Downsides to Shake Reduction

right, which is why you should turn it off. I've seen the results, and they're not nice. When I have some time I will replicate them.

SR (at least on a K10D) also works pretty well on horseback.

Bret

No real need as it's automatic. As Pete says, if you use a 2 or 12 second timer SR is automatically turned off.
Not only that if you use IR remote and/or mirror lockup, bracketing by timer or IR remote again SR is automatically turned off with the K-7.

have been doing some searching and it seems some people have had to return there k7's for faulty SR systems and have been pleased with the replacements.

Will have another look tomorrow and take a few comparison pictures. If i see no improvement i might be returnig the camera

If you use PhotoMe exif viewer to view your shots you can see which ones are stabilized and which are not. I have yet to see any difference in shots taken with SR on or off when the camera is mounted on a tripod, other than the data saying stabilized or not ready.

Well said, sir.

People that have known me in these parts for a while will remember that I have a bit of a hobby horse about this topic, so I trust they'll forgive me for jumping back on my favourite bandwagon

But really, the fuzzy logic is what goes on in people's heads, methinks. This 'explanation' - that so many people seem to be able to 'remember' but can never cite an actual reference for - is one without a shred of actual evidence for it.

The whole point of SR is that it measures the shake, using accelerometers in the body, and then decides how much to move the sensor to counter the shake.

Whether it then applies some algorithm based on the type of shake expected, I don't know, but despite looking quite carefully for empirical evidence of SR causing blur, I've found none (I wrote it all up here a long time ago, so it'll be searchable if you're interested).

I recall that MattMatic also did some tests, and established that unstabilised SR can be worse than no SR at all, but to the best of my knowledge, there's really no evidence that stabilised SR causes any shake at all.

The trouble is, there has never been an official explanation for Pentax's recommendation to turn SR off on the tripod, and thus, mixed with anecdotal evidence of people getting blurred pictures for unknown reasons, urban myth is born.


As I've said before, I concede that the recommendation in the manuals is suspicious, and of course I may be completely wrong. But in the absence of some careful repeatable experimentation that gives actual evidence, I shall continue to hold the view that stabilised SR does not cause shake.

I'd be extremely interested in both your method, and your results, Bret, if you care to repeat them.

Yep. My experience to now is that hand-held, SR works very very nicely up to around 1/250. With a flash it can cause issues.

Strangely enough, I just tested on a 190B at minimum height and see no difference between on or off.

Tomorrow I will have more chance to play, I'll test some more. I was sure I'd seen issues where there was blur for no apparent reason; 2s delay, no-one near the tripod and such like but still blur. Hmm.....

Bret

Does the SR system have some form of intelligence?
Will it for example only stabilize vertical movement when you are panning or will it get fooled by a panning motion.

Other manufacturers who have lens based SR systems doe support a panning motion. Nikon only stabilises vertical movement when you pan.

I believe that's an option, you can set the SR of the Nikon on several settings and one is panning I believe.

No such thing with Pentax though.

If you were panning correctly, there would be no acceleration, and no reaction from the SR. It cannot detect velocity (unless it has gyros).

I have a shot from horseback where it's trying to compensate for rotary motion, resulting in a very interesting psychedlic effect.

I've just had SR on for the "play" in the garage with my new D-Lites and it doesn't appear to make any difference. Well, well?!

Bret

Before this thread disappears into history, with my view that there's no evidence that SR causes shake still firmly in place, I just thought I'd ask...

... did anyone who claimed it does, carry out any repeatable tests after we talked here last?

no. I've tried a lot of stuff and SR has simply been on since. I see no reason to turn it off. I was obviously seeing something else, since now it's fine, even on a rocksolid surface and at 1/320 @ 200mm.

Bret

The SR is automatically switched off when using the 2s timer.

Dan

Dunno what to say... I had what I felt was a concrete reproducible example where SR seemed to cause shake in the image while on a tripod. (performed the test a few times in the field, but not very scientific)

Anyway, I just did tests from .5 sec through 15 seconds with SR on/off. No discernible difference in any of the images could be noted. I just did this with both the K10D and the K7.
With SR On, I tried both before the SR hand symbol was lit and after the symbol was lit. Still no difference regardless of SR being locked on.

Unless there is a bug or some other factor that might affect how SR behaves, I have to now agree that simply being on a tripod should not require SR to be turned off.

FYI, set up:
Page of black text on white paper taped to the wall.
Camera on a tripod
Corded remote.