shake reduction

oatman911 · 03-04-2009, 09:47 AM

Question, why are you not supposed to have shake reduction on when the camera is on the tripod? Will it hurt my camera if I forget to turn it off?

Oren

Gooshin · 03-04-2009, 09:57 AM

it will hurt your image, not the camera.

although i still dont know why

.

Saaby · 03-04-2009, 10:42 AM

To understand why, you need to understand how SR works.

The CCD is suspended in air by magnetic forces. Then a computer controls electromagnetic fields to keep the CCD suspended, and slightly adjust the location of it's suspension to counter-act the camera shake.

This suspending of the CCD introduces "shake" of it's own into the image.

When you're hand-holding the camera, the shake your hands introduce to the camera is far greater than the shake introduced by the Shake-Reduction system, so overall you're better with the system on.

If you're on a tripod, however, shake should be effectively 0, so the shake introduced by having the CCD suspended is greater than the shake if the system is off (And the CCD is locked in place).

Trivia:

A similar phenomenon can be seen in noise canceling headphones, if you've ever used them before. They use white-noise to cancel other noises. In a noisy environment the white-noise is preferable over other noises, but if you're listening in an already-quiet environment, the white-noise introduced is now additional noise, and you'd be better off served by turning the noise-canceling circuitry off.

Though we don't do video with our cameras, if you're doing video using a camera with OIS or SR, you need to turn that off when you're on sticks because as you start to pan or tilt, the camera will try to compensate, and then when you stop panning or tilting, the camera will compensate again, and your footage looks goofy for an instant at the beginning and end of your pan.

SOldBear · 03-04-2009, 12:13 PM

If you use the timer (either 2 sec or 10 sec delay), the camera will automatically turn off the vibration reduction for that exposure.

ChrisA · 03-04-2009, 12:55 PM

Originally posted by Saaby

If you're on a tripod, however, shake should be effectively 0, so the shake introduced by having the CCD suspended is greater than the shake if the system is off (And the CCD is locked in place).

The only way of taking a picture on the tripod, with SR on, is to release the shutter with your finger, in non-delayed mode.

All the delayed modes, and all cable release modes turn SR off whether you have the switch set to on or not.

The amount of shake introduced by your finger, during a manual shutter release, is substantial, and much more than this hypothetical electro-magnetism-induced non-zero shake.

I did a careful investigation into SR on a tripod, some time ago, and concluded that having SR on was no worse than having SR off, if you release the shutter with your finger.

What has also been shown quite convincingly, is that with SR on, taking the picture before you get the little 'hand' icon in the viewfinder, ie before SR has stabilised, gives results that are much worse that having SR off.

Gimbal · 03-04-2009, 02:57 PM

Originally posted by Saaby

To understand why, you need to understand how SR works.

The CCD is suspended in air by magnetic forces. Then a computer controls electromagnetic fields to keep the CCD suspended, and slightly adjust the location of it's suspension to counter-act the camera shake.

This suspending of the CCD introduces "shake" of it's own into the image.

When you're hand-holding the camera, the shake your hands introduce to the camera is far greater than the shake introduced by the Shake-Reduction system, so overall you're better with the system on.

If you're on a tripod, however, shake should be effectively 0, so the shake introduced by having the CCD suspended is greater than the shake if the system is off (And the CCD is locked in place).

Not entirely true, since it’s the same magnetic field that holds the sensor static when SR is off. The sensor is never floating lose in a magnetic field, it’s positioned firmly at positions computed by the SR-system. The actual reason for shutting SR off is still “under debate” as far as I know.

Gimbal · 03-04-2009, 03:02 PM

Originally posted by ChrisA

What has also been shown quite convincingly, is that with SR on, taking the picture before you get the little 'hand' icon in the viewfinder, ie before SR has stabilised, gives results that are much worse that having SR off.

Really? That is surprising, I thought it just wouldn't activate if you are to fast on the shutter. Any sources?

ChrisA · 03-04-2009, 04:02 PM

Originally posted by Gimbal

Really? That is surprising, I thought it just wouldn't activate if you are to fast on the shutter. Any sources?

This thread on the UK Pentax forum has quite a lot of discussion about this one - MattMatic is extremely highly regarded over there and did some quite convincing work on the topic. IIRC, it started with some observations that surprised him, and he then looked into it more closely and reported the results.

I wrote up my investigation into whether having SR on with the camera on the tripod made things better or worse, over there too.

Both threads are quite long, so beware

I certainly came away pretty convinced that the stories you hear about SR causing shake on the tripod are just urban myth, but that taking pictures with SR unstabilised is bad.

Quicksand · 03-04-2009, 05:31 PM

I haven't read the UK forum thread in detail, but I'll state my understanding of the situation:

Just think of Newtonian physics. You know, every action has an equal and opposite reaction.

When you hand-hold the camera, you hold it pretty tightly -- and you have a lot of mass applied to the camera. When the SR mechanism moves the sensor around, it has essentially no influence on your hands and arms because of the tremendous difference in mass. Also, your hands and arms are somewhat flexible, and would tend to damp any oscillation if there were any.

A tripod, especially a light one, CAN be moved around slightly by the SR mechanism. The SR moves the sensor, the mass of which also shakes the camera body (you can feel this during dust removal, for example), and that shake is transferred to the tripod. The tripod then moves, and the SR has to compensate for that too. Tripods are not flexible, and there's some possibility you'll get a feedback loop where the SR tries harder and harder to stop the tripod from moving, when in fact it's exacerbating it. There's not a HIGH likelihood of this occurring, but I believe it has been observed "in the wild."

That's an oversimplification, but you get the idea.

It shouldn't hurt the camera (the SR can only do so much, and it won't continue to increase past its design limit), but it can hurt your pictures!

Saaby · 03-04-2009, 10:44 PM

Me: 0
Forum: 1

Gimbal · 03-05-2009, 12:43 AM

Originally posted by Quicksand

A tripod, especially a light one, CAN be moved around slightly by the SR mechanism.

That’s what I think too, but there are always those who don’t agree. It’ also means that there is no problem with short focal lengths as the sensor movement is relatively small and smooth.

Gimbal · 03-05-2009, 12:44 AM

Originally posted by ChrisA

This thread on the UK Pentax forum has quite a lot of discussion about this one - MattMatic is extremely highly regarded over there and did some quite convincing work on the topic. IIRC, it started with some observations that surprised him, and he then looked into it more closely and reported the results.

I wrote up my investigation into whether having SR on with the camera on the tripod made things better or worse, over there too.

Thanks, will check it out.

**Douglas_of_Sweden** · 03-05-2009, 03:32 AM

Originally posted by Quicksand

I haven't read the UK forum thread in detail, but I'll state my understanding of the situation:

Just think of Newtonian physics. You know, every action has an equal and opposite reaction.

When you hand-hold the camera, you hold it pretty tightly -- and you have a lot of mass applied to the camera. When the SR mechanism moves the sensor around, it has essentially no influence on your hands and arms because of the tremendous difference in mass. Also, your hands and arms are somewhat flexible, and would tend to damp any oscillation if there were any.

A tripod, especially a light one, CAN be moved around slightly by the SR mechanism. The SR moves the sensor, the mass of which also shakes the camera body (you can feel this during dust removal, for example), and that shake is transferred to the tripod. The tripod then moves, and the SR has to compensate for that too. Tripods are not flexible, and there's some possibility you'll get a feedback loop where the SR tries harder and harder to stop the tripod from moving, when in fact it's exacerbating it. There's not a HIGH likelihood of this occurring, but I believe it has been observed "in the wild."

That's an oversimplification, but you get the idea.

It shouldn't hurt the camera (the SR can only do so much, and it won't continue to increase past its design limit), but it can hurt your pictures!

Unless it shake so much that the camera fall over on your ultra light tripod and bam crash the lens right on some rocky surface

Class A · 03-05-2009, 03:04 PM

Originally posted by Quicksand

A tripod, especially a light one, CAN be moved around slightly by the SR mechanism.

Yes, one night I wanted to make a 12 hour exposure of the night sky, put the camera on the tripod and released with the finger but forgot to turn off SR. Then next morning, the camera and tripod had walked away 50m or so!

Sorry, tried to do a Gooshin.

Timd · 03-06-2009, 01:38 PM

Remembering my control theory: you need to have only one dominant time period in any control system, and that needs to be the external shake.

On a tripod, the camera is far more rigid, in which case the natural frequency of movement is much greater than when hand held (your hands make very good shock absorbers). Thus if the sensor were to move, it might introduce vibration in the camera which might be at a similar frequency to the sensor movement. With these 2 movements at similar (high) frequencies, you can get positive feedback (classic case of 2 similar time periods in the control and the feedback loop, introducing instability).

I also think the bottom line is that you should not have any shake while on a tripod!

If you take a photo before the hand is showing, the SR has not managed to work out the movement of the camera, hence when the photo is taken, the sensor is not moving in time to the camera movement.

My gut feel is that the SR is only really effective in the 1/15th to 1/500th sec shutter speeds: basically below 1/15th sec the movement is too big to compensate and faster than 1/500th the movement is too fast to respond to.

Just my penny's worth...

03-04-2009, 09:47 AM	#1
oatman911 Forum Member Join Date: Dec 2008 Location: Bronx Posts: 70	shake reduction Question, why are you not supposed to have shake reduction on when the camera is on the tripod? Will it hurt my camera if I forget to turn it off? Oren

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Do you always have shake reduction ON?	gawan	Photographic Technique	18	08-01-2010 02:36 PM
Shake Reduction Help	TimShadler	Video Recording and Processing	10	11-30-2009 01:56 PM
Thank you Shake Reduction!	tcdk	Post Your Photos!	4	09-03-2008 09:29 AM
Shake reduction not on?	Iowa Tom	Pentax DSLR Discussion	9	08-11-2008 04:00 PM
shake reduction	jay legere	Pentax DSLR Discussion	7	01-20-2008 07:45 AM

03-04-2009, 09:57 AM	#2
Gooshin Veteran Member Join Date: Sep 2007 Location: Toronto, the one in Canada. Posts: 5,610	it will hurt your image, not the camera. although i still dont know why .

03-04-2009, 10:42 AM	#3
Saaby Forum Member Join Date: Dec 2008 Posts: 70	To understand why, you need to understand how SR works. The CCD is suspended in air by magnetic forces. Then a computer controls electromagnetic fields to keep the CCD suspended, and slightly adjust the location of it's suspension to counter-act the camera shake. This suspending of the CCD introduces "shake" of it's own into the image. When you're hand-holding the camera, the shake your hands introduce to the camera is far greater than the shake introduced by the Shake-Reduction system, so overall you're better with the system on. If you're on a tripod, however, shake should be effectively 0, so the shake introduced by having the CCD suspended is greater than the shake if the system is off (And the CCD is locked in place). Trivia: A similar phenomenon can be seen in noise canceling headphones, if you've ever used them before. They use white-noise to cancel other noises. In a noisy environment the white-noise is preferable over other noises, but if you're listening in an already-quiet environment, the white-noise introduced is now additional noise, and you'd be better off served by turning the noise-canceling circuitry off. Though we don't do video with our cameras, if you're doing video using a camera with OIS or SR, you need to turn that off when you're on sticks because as you start to pan or tilt, the camera will try to compensate, and then when you stop panning or tilting, the camera will compensate again, and your footage looks goofy for an instant at the beginning and end of your pan.

03-04-2009, 12:13 PM	#4
SOldBear Veteran Member Join Date: Feb 2009 Location: Alameda, CA Posts: 3,206	If you use the timer (either 2 sec or 10 sec delay), the camera will automatically turn off the vibration reduction for that exposure.

03-04-2009, 12:55 PM	#5
ChrisA Veteran Member Join Date: Feb 2007 Location: London Posts: 393	Originally posted by Saaby If you're on a tripod, however, shake should be effectively 0, so the shake introduced by having the CCD suspended is greater than the shake if the system is off (And the CCD is locked in place). The only way of taking a picture on the tripod, with SR on, is to release the shutter with your finger, in non-delayed mode. All the delayed modes, and all cable release modes turn SR off whether you have the switch set to on or not. The amount of shake introduced by your finger, during a manual shutter release, is substantial, and much more than this hypothetical electro-magnetism-induced non-zero shake. I did a careful investigation into SR on a tripod, some time ago, and concluded that having SR on was no worse than having SR off, if you release the shutter with your finger. What has also been shown quite convincingly, is that with SR on, taking the picture before you get the little 'hand' icon in the viewfinder, ie before SR has stabilised, gives results that are much worse that having SR off.

03-04-2009, 02:57 PM	#6
Gimbal Pentaxian Join Date: Apr 2007 Location: Sweden Posts: 2,106	Originally posted by Saaby To understand why, you need to understand how SR works. The CCD is suspended in air by magnetic forces. Then a computer controls electromagnetic fields to keep the CCD suspended, and slightly adjust the location of it's suspension to counter-act the camera shake. This suspending of the CCD introduces "shake" of it's own into the image. When you're hand-holding the camera, the shake your hands introduce to the camera is far greater than the shake introduced by the Shake-Reduction system, so overall you're better with the system on. If you're on a tripod, however, shake should be effectively 0, so the shake introduced by having the CCD suspended is greater than the shake if the system is off (And the CCD is locked in place). Not entirely true, since it’s the same magnetic field that holds the sensor static when SR is off. The sensor is never floating lose in a magnetic field, it’s positioned firmly at positions computed by the SR-system. The actual reason for shutting SR off is still “under debate” as far as I know.

03-04-2009, 03:02 PM	#7
Gimbal Pentaxian Join Date: Apr 2007 Location: Sweden Posts: 2,106	Originally posted by ChrisA What has also been shown quite convincingly, is that with SR on, taking the picture before you get the little 'hand' icon in the viewfinder, ie before SR has stabilised, gives results that are much worse that having SR off. Really? That is surprising, I thought it just wouldn't activate if you are to fast on the shutter. Any sources?

03-04-2009, 04:02 PM	#8
ChrisA Veteran Member Join Date: Feb 2007 Location: London Posts: 393	Originally posted by Gimbal Really? That is surprising, I thought it just wouldn't activate if you are to fast on the shutter. Any sources? This thread on the UK Pentax forum has quite a lot of discussion about this one - MattMatic is extremely highly regarded over there and did some quite convincing work on the topic. IIRC, it started with some observations that surprised him, and he then looked into it more closely and reported the results. I wrote up my investigation into whether having SR on with the camera on the tripod made things better or worse, over there too. Both threads are quite long, so beware I certainly came away pretty convinced that the stories you hear about SR causing shake on the tripod are just urban myth, but that taking pictures with SR unstabilised is bad.

03-04-2009, 05:31 PM	#9
Quicksand Pentaxian Join Date: Apr 2008 Location: Central Ohio (formerly SF Bay Area) Posts: 1,519	I haven't read the UK forum thread in detail, but I'll state my understanding of the situation: Just think of Newtonian physics. You know, every action has an equal and opposite reaction. When you hand-hold the camera, you hold it pretty tightly -- and you have a lot of mass applied to the camera. When the SR mechanism moves the sensor around, it has essentially no influence on your hands and arms because of the tremendous difference in mass. Also, your hands and arms are somewhat flexible, and would tend to damp any oscillation if there were any. A tripod, especially a light one, CAN be moved around slightly by the SR mechanism. The SR moves the sensor, the mass of which also shakes the camera body (you can feel this during dust removal, for example), and that shake is transferred to the tripod. The tripod then moves, and the SR has to compensate for that too. Tripods are not flexible, and there's some possibility you'll get a feedback loop where the SR tries harder and harder to stop the tripod from moving, when in fact it's exacerbating it. There's not a HIGH likelihood of this occurring, but I believe it has been observed "in the wild." That's an oversimplification, but you get the idea. It shouldn't hurt the camera (the SR can only do so much, and it won't continue to increase past its design limit), but it can hurt your pictures!

03-04-2009, 10:44 PM	#10
Saaby Forum Member Join Date: Dec 2008 Posts: 70	Me: 0 Forum: 1

03-05-2009, 12:43 AM	#11
Gimbal Pentaxian Join Date: Apr 2007 Location: Sweden Posts: 2,106	Originally posted by Quicksand A tripod, especially a light one, CAN be moved around slightly by the SR mechanism. That’s what I think too, but there are always those who don’t agree. It’ also means that there is no problem with short focal lengths as the sensor movement is relatively small and smooth.

Thread Tools	Search this Thread
	Search this Thread: Advanced Search

03-05-2009, 12:44 AM	#12
Gimbal Pentaxian Join Date: Apr 2007 Location: Sweden Posts: 2,106	Originally posted by ChrisA This thread on the UK Pentax forum has quite a lot of discussion about this one - MattMatic is extremely highly regarded over there and did some quite convincing work on the topic. IIRC, it started with some observations that surprised him, and he then looked into it more closely and reported the results. I wrote up my investigation into whether having SR on with the camera on the tripod made things better or worse, over there too. Thanks, will check it out.

03-05-2009, 03:32 AM	#13
Douglas_of_Sweden Site Supporter Join Date: Apr 2008 Location: Stockholm Photos: Gallery Posts: 3,374	Originally posted by Quicksand I haven't read the UK forum thread in detail, but I'll state my understanding of the situation: Just think of Newtonian physics. You know, every action has an equal and opposite reaction. When you hand-hold the camera, you hold it pretty tightly -- and you have a lot of mass applied to the camera. When the SR mechanism moves the sensor around, it has essentially no influence on your hands and arms because of the tremendous difference in mass. Also, your hands and arms are somewhat flexible, and would tend to damp any oscillation if there were any. A tripod, especially a light one, CAN be moved around slightly by the SR mechanism. The SR moves the sensor, the mass of which also shakes the camera body (you can feel this during dust removal, for example), and that shake is transferred to the tripod. The tripod then moves, and the SR has to compensate for that too. Tripods are not flexible, and there's some possibility you'll get a feedback loop where the SR tries harder and harder to stop the tripod from moving, when in fact it's exacerbating it. There's not a HIGH likelihood of this occurring, but I believe it has been observed "in the wild." That's an oversimplification, but you get the idea. It shouldn't hurt the camera (the SR can only do so much, and it won't continue to increase past its design limit), but it can hurt your pictures! Unless it shake so much that the camera fall over on your ultra light tripod and bam crash the lens right on some rocky surface

03-05-2009, 03:04 PM	#14
Class A Pentaxian Join Date: Aug 2008 Location: Wellington, New Zealand Posts: 11,251	Originally posted by Quicksand A tripod, especially a light one, CAN be moved around slightly by the SR mechanism. Yes, one night I wanted to make a 12 hour exposure of the night sky, put the camera on the tripod and released with the finger but forgot to turn off SR. Then next morning, the camera and tripod had walked away 50m or so! Sorry, tried to do a Gooshin.

03-06-2009, 01:38 PM	#15
Timd Senior Member Join Date: Dec 2008 Location: Cape Town, SA Posts: 262	Remembering my control theory: you need to have only one dominant time period in any control system, and that needs to be the external shake. On a tripod, the camera is far more rigid, in which case the natural frequency of movement is much greater than when hand held (your hands make very good shock absorbers). Thus if the sensor were to move, it might introduce vibration in the camera which might be at a similar frequency to the sensor movement. With these 2 movements at similar (high) frequencies, you can get positive feedback (classic case of 2 similar time periods in the control and the feedback loop, introducing instability). I also think the bottom line is that you should not have any shake while on a tripod! If you take a photo before the hand is showing, the SR has not managed to work out the movement of the camera, hence when the photo is taken, the sensor is not moving in time to the camera movement. My gut feel is that the SR is only really effective in the 1/15th to 1/500th sec shutter speeds: basically below 1/15th sec the movement is too big to compensate and faster than 1/500th the movement is too fast to respond to. Just my penny's worth...