[jlaubza]

Hi, I have one of those KX's that blurs at 1/60s - the typical double image blur with SR on that so many have complained about.

I did a post over at DPReview on what I started to do to pin down conditions etc.

I cracked the cause eventually by putting the KX on a tripod and using it with SR on - one is advised not to use SR when mounted on a tripod but I was desperate.

The shots were all fine! Off the camera, I then rested my 300mm lens barrel on a bean bag and tried again - still all OK.

But handheld, not a chance - double image blurring.

Conclusion - the cause is the user - i.e. me!

Maybe your camera has a real problem but try it on a tripod with SR on and make sure the camera is really stable.

I also think the SR is effective for only 2 stops handheld - a 300mm lens (i.e. 450mm equivalent) is 2 stops advantaged at 1/120s and 3 stops at 1/60s, which for me at any rate, is a stop too far.

I'd love to read about what happens with your problem KX when you put it on a tripod with SR on. If this sorts out your problem, then great and if it doesn't, then my commiserations.

Cheers.

[Zebooka]

You are wrong with SR effectiveness.
It is just not designed for long telephoto lens. At wide angle it is more effective.

Also on my K10D SR won't turn on if cam detects no movement. You can check this by M mode with long shutter speed without a lens. Put your cam on a table. After shutter is opened, try to slightly tilt/move your cam - it won't adjust sensor position. But if you shake slowly you cam before shutter release, then after it is opened sensor will be floating freely.

[Eruditass]

You are wrong with SR effectiveness.
It is just not designed for long telephoto lens. At wide angle it is more effective.

Also on my K10D SR won't turn on if cam detects no movement. You can check this by M mode with long shutter speed without a lens. Put your cam on a table. After shutter is opened, try to slightly tilt/move your cam - it won't adjust sensor position. But if you shake slowly you cam before shutter release, then after it is opened sensor will be floating freely.

Says who? One german tester shows it being more effective at telephoto.

I think it all depends on your personal shake profile..

[jlaubza]

Hi Zebooka, thanks for your reply. To answer:
1) Of course SR works with all lens focal lengths (FLs) - I used the 300mm length as longer FLs show the problem better.

2) You have not commented on my observation that I get only a two stop advantage. If I use a 100mm FL, I dont get blurring at 1/60s but do get blurring at 1/15s. Do you get better than 2 stop advantage?

3) Although the sensor does not move if the camera does not move, it is activated and ready to move. Many users believe the problem is caused by the physical shock of the mirror hitting the padding inside the camera when the mirror flips out of the way.

If they are correct, then even if the camera is stabilised on a tripod or a bag, the sensor will be vibrated by the mirror and the image will show a blur.

Since this does not happen with my camera under the conditions I tested, I can rule out the 'mirror slap' theory as the cause of the problem.

4) I want you now to think for a minute about which part of the camera needs to be stabilised - is it the body of is it the lens that needs the stabilisation the most?

Obviously you must say it is the lens - let us assume that the front of the lens moves 0.1mm as we take the picture - that will introduce a fast image movement on the sensor - it is speed of the movement that matters. Assume now that the body moves 0.1mm as we take the picture but that the front of the lens was fixed. The speed of the image movement on the sensor will be less. So it is not the body movement that we are trying to stabilise, it is the leverage effect of the body movement on the front lens element that we are trying to stop.

When I stabilise my hands with a camera in them on a bean bag, I can still get blurred shots (from hand vibration magnified by leverage advantage to the front of the lens) but when i hold the camera and rest the lens itself on the beanbag, I get clear shots.

I hope this explanation will help you and other readers to understand this problem better.

[jlaubza]

Hi Eruditass, yes I think you are right, every person has their own shake profile.

A recent poll on this forum indicated that 34% of users who responded have a shake problem on the KX. This is not a high enough error rate to be built into a camera - the error rate should be higher if it is built in because it means a design fault - design faults usually result in all units failing.

Manufacturing/assembly errors usually result in lower error rates - up to 5% but with good QC, less than this.

Manufacturing tolerances resulting in elements in a unit not fitting properly or being misaligned are really very small these days - of course it happens but QC on expensive cameras is usually very good and this type of problem would not generate 34% problem cameras.

The logical conclusion is that most of the reported SR errors are user related - users are expecting too much from the SR system or they are not as steady as they think they are - obviously I was not as steady as I thought I was and i accept that the fault lies in my technique.

"Physician, know thyself" is a valuable first step to fixing a problem. I have learned this for myself.

[Lowell Goudge]

For those who wonder about SR and focal length

check the forums for posts by me using a 500 mm lens at 1/40. It is with a k7 not kx but SR effectivity is all about technique

edit note. see the photo, not KX but K7, at 1/40 with a 300F4 and 1.7x TC for 510mm effective focal lenght.
considering the "rule of thumb" should be 1/750th for this focal length on an ASP-C sensor, I would have to say SR does do something. This shot was free standing, with nothing to lean on, no rest for the camera, but one hand on the camera, the other at the extreme of the lens, legs apart and staggered front to back, and arms tight in at my side.

this is a 100% crop, i.e. pixel to pixel crop out of a larger image to show detail

there is no way, SR can claim all of the 4 stops here, part technique part SR.

[DanWeso]

Are there any tips for people who may have a high shake profile besides lugging a tri/monopod for every shot?

[jlaubza]

Wow Lowell, you must be the king of steady. I don't know whether I should have more or less water in my drink before I take the next shot but that is brilliant. Come to think of it, I'll have a drink anyway, a toast if you like to the king of steady. Then I'll go and try again but I suspect that I'm going to have to limit myself to higher speeds than what you used.

[Lowell Goudge]

Wow Lowell, you must be the king of steady. I don't know whether I should have more or less water in my drink before I take the next shot but that is brilliant. Come to think of it, I'll have a drink anyway, a toast if you like to the king of steady. Then I'll go and try again but I suspect that I'm going to have to limit myself to higher speeds than what you used.

Don't praise me, it was all the bird!

while SR might help with a steady camera, it did nothing to keep the bird perfectly still.

One thing people may wish to remember is that SR eliminates camera movement, but only shutter speed eliminates subject movement (or minimizes it)

Also note that if you are panning, SR does not work because it can't figure out totally the panning motion.

It also does not work for the third axis of motion, towards and away from the subject.

SR is good but it is not 100%, it simply can't be.

[xjx]

The SR problem that was debated a lot is probably not what you are observing. They have the problem at wide angle.

Trying to handhold 300mm at 1/60s of course can produce a large percentage of blurred image, regardless SR on or off. You get a real SR issue if the blur also happens at 18mm.

[reeftool]

The results of SR are going to vary with the user. The longer the lens, the harder it is going to be to hold it steady. I can get some good shots at 300mm handheld but not consistantly. I don't even think I would even try for a shot at 1/40 with a 500mm. Great job, Lowell! I have determined 1/500 as my minimum with the 500mm mirror and will use a monopod. With my Sigma 70-300, I can get 1/250 with the lens fully extended. I might get a few lucky shots to come out at slower speeds but it's usually just luck. I have had much better results with short lenses, having shots come out at 1/4 sec.

[Parallax]

Are there any tips for people who may have a high shake profile besides lugging a tri/monopod for every shot?

I find scotch or brandy works well.

[jlaubza]

XJX, it is quite difficult to see blur at 18mm unless it is really extreme so if people are getting blurred shots at 18mm at 1/60s then yes, it may well be a camera problem.

I'm not talking about fuzzy edges but about the double image blur that has been widely reported.

If any readers have blurred shots taken at 18mm with a speed ranging from 120 to 60, please post them, with exif data to prove that the SR system was ready or at least check the exif to ensure that SR was ready.

Thanks

[Marv Toler]

Are there any tips for people who may have a high shake profile besides lugging a tri/monopod for every shot?

Dan, I have had some success with shooting the camera using a technique similar to shooting a rifle. My right hand grips the camera grip and shutter. The left hand supports the camera body and lens from below. Inhale, then exhale partially, then squeeze the shutter release. At slow shutter speeds breath control can help a lot. After a while the technique will be a habit. Good luck.

[cfraz]

500 mm lens at 1/40

Lowell - do you give private lessons?