SR effectiveness at 300mm plus

derekkite · 11-26-2011, 01:36 AM

Has anyone done any tests to determine how effective K-5 SR is when hand holding long lenses?

I used a canon super zoom for a while and it had the ability to shut off part of the sr when panning. I'm curious whether it makes sense to have SR on at all when using my 300mm lenses, especially when following moving wildlife.

My understanding of the operation of SR in the body is that the sensor is shifted to follow the moving image. It would be effective only as far as the sensor can move, and probably not be terribly effective at long focal lengths.

Gimbal · 11-26-2011, 02:57 AM

The limited movement of the sensor is only a problem when using live view or video, for photo the sensor only needs to move a few pixels during the exposure.

dosdan · 11-26-2011, 06:07 AM

Originally posted by derekkite

Has anyone done any tests to determine how effective K-5 SR is when hand holding long lenses?

Falconeye collated and analysed K-7 results and and got an average improvement of 1.7 stops with SR over all tested FLs. See section 2.5.3 here:

LumoLabs Article -- Understanding Image Sharpness

Specifically, at 2ms exposure time (1/500s) the reduction with the light blue 300mm curve is about 2 stops (4x).

The red 300mm curve goes down to approx 3ms (1/320s). Here the improvement again is about 2 stops (4x).

The K-5 should equal or better this.

Dan.

creampuff · 11-26-2011, 08:36 AM

I'm no expert in interpreting pseudo scientific verbiage about SR.
What I know is SR is needed even more when the focal length gets longer because any little handshake is magnified.
Shooting very fast moving small sunbirds with AF handheld, I definitely welcome every little bit of shake reduction.

K-5 DA 55-300mm uncropped

ChrisA · 11-26-2011, 09:18 AM

Originally posted by derekkite

My understanding of the operation of SR in the body is that the sensor is shifted to follow the moving image.

This isn't quite accurate - the sensor is shifted in response to accelerations experienced by the body (and hence by the SR accelerometers).

The SR mechanism isn't aware of the actual image at all.

derekkite · 11-26-2011, 09:33 AM

That makes more sense. In a smooth panning motion it would not detect or correct for the movement.

Thanks all. Getting to know the tool.

jamesm007 · 11-26-2011, 10:04 AM

Originally posted by derekkite

Has anyone done any tests to determine how effective K-5 SR is when hand holding long lenses?

I used a canon super zoom for a while and it had the ability to shut off part of the sr when panning. I'm curious whether it makes sense to have SR on at all when using my 300mm lenses, especially when following moving wildlife.

My understanding of the operation of SR in the body is that the sensor is shifted to follow the moving image. It would be effective only as far as the sensor can move, and probably not be terribly effective at long focal lengths.

Using a zoom lens that can go to 300mm on my K20D (FF= 450mm) the K20D will default to a minimum speed of 1/180. This means the engineers were confident most people could take a hand held pic at that speed with good results. Thats about a two stop advantage using 1.5X as the rule. Of course most people can shoot at lower speed than this at 300mm. Its just a camera default. Shake Reduction is very effective at long focal lengths, even moving, it seems!

This goes against everything I have read. I shoot BIF with SR on. I have no problem tracking and or getting good sharp results even during high body speed movements such as trying to track a gull up close flying around me, toward me, above me, to the point I am about to lose my balance. Shake Reduction has no negative effects in fact seems to help! I use 11 point AF-C mode with a K20D and DA55-300mm. I am swinging the camera back and forth, up/down. How is Shake Reduction working for me? I really would expect the system to be overwhelmed by my body movements. Then again even with SR turned off its not physically locked down but magnetically held perpendicular to the lens and must compensate (but not correct shake) to keep it square with the lens by moving the SR platter in all directions including diagonally even in the K20D. The difference is it does not compensate for shake in this mode, but only keeps the sensor lined up with the back of the lens. Pretty strong magnets and fast, no? It may be this fact of how powerful the system is that enables very good results using it even panning up/down all around with it turned on and compensating for shake. In some of my pics I see a level of detail that I really believe is the result of Shake Reduction smoothing my movements. Although some of these are not at 300mm I still get the same results at that focal lenght. Not only does Shake Reduction help at 300mm, SR at the very least does not hurt even when panning at 300mm.

**davidreilly3207** · 11-26-2011, 12:53 PM

SR should be turned off when panning as it will probably not achieve a lock anyway with a moving object. A lock is critical to the function of the SR.

**davidreilly3207** · 11-26-2011, 12:59 PM

From Wikipedia:
The required sensor movement (both speed and range) increase with the focal length of the lens being used, making sensor-shift technology less suited for very long telephoto lenses, especially when using slower shutter speeds because the available motion range of the sensor quickly becomss insufficient to cope with the increasing image displacement.

SteveM · 11-26-2011, 01:41 PM

I don't have anything technical to offer but I have found SR to help with panning. The highest focal length is 800mm, so I've set it to 800 when I have the 1200mm bolted on the camera. This issue came up in another thread and as was pointed out, if you set the camera to use a remote or the timer, SR will be automatically shut off.

derekkite · 11-26-2011, 01:52 PM

Originally posted by mdave13

SR should be turned off when panning as it will probably not achieve a lock anyway with a moving object. A lock is critical to the function of the SR.

Now I'm confused. A lock implies that the algorithm finds edges in the image and attempts by mechanical adjustments to maintain the position of the edge on the sensor for the duration of the exposure. I suspect that the Canon that I used did something similar. It was mirrorless, and had settings for IS during panning. If you didn't turn it off, the image would jump as you panned, as it attempted to maintain the position until it couldn't.

If it uses gyros, which seems to be the case, then movement on two axis' would be detected, yaw and pitch, and the sensor would be moved in response.

The question is whether there are circumstances where you would get image degradation from SR. Other than tripod use of course.

Gimbal · 11-26-2011, 02:26 PM

The gyros/accelerometers measures relative movement, so if you move/pan at a steady pace then it will work, just as it will work taking pictures while trawling in an airplane even though you are "moving" at a very high speed.

However you might fool the SR if right before shooting you change speed or aim as the camera will have no way of knowing if that sudden change was deliberate or if it where a shake that should be counteracted. If the change is small it will try to counteract it, chances are that it will get things wrong. So if you are shooting subjects that alter speed and course suddenly thus resulting in unpredictable panning, turn off SR.

**Lowell Goudge** · 11-26-2011, 03:02 PM

I think it is the same as the K7. So I have attached a shot with theb300/4 and 1.7x AFTc

Forgot to mention 1/40 th hand held

Rupert · 11-26-2011, 03:32 PM

I have tons of shots at 500mm and very low speeds that tell me SR is invaluable at long ranges. I think this one I have posted before is when I realized just how important SR could be. It was with the K10D, but I find the K5 to be slightly better.......

1/20 @ 500mm Hand held
[IMG]

[/IMG]

RED ALERT! Squirrel shot coming up for those with Squirrel allergies! I warned you!

It works for Squirrels too, but with the K5 you get the added benefit of speed with higher ISO values, so it is double good!
Corn is double good too! Sorry about the nudity...her bras were a little uncomfortable, so she took them off!

1/200 f6.7 500mm ISO 1600
[IMG]

[/IMG]

Best Regards!

creampuff · 11-27-2011, 07:06 AM

Originally posted by mdave13

SR should be turned off when panning as it will probably not achieve a lock anyway with a moving object. A lock is critical to the function of the SR.

I think it would help if you can come up with a photo taken with a Pentax (preferably the K-5 to answer the OP's post) to prove this instead of just stating Wikipedia or making assertions without actual proof.

11-26-2011, 01:36 AM	#1
derekkite Pentaxian Join Date: Sep 2011 Location: Nelson B.C. Posts: 3,782	SR effectiveness at 300mm plus Has anyone done any tests to determine how effective K-5 SR is when hand holding long lenses? I used a canon super zoom for a while and it had the ability to shut off part of the sr when panning. I'm curious whether it makes sense to have SR on at all when using my 300mm lenses, especially when following moving wildlife. My understanding of the operation of SR in the body is that the sensor is shifted to follow the moving image. It would be effective only as far as the sensor can move, and probably not be terribly effective at long focal lengths.

11-26-2011, 03:02 PM	#13
Lowell Goudge Site Supporter Join Date: Jan 2007 Location: Toronto Photos: Gallery \| Albums Posts: 17,891	I think it is the same as the K7. So I have attached a shot with theb300/4 and 1.7x AFTc Forgot to mention 1/40 th hand held Last edited by Lowell Goudge; 11-26-2011 at 05:50 PM.

11-26-2011, 03:32 PM	#14
Rupert Veteran Member Join Date: Sep 2006 Location: Texas Photos: Gallery \| Albums Posts: 25,123	I have tons of shots at 500mm and very low speeds that tell me SR is invaluable at long ranges. I think this one I have posted before is when I realized just how important SR could be. It was with the K10D, but I find the K5 to be slightly better....... 1/20 @ 500mm Hand held [IMG] [/IMG] RED ALERT! Squirrel shot coming up for those with Squirrel allergies! I warned you! It works for Squirrels too, but with the K5 you get the added benefit of speed with higher ISO values, so it is double good! Corn is double good too! Sorry about the nudity...her bras were a little uncomfortable, so she took them off! 1/200 f6.7 500mm ISO 1600 [IMG] [/IMG] Best Regards! Last edited by Rupert; 11-26-2011 at 03:40 PM.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Cost Effectiveness of Printing Your Own Photos	reivax	Digital Processing, Software, and Printing	4	11-16-2011 03:01 AM
Poll: SR effectiveness	bwDraco	Pentax DSLR Discussion	17	12-05-2010 10:34 AM
Effectiveness of SR in K-5 vs. K-7	conradj	Pentax K-5 & K-5 II	12	11-27-2010 06:07 AM
DA 55-300mm shoots some pretty decent pseudo-macro shots at 300mm	G-Diesel	Pentax SLR Lens Discussion	14	10-19-2010 07:47 PM
Weather Sealing Effectiveness	smcook99	Pentax SLR Lens Discussion	34	04-14-2010 03:32 PM

11-26-2011, 02:57 AM	#2
Gimbal Pentaxian Join Date: Apr 2007 Location: Sweden Posts: 2,106	The limited movement of the sensor is only a problem when using live view or video, for photo the sensor only needs to move a few pixels during the exposure.

11-26-2011, 06:07 AM	#3
dosdan Pentaxian Join Date: Dec 2007 Location: Brisbane, Australia Photos: Gallery \| Albums Posts: 1,741	Originally posted by derekkite Has anyone done any tests to determine how effective K-5 SR is when hand holding long lenses? Falconeye collated and analysed K-7 results and and got an average improvement of 1.7 stops with SR over all tested FLs. See section 2.5.3 here: LumoLabs Article -- Understanding Image Sharpness Specifically, at 2ms exposure time (1/500s) the reduction with the light blue 300mm curve is about 2 stops (4x). The red 300mm curve goes down to approx 3ms (1/320s). Here the improvement again is about 2 stops (4x). The K-5 should equal or better this. Dan.

11-26-2011, 08:36 AM	#4
creampuff Veteran Member Join Date: May 2007 Location: Singapore Photos: Gallery Posts: 3,953	I'm no expert in interpreting pseudo scientific verbiage about SR. What I know is SR is needed even more when the focal length gets longer because any little handshake is magnified. Shooting very fast moving small sunbirds with AF handheld, I definitely welcome every little bit of shake reduction. K-5 DA 55-300mm uncropped

11-26-2011, 09:18 AM	#5
ChrisA Veteran Member Join Date: Feb 2007 Location: London Posts: 393	Originally posted by derekkite My understanding of the operation of SR in the body is that the sensor is shifted to follow the moving image. This isn't quite accurate - the sensor is shifted in response to accelerations experienced by the body (and hence by the SR accelerometers). The SR mechanism isn't aware of the actual image at all.

11-26-2011, 09:33 AM	#6
derekkite Pentaxian Join Date: Sep 2011 Location: Nelson B.C. Posts: 3,782 Original Poster	That makes more sense. In a smooth panning motion it would not detect or correct for the movement. Thanks all. Getting to know the tool.

11-26-2011, 10:04 AM	#7
jamesm007 Veteran Member Join Date: Jul 2008 Location: Detroit MI, USA Posts: 508	Originally posted by derekkite Has anyone done any tests to determine how effective K-5 SR is when hand holding long lenses? I used a canon super zoom for a while and it had the ability to shut off part of the sr when panning. I'm curious whether it makes sense to have SR on at all when using my 300mm lenses, especially when following moving wildlife. My understanding of the operation of SR in the body is that the sensor is shifted to follow the moving image. It would be effective only as far as the sensor can move, and probably not be terribly effective at long focal lengths. Using a zoom lens that can go to 300mm on my K20D (FF= 450mm) the K20D will default to a minimum speed of 1/180. This means the engineers were confident most people could take a hand held pic at that speed with good results. Thats about a two stop advantage using 1.5X as the rule. Of course most people can shoot at lower speed than this at 300mm. Its just a camera default. Shake Reduction is very effective at long focal lengths, even moving, it seems! This goes against everything I have read. I shoot BIF with SR on. I have no problem tracking and or getting good sharp results even during high body speed movements such as trying to track a gull up close flying around me, toward me, above me, to the point I am about to lose my balance. Shake Reduction has no negative effects in fact seems to help! I use 11 point AF-C mode with a K20D and DA55-300mm. I am swinging the camera back and forth, up/down. How is Shake Reduction working for me? I really would expect the system to be overwhelmed by my body movements. Then again even with SR turned off its not physically locked down but magnetically held perpendicular to the lens and must compensate (but not correct shake) to keep it square with the lens by moving the SR platter in all directions including diagonally even in the K20D. The difference is it does not compensate for shake in this mode, but only keeps the sensor lined up with the back of the lens. Pretty strong magnets and fast, no? It may be this fact of how powerful the system is that enables very good results using it even panning up/down all around with it turned on and compensating for shake. In some of my pics I see a level of detail that I really believe is the result of Shake Reduction smoothing my movements. Although some of these are not at 300mm I still get the same results at that focal lenght. Not only does Shake Reduction help at 300mm, SR at the very least does not hurt even when panning at 300mm.

11-26-2011, 12:53 PM	#8
davidreilly3207 Loyal Site Supporter Join Date: Sep 2009 Location: South Jersey Photos: Gallery \| Albums Posts: 720	SR should be turned off when panning as it will probably not achieve a lock anyway with a moving object. A lock is critical to the function of the SR.

11-26-2011, 12:59 PM	#9
davidreilly3207 Loyal Site Supporter Join Date: Sep 2009 Location: South Jersey Photos: Gallery \| Albums Posts: 720	From Wikipedia: The required sensor movement (both speed and range) increase with the focal length of the lens being used, making sensor-shift technology less suited for very long telephoto lenses, especially when using slower shutter speeds because the available motion range of the sensor quickly becomss insufficient to cope with the increasing image displacement.

11-26-2011, 01:41 PM	#10
SteveM Veteran Member Join Date: Jan 2009 Location: Vancouver Island, BC, Canada Photos: Gallery Posts: 3,294	I don't have anything technical to offer but I have found SR to help with panning. The highest focal length is 800mm, so I've set it to 800 when I have the 1200mm bolted on the camera. This issue came up in another thread and as was pointed out, if you set the camera to use a remote or the timer, SR will be automatically shut off.

11-26-2011, 01:52 PM	#11
derekkite Pentaxian Join Date: Sep 2011 Location: Nelson B.C. Posts: 3,782 Original Poster	Originally posted by mdave13 SR should be turned off when panning as it will probably not achieve a lock anyway with a moving object. A lock is critical to the function of the SR. Now I'm confused. A lock implies that the algorithm finds edges in the image and attempts by mechanical adjustments to maintain the position of the edge on the sensor for the duration of the exposure. I suspect that the Canon that I used did something similar. It was mirrorless, and had settings for IS during panning. If you didn't turn it off, the image would jump as you panned, as it attempted to maintain the position until it couldn't. If it uses gyros, which seems to be the case, then movement on two axis' would be detected, yaw and pitch, and the sensor would be moved in response. The question is whether there are circumstances where you would get image degradation from SR. Other than tripod use of course.

11-26-2011, 02:26 PM	#12
Gimbal Pentaxian Join Date: Apr 2007 Location: Sweden Posts: 2,106	The gyros/accelerometers measures relative movement, so if you move/pan at a steady pace then it will work, just as it will work taking pictures while trawling in an airplane even though you are "moving" at a very high speed. However you might fool the SR if right before shooting you change speed or aim as the camera will have no way of knowing if that sudden change was deliberate or if it where a shake that should be counteracted. If the change is small it will try to counteract it, chances are that it will get things wrong. So if you are shooting subjects that alter speed and course suddenly thus resulting in unpredictable panning, turn off SR.

11-27-2011, 07:06 AM	#15
creampuff Veteran Member Join Date: May 2007 Location: Singapore Photos: Gallery Posts: 3,953	Originally posted by mdave13 SR should be turned off when panning as it will probably not achieve a lock anyway with a moving object. A lock is critical to the function of the SR. I think it would help if you can come up with a photo taken with a Pentax (preferably the K-5 to answer the OP's post) to prove this instead of just stating Wikipedia or making assertions without actual proof.