[windhorse]

Is there any way to tell what effect if any the IS is having when it is on, apart from the IS confirmation hand signal icon in the viewfinder?

Also can having IS switched on have any (negative) effect on IQ when hand-holding at slow shutter speeds in low light?

I think I got better results using the in-body IS system of my former camera, a Minolta 5D, though maybe this was because the sensor was only 5mp?

Maybe I just need to drink less coffee?

[Adam]

Can you get decently sharp hand-helds at 1/10s or 1/15s (for standard focal lengths)? If so, then the SR is working just fine.

[Steve.Ledger]

Any shutter speed greater than the focal length doesn't need SR unless you are seriously shaky.
I tend to leave it disabled until I need it.

[npc]

>>Is there any way to tell what effect if any the IS is having when it is on, apart from the IS confirmation hand signal icon in the viewfinder?
Take one picture with IS on , then one with IS off under the same (dim) conditions - compare the pictures.

I thought SR in my K-5IIs is big deal until I got the sony A7R - no IS, roughly same pixel pitch, same lens - turns out I'm not getting any worse results on same speeds with it due to shake.
Got curious, did some tests - SR might save the picture in some corner cases but for the most part when I start getting blurry pictures on a7r I just get less blurry but still blurry pictures on k-5iis. Some people might still find them acceptalbe, some not.
I guess when it is time to use a flash or a tripod you just have to use them.
However, I don't think I've noticed any negative results from leaving SR on all these years (i used to have the original k-5 before k-5iis) - I only turn it off when on a tripod.

---

Last edited by npc; 04-26-2015 at 01:55 AM.

[Fauxton]

Thom Hogan analyzed Nikon VR back awhile ago. The principles would be roughly the same - the Pentax system would have a frequency above which it wouldn't add but it may detract. Anyway you might find the article interesting even though its lens stab rather than in body and Nikon's rather than Pentax.

I leave SR off unless I need it.

---------- Post added 04-26-15 at 06:35 AM ----------

Nikon VR explained

Forgot the link in the last post....

---------- Post added 04-26-15 at 06:39 AM ----------

Falk Lumo: Pentax shake reduction revisited

[Rupert]

I leave SR on at all times, I couldn't do much without it. Most of my shooting is with the Bigma 50-500 and most of that at 500mm. All my shots are handheld, and I have found that the optimum speed at 500mm is 1/200. It keeps the ISO reasonably low in most cases, and still delivers a sharp shot.




If you have steady hands...and a little luck...you can go with a slower speed, but your keepers will decrease.

I was first thrilled by SR with the K10D when I got this shot and then realized how slow it was shot.

1/20@500mm


Except for certain circumstances, I can't see why anyone would not leave it on...I couldn't count the times it has made all the difference, but it must be thousands...many thousands!

Regards! Love SR! Do not be an SR hater, Otis will hunt you down like a rabid dog!

[windhorse]

Can you get decently sharp hand-helds at 1/10s or 1/15s (for standard focal lengths)? If so, then the SR is working just fine.

It depends on what you mean by "decently" sharp doesn´t it? A problem is the only way to test this seems to compare images taken hand-held at different shutter speeds. My problem with this is when I am hand-holding there is quite a bit of variation from one shot to the next. I guess I just have to take a lot more shots of the same thing until a pattern starts to emerge that I can draw conclusions from.

I thought SR in my K-5IIs is big deal until I got the sony A7R - no IS, roughly same pixel pitch, same lens - turns out I'm not getting any worse results on same speeds with it due to shake.

Wow that is a big difference in cameras, even if the pixel pitch is roughly the same. Most of what I have read elsewhere suggests A7R is not a good candidate for hand-holding but maybe you have more steady hands than most? If not how do you explain that SR makes not much difference?

I was first thrilled by SR with the K10D when I got this shot and then realized how slow it was shot.

Correct me if I am wrong but there seems to be a big difference between relevance and working of SR with long lenses compared to short lenses. Also, one thing is to get a sharp shot once in a while, another is to be able to get consistently sharp images, right?

From what I had read here and elsewhere I had thought SR would give me consistently sharp (ie no camera shake) images at shutter speeds down to 1/15 sec with my 28mm lens (36mm equivalent on my k-5 but, like NPC says, am not so sure if that is correct. Maybe I am pixel peeping too much. But I figure even when the image is small like your beaut bird shot the difference that camera shake makes is still noticable, even if not so obvious.

Now I realise there is no magic wand, so I try to find out more about how it works.

---

Last edited by windhorse; 04-26-2015 at 06:55 AM.

[Homo_erectus]

I've found that for general shooting SR seems to actually introduce a small amount of blur that looks like slightly missed focus. So, I pretty much never use SR anymore. The only time I ever turn it on is when I am shooting at a very low shutter speed.

[normhead]

I just turn it on and leave it on. It's automatically turned off when I use the 2 second delay which is what I use when I want "really sharp." But I'm curious about the "SR ruins my pictures" thing. Anyone care to post some images? Like a test chart with SR on and SR off? Seems like that would be pretty simple.

[dakight]

I just turn it on and leave it on. It's automatically turned off when I use the 2 second delay which is what I use when I want "really sharp." But I'm curious about the "SR ruins my pictures" thing. Anyone care to post some images? Like a test chart with SR on and SR off? Seems like that would be pretty simple.

This pretty much hits the nail on the head. Heretofore all we've seen is subjective opinions and hearsay evidence. The only way to definitively answer the question is to test using scientific methods to minimize or eliminate bias. A set of images with measurable results should be taken using the same camera/lens combination but with SR on and again with it off. Ideally the shooter should not know which setting is used; otherwise bias could be introduced. An objective examination of the results should yield the answer. There's a reason that drugs are tested using a "double blind" protocol and the same sort of method could be adopted in this case to give us an answer that we can depend upon.

[windhorse]

This pretty much hits the nail on the head. Heretofore all we've seen is subjective opinions and hearsay evidence. The only way to definitively answer the question is to test using scientific methods to minimize or eliminate bias. A set of images with measurable results should be taken using the same camera/lens combination but with SR on and again with it off. Ideally the shooter should not know which setting is used; otherwise bias could be introduced. An objective examination of the results should yield the answer. There's a reason that drugs are tested using a "double blind" protocol and the same sort of method could be adopted in this case to give us an answer that we can depend upon.

In theory this makes sense but in reality no two hand held shots can be replicated the same due to the variation in hand shake and doing such a test on a tripod is a different kettle of fish

[normhead]

In theory this makes sense but in reality no two hand held shots can be replicated the same due to the variation in hand shake and doing such a test on a tripod is a different kettle of fish

I'm not sure why... if SR messes up IQ, it should mess it up on a tripod. Yesterday I had a whole 13 burst sequences in sharp focus using my DA*60-250 and 1.4 TC at 350mm and 1/25 of a second. So I am really sceptical. I'm going to say right up front here, without fear of contradition, these images taken without SR would be a wash. Probably 100% of them.

If you're arguing that SR doesn't help images hand held, ok, now I'm really sceptical. First of all, up to 1/1000 of a second, you can tell the difference between hand held and tripod mounted images. (You may be able to tell up to 1/8000 second too, I just haven't seen a test for that.) So, hand held for most images, you are already talking about a certain amount of blur....now you're going to try and differentiate between motion blur caused by camera movement and blur caused by SR?

I simply am not seeing the kind of rigour you'd need to make a statement like "SR blurs your images at high shutter speeds." What are you talking about? Your impressions based on what?

So what I see, is leaving SR off for dubious reasons, as opposed to leaving it on for obvious reasons, and probably missing shots because of it.

Has anyone an example of an image they think was ruined by in camera SR?

[windhorse]

I am getting a little camera shake at slow shutter speeds with SR turned on but not a lot and not all the time. It is about the same as I used to get with my Canon 5D that had no SR a couple of years ago. The only reason I can think of why is either I am too old to hold a camera still like I used to be able to do or I am drinking too much coffee. Unless of course the mechanical SR is not all it is cracked up to be and maybe the type of SR used in Pentax cameras does not perform the same as optical SR in other camera bodies ie it is better than nothing but not as good as other IS solutions. I don´t know that is why I am asking. Normhead you seem to say for you it makes a lot of difference, others say to them it does not make a lot of difference. So maybe this depends so much more on who is holding the camera than the mechanical bits inside it that there is no scientific way to answer my question apart from trial and error.

[Adam]

It depends on what you mean by "decently" sharp doesn´t it? A problem is the only way to test this seems to compare images taken hand-held at different shutter speeds. My problem with this is when I am hand-holding there is quite a bit of variation from one shot to the next. I guess I just have to take a lot more shots of the same thing until a pattern starts to emerge that I can draw conclusions from.

You have to let the SR system arm / reset itself. If you don't see the little hand symbol in the VF at the time the shot is captured, it will not be stabilized. This could be one potential cause of the inconsistency.

Good enough results at those shutter speeds, to me, are those that can be scaled to web resolution without appearing blurry.

[windhorse]

Good enough results at those shutter speeds, to me, are those that can be scaled to web resolution without appearing blurry.

Thanks Adam for the useful clarification. Excuse my ignorance but not sure what you mean by "scaled to web resolution"? When I open a K5 file in Photoshop I get an image size of 52.15 x 34.54cm @ 240 dpi. Is this at actual size similar to what you refer to? Or do you mean scaled to 72dpi that if I understand right used to be the maximum reolution a computer screen could handle?