOV spread of 90.0°
vs. 81.2° becomes 97.7° vs. 94.6°. Hardly worth the trouble, and that's within
expected normal variation, so there will
be outliers beyond that.
It's easy to get carried away with theoretical calculations and ideas, based on
a nominal values, and ignoring normal
manufacturing tolerances. I have no idea and will never know, whether my 9-18
has average distortion for the model.
Based on the results on this one image, at one focal length, I might guess
slightly greater than average distortion?
Given these facts, I think Adobe has made exactly the correct call, relying on
manufacturer knowledge, and, as below,
providing the means to find tune corrections
> ... but no slider adjustments are available.
This is not correct, in two ways. First, the Manual correction tab has the
sliders available.
Second, If I select "Enable Lens Profile Corrections" on the Profile Tab of the
Lens Corrections menu, I get a profile
for this lens, made on an E-P2, which nicely over corrects to a convex horizon
and more vignetting correction. With this
profile active, the sliders are active, too. Pull "distortion" down to zero,
and it looks just the same with the enable
box checked or not. One could very easily set it quite low to correct for the
residual distortion of my particular lens
at 9 mm.
My assumption is that this profile was done back before ACR incorporated the
metadata corrections. The E-P2 was
introduced about 3 1/2 years ago.
Is it possible that you have done a little too much on line research relying on
folks who write with apparent authority
without adequate knowledge and experience? ;-)
> Last I checked the LCA correction metadata was only valid for Panny lenses on
> Panny bodies. Perhaps that is still true.
That is what I have read. I've read that the Oly lenses don't provide LCA data
to the camera. Given my caveat above,
I'll not claim that myself. If true, it could, one supposes, be added in a
firmware update if a future Oly body that
can use it is produced. It seems that Oly bodies don't use LCA data from Panny
lenses. Whether they record it in the
metadata, I have no idea.
I can, if you want, provide the ORF for this image, and another one from the
Panny 20/1.7, if you want to do your own
research. :-)
> I still see some use the supplemental profiles to
> correct the residual geometric distortion or still use PT lens. I see
> that the horizon you corrected with the metadata still has about 10-20%
> residual barrel distortion left as I analyze with my trusty straight
> edge.
Yup, but, as I discussed above, that may not be generalizable.
> Enjoyed the roll-over.
A picture is worth ...
Thousand Words Moose
--
What if the Hokey Pokey *IS* what it's all about?
--
_________________________________________________________________
Options: http://lists.thomasclausen.net/mailman/listinfo/olympus
Archives: http://lists.thomasclausen.net/mailman/private/olympus/
Themed Olympus Photo Exhibition: http://www.tope.nl/
|