On Thu, Apr 30, 2009 at 10:35:15AM +0300, Margus Männik scripsit:
> Hi,

Hi!

> I don't know what EXIF-viewer do you use, but here is the screenshot  
> from Pentax Photo Browser:
> http://nagi.ee/photos/photo_sizes.php?id=10410167&size=o&album_id=148643

I'm using exiftool.  It has the interesting property of listing all of
the EXIF information, rather than the plausibly interesting or useful
exif information.  (Although I can see a use for camera temperature.)

So one can get both "Focal Length" and "Shake Reduction Focal Length".

So if I run exiftool on a downloaded copy of your lovely sample JPEG, I
get:

8:40 K-7 % exiftool -FocalLength -SRFocalLength 60-250.jpg
Focal Length                    : 250.0 mm
SR Focal Length                 : 196 mm

It's quite clear you had the 60-250 all the way out at 250mm; "Focal
Length" is 250mm.  The "SR Focal Length" isn't; it comes out at 196mm.

My immediate question was "that seems odd; does anyone know why it's
like that?".

If I chew through 2009's photos on my hard drive:

8:35 2009 % exiftool -T -LensID -FocalLength -SRFocalLength */*/*dng > list

(Get the Lens ID, the Focal Length, and the Shake Reduction Focal Length
from each of the dng files below the current (2009) directory, get this
information in Table format, and stuff that in a file called "list".)

Then sort the list, removing duplicates:

8:40 2009 % sort -u list > sorted_list

This gives me a list of Lens, Focal Length, Shake Reducation Focal
Length, separated by tabs.  Since tabs are doubtful creatures in email,
I've replaced all the tabs with 8 spaces and cleaned up the columns a
bit.

I find:

Lens                              Focal Length    SR Focal Length
smc PENTAX-DA 14mm F2.8 ED[IF]        14.0 mm        14 mm
smc PENTAX-FA 31mm F1.8AL Limited     31.0 mm        32 mm
smc PENTAX-FA 50mm F1.4               50.0 mm        52 mm
smc PENTAX-FA 77mm F1.8 Limited       77.0 mm        77 mm
smc PENTAX-FA MACRO 100mm F2.8       100.0 mm       100 mm

Which is what I'd expect; though the mirror lenses, where I have to
manually input the focal length, are interesting:

M-42 or No Lens        500.0 mm        500 mm
M-42 or No Lens        800.0 mm        804 mm

The 800mm Rokinon comes out at 804mm in SR, for no reason I can
plausibly suggest.

The DA 35 Ltd, on the other hand, was a surprise:

Lens                                      Focal Length    SR Focal Length
smc PENTAX-DA 35mm F2.8 Macro Limited        35.0 mm        34 mm
smc PENTAX-DA 35mm F2.8 Macro Limited        35.0 mm        35 mm
smc PENTAX-DA 35mm F2.8 Macro Limited        35.0 mm        37 mm
smc PENTAX-DA 35mm F2.8 Macro Limited        35.0 mm        41 mm
smc PENTAX-DA 35mm F2.8 Macro Limited        35.0 mm        46 mm

Shake reduction focal length appears to be interactive in some way
depending on *something*, I'd assume how far the front element has
moved.  (This lends credence to the view that it's really a small zoom.
:)

Rather than include all of the *many* focal length and SR focal length
combinations for the 55-300, I'll just note:

Lens                                  Focal Length    SR Focal Length
smc PENTAX-DA 55-300mm F4-5.8 ED        300.0 mm        252 mm
smc PENTAX-DA 55-300mm F4-5.8 ED        300.0 mm        260 mm
smc PENTAX-DA 55-300mm F4-5.8 ED        300.0 mm        268 mm
smc PENTAX-DA 55-300mm F4-5.8 ED        300.0 mm        276 mm
smc PENTAX-DA 55-300mm F4-5.8 ED        300.0 mm        284 mm
smc PENTAX-DA 55-300mm F4-5.8 ED        300.0 mm        292 mm

Shake reduction more or less has to be doing something based on a
detected property of the image, presumably the actual distance to the
subject (meaning the thing the camera has a focus lock on), rather than
purely the set focal length of the lens.

I was initially curious about what the different numbers implied about
the optical design of the 60-250; I'm now (having extracted all those
numbers) finding myself curious about how shake reduction actually
works.  This does provide an explanation for how improving shake
reduction (which is what Pentax said the 1.03 firmware for the K20D did)
could improve auto-focus performance, if the Shake Reduction has to
decide what SR Focal Length is *first*, and then autofocus has to try
for a lock.

I don't suppose you'd be willing to take a 60-250 shot focused on
infinity? :)

-- Graydon

--
PDML Pentax-Discuss Mail List
[email protected]
http://pdml.net/mailman/listinfo/pdml_pdml.net
to UNSUBSCRIBE from the PDML, please visit the link directly above and follow 
the directions.

Reply via email to