[bluebronco]

Hello,

How much is the K20Ds shake reduction on average?
I was reading that the Sigma 70-200 2.8 with OS up to 4.5 stops.
I was wondering if that was inline with the K20Ds in camera SR.

Thanks

[Adam]

There are a few threads already on the topic- I'd do a search and take a look at the similar threads at the bottom of this page.

On average both systems are the same. Basically, in-camera SR saves power while in-lens SR has in-viewfinder stabilization, which some people like. For most users, sticking to in-camera is fine.

[ChaseTheLight]

Four and a half stops might be reachable, but I doubt it in everyday situations. I am not a big lens shooter, so take my views with a grain or two of salt. I find in most situations with wide to short tele lenses, in-body stabilization gives me consistently two stops, sometimes more. In non-scientific tests, I got slightly better results with in-lens stabilization, not noticeable in most circumstances.

[Philoslothical]

That sounds about right. Lens stabilization isn't much (if any) better than the built in stabilization in Pentax DSLRs.

[RioRico]

With SR aka IBIS (in-body image stabilization) EVERY lens mounted is stabilized. With ILIS (in-lens IS) only specific (and usually longer) lenses are stabilized, so I'd be SOL (sh!t outa luck) with my glass accumulation. Also, claims of 4+ stops should be taken with several grams of salt. In real-world shooting, figure on 2 stops.

Here's the paranoid conservative ROT (rule of thumb) for shooting handheld: If detail doesn't matter much, shoot at any speed with SR on. If detail DOES matter, shoot at 1/FL with SR on, or 1/(5*FL) with SR off. And if detail REALLY matters, switch SR off and use a tripod. Go ahead, ask me why.

[hcc]

You already received many good advice. Let me add a relevant thread:
Image Stabilization Test: Olympus E-520 SLR Body - SLRgear.com!

Some test with an Oly 520 were done to test in-camera SR versus lens OS. The results were nearly identical :

" the performance of the Olympus E-520's IS system certainly demonstrated that there's no inherent reason why sensor-based IS systems should underperform lens-based ones";
"The bottom line on the Olympus E-520's IS system is that it turned in a superb performance, very much on par with the best lens-based IS systems we've looked at.".

As mentioned earlier, lenses with OS are more expensive than non-OS lenses. In turn a camera with SR gives you more choice of cheaper (and as good) lenses.

Hope that the comment may help.

[dms]

I find 200mm lens hand held give me acceptably sharp pictures (reliably) at 1/15s. While conventional wisdom would say 1/300 sec on a pentax dslr. I don't know what I could do w/o sr--my memory(?) says I used to be happy with 1/30-1/60 sec with 200mm on 35mm fim--so maybe it is about 1.5 to 2.5 stops for me. But not so at shorter focal lengths (again for me). At 28mm-50mm I don't find less than 1/8s to be reliably sharp with SR. And at 28mm-50mm I believe I could use 1/4-1/8s on film w/o sr. Not sure why--but seems likely the SR cannot be used for complex motion that may be present at (about) 1/4sec or longer!

[dms]

Vis-a-vis my reply above I shoot college theatre stuff--and my observation is based on several thousand pictures a year at the 200 mm focal length--which I review to pick the best. And generally camera motion is not the limiting problem--depth of field/missed focus and subject movement are. So for me the standard setting I use 1/15s with SR on (on a K20D) is "reliable." If the conditions were different (static and depth of field not a problem) then maybe 1/15s would not cut it. The question is very hard to answer--except to say qualitatively the K20D SR makes a very noticeable difference (for longer focal lengths)--for me. And that it works with every lens is amazing!!

[Anvh]

I find 200mm lens hand held give me acceptably sharp pictures (reliably) at 1/15s. While conventional wisdom would say 1/300 sec on a pentax dslr. I don't know what I could do w/o sr--my memory(?) says I used to be happy with 1/30-1/60 sec with 200mm on 35mm fim--so maybe it is about 1.5 to 2.5 stops for me. But not so at shorter focal lengths (again for me). At 28mm-50mm I don't find less than 1/8s to be reliably sharp with SR. And at 28mm-50mm I believe I could use 1/4-1/8s on film w/o sr. Not sure why--but seems likely the SR cannot be used for complex motion that may be present at (about) 1/4sec or longer!

It's the amount (length/range) of movement that's a problem, there is only so much room the sensor can move.
If you can steady things a bit, limiting the amount you move it can do very well.
I've done some 4 seconds shot around 18mm while leaning against something.

[asp1880]

This thread has already touched on the advantages held by sensor shift stabilization.
Lens stabilization has two advantages over sensor shift:
1. It also stabilizes the image seen by the AF sensor, giving potentially better AF-C performance.
2. It stabilizes the viewfinder image.

All points are debatable, of course. A few people get motion sickness with looking into a
stabilized viewfinder.

Regards,
--Anders.

[Stone G.]

I haven't got any experiences with ILS but for Pentax SR I would surely count on 2 stops and up to 3 stops if you have a steady hand. (On a good day perhaps a bit more).

And don't forget: SR also works with all those fine, vintage MF-lenses!!!

[Lowell Goudge]

I cant speak of the K20 but I have shot using 500mm with my K7 and K5 in the 1/40-1/60 and posted results from this. 1/40 with 500mm on a croped sensor is just over 4 stops below the rule of thumb. in reality regardless of OS or SR, what this means is that good technique is rewarded. for me, the other benefits of OS, are more important, a stable image presented to the AF system, and a stable image in the fiewfinder