[TwoUptons]

Does anyone know when distance information was added to the AF lenses?

I seem to recall being told by a Pentax rep in the early 90s (about the time Nikon was going on about 3D matrix metering) that Pentax had been doing focus distance in the lenses for years.

Does that mean Pentax F lenses have distance information or is it strictly a FA thing?

-Eric

According to Bojidar Dimitrov's excellent K-mount page, F lenses do distance but FA lenses add the MTF data for the lenses and power zoom possibilities. I hope, then, that F lenses get full five-axis SR...

[SeattleDucks]

According to Bojidar Dimitrov's excellent K-mount page, F lenses do distance but FA lenses add the MTF data for the lenses and power zoom possibilities. I hope, then, that F lenses get full five-axis SR...

Thanks for the info. Any idea what will happen with a recent design third party lens such as the Sigma ART 35/1.4? That's the lens with K-1 I'm craving for travel and hoping it is compatible with the full 5-axis SR.

Ross

[Adam]

Thanks for the info. Any idea what will happen with a recent design third party lens such as the Sigma ART 35/1.4? That's the lens with K-1 I'm craving for travel and hoping it is compatible with the full 5-axis SR.

Ross

I wouldn't expect there to be any compatibility issues for autofocus lenses, so the F series and third-party lenses should be fine.

[Fogel70]

Yes, I'm sure it seemed like a dumb question But in Pentax's official specs they stated SR was available with "compatible Pentax lenses" which made me wonder. Also, other manufacturers with a sensor stabilizer do not necessarily have full compatibility with all lenses - for example the Sony A-series with 5-axis is not available with every Sony lens - some of their lenses drop to 3-axis mode.

It will be great if the full effective 5-axis SR is available with virtually any lens.

Thanks,
Ross

I believe we will only get 3-axis of stabilization on older lenses, but on other hand those 2 extra axis of stabilization K-1 offer is only an advantage if you focus very close to the subject.
So it's only going to make a major benefit when shooting macro. Had K-1 offered mechanical stabilization during video the 5 axis stabilization would also be a benefit as the x y motion can be noticed during video.

I normal shooting condition there will most likely not be possible to notice any difference if the camera use 5- or 3-axis SR.

[i5_david]

Why do we think we will only have 3axis?! I haven't read this anywhere... Why won't we notice the difference between 5 and 3 when Ricoh is making a big deal out of it (5 stops...). The focal distance is important? Why? I think I am having a little oroblem thinking this through (and understanding). Thanks in advanced.

[Fogel70]

Why do we think we will only have 3axis?! I haven't read this anywhere... Why won't we notice the difference between 5 and 3 when Ricoh is making a big deal out of it (5 stops...). The focal distance is important? Why? I think I am having a little oroblem thinking this through (and understanding). Thanks in advanced.

The X/Y movement affect the scene as much independent of distance. If you capture a scene that is 10 m wide, and the X/Y camera shake is 1 mm, then the camera shake is less than a pixel (1 mm movement on a scene that is 10000 mm wide), so it will not affect the captured image. But if the scene you capture is 10 cm wide then the same X/Y camera shake of 1 mm will affect the scene 100 times more and will be visible (1 mm movement on a scene that is 100 mm wide).

So to know how much to move the sensor, the camera need to know the distance the lens is focused on and which focal length is used. (magnification of the shot).

(In real life the X/Y camera motion is probably much less than I described above)

Btw, the 5 stops correction is verified by using the CIPA standard test, which only apply for pitch and jaw motion (2-axis).
The number of axis compensated for does not necessarily say anything on how well the stabilization work (numbers of stops).

[fotomarc]

Older zoom lenses that don’t communicate their current focal length will not get the full SR effect on all focal lengths. As for the two “new” axis added to the SR function, it will need the distance to the subject in focus to work. How the camera get that distance is still unknown, so maybe that will only work with newer lenses.

Does that mean Pentax F lenses have distance information or is it strictly a FA thing?

Focus distance is part of the information available through the data pin of AF lenses (KAF/KAF2 mounts). I am not sure how granular this information is or whether it has changed with lens generation.

I know this is an old thread, but I came across a bit of information that clarified/confirmed some of the above discussion about which types of lenses are supported for 5-axis SR (and I didn't see it elsewhere on PF), so I thought I'd add it here.

I happened to look at Ricoh's K-1 FAQ, and noticed that it says [with emphasis added]:

Q. Are there any lenses that the Shake Reduction (SR) function cannot be used with?

A. No, there are not. The SR function on the K-1 can be used with all supported lenses. Input a focal length in the range from 8 mm to 2000 mm when using a lens which cannot obtain the focal length information to the camera.

*The lens which can obtain the focal length information such as F/FAJ/FA series lenses are corresponded to five-axis mechanism (horizontal shift, vertical shift, roll, pitch and yaw), and the other older lenses are corresponded to three –axis mechanism (roll, pitch and yaw).

So, since the two additional SR axes supported by the 5-axis system are most useful for handheld macro work, it seems that using a Pentax AF lens for such work would have some SR advantage over an older, non-AF lens.

[stevebrot]

I know this is an old thread, but I came across a bit of information that clarified/confirmed some of the above discussion about which types of lenses are supported for 5-axis SR (and I didn't see it elsewhere on PF), so I thought I'd add it here.

Thanks for the clarification. The same FAQ note is in place for the KP and would apply to 5-axis SR on that model as well. I am curious as to whether 3rd-party manual focus lenses having the data pin will allow 5-axis SR. I am also curious as to whether the EXIF provides clues as to whether 3-axis or 5-axis is being used.


Steve

[Fenwoodian]

.
SR on the K1 works GREAT!

I shot this photo today from my open car window, traveling 65mph across a bumpy bridge. I was using a manual focus 80mm f/1.4 lens on my K1 camera. I don't believe that I'd have gotten such nice detail if SR would have been off.

[IMG][/IMG]

[photoptimist]

Thanks for the clarification. The same FAQ note is in place for the KP and would apply to 5-axis SR on that model as well. I am curious as to whether 3rd-party manual focus lenses having the data pin will allow 5-axis SR. I am also curious as to whether the EXIF provides clues as to whether 3-axis or 5-axis is being used.


Steve

What's needed for effective use of 5-axis is data from the lens on the focusing distance and that distance seems to be stored in EXIF. If the focusing distance is large relative to the focal length, horizontal and vertical shifting of the camera body has almost no effect on the image, so there's no reason use all 5 axes. Thus, a lens at infinity focus only used the 3-axis system.

Preliminary peeking at some EXIF entries shows that K and M lenses (which don't have a data in or any way to communicate the focusing distance) are reported in EXIF as "distant view" which would mean the camera only used 3-axis.

Third party lenses might report distance. I once took about a dead Sigma zoom that did have the K-mount data pin. The lens contained a 4-bit linear encoder that would have reported 16 possible distance values between infinity and a fraction of a meter.

Thus, it seems 3rd party lenses might report distance and might enable the 5-axis SR. Probably the easiest test is to mount the lens, enable SR, take a shot at minimum focus distance and see if the EXIF reports it correctly.

[stevebrot]

Third party lenses might report distance. I once took about a dead Sigma zoom that did have the K-mount data pin. The lens contained a 4-bit linear encoder that would have reported 16 possible distance values between infinity and a fraction of a meter.

Thus, it seems 3rd party lenses might report distance and might enable the 5-axis SR. Probably the easiest test is to mount the lens, enable SR, take a shot at minimum focus distance and see if the EXIF reports it correctly.

There are five known locations related to subject distance in the EXIF, two in the standard EXIF and three in the Pentax makernotes.

EXIF

• SubjectDistance (64-bit floating point, values not defined/constrained)
• SubjectDistantRange (16-bit int, four values allowed 0 (unk), 1 (macro), 2 (close), 3 (distant))1

Makernotes (ExifTool tag designations)

• FocusMode (16-bit int, multiple values defined with tag apparently overloaded for both distance and configuration settings)1
• FocusPosition (16-bit int, values/encoding not documented by ExifTool)
• FocusRangeIndex (8-bit int, apparent bitwise mask)1

1 Stored in EXIF by my K-3

I would be curious as to whether the K-1 and KP use the FocusPosition or SubjectDistance tags and whether there is 3rd-party support.


Steve

[Kerrowdown]

I'm hoping SR works with the classic Pentax A and M series manual focus lenses.

Rest assured it works just fine as that's the only Pentax glass I own.