Velocity Reviews - Computer Hardware Reviews

Velocity Reviews > Newsgroups > Computing > Digital Photography > Re: Help please - Optical image stabilization with telephoto adapter?

Reply
Thread Tools

Re: Help please - Optical image stabilization with telephoto adapter?

 
 
David J Taylor
Guest
Posts: n/a
 
      05-15-2008
Bob Cunningham wrote:
> Some cameras, like Canon G9 and Canon S5 IS, have provision
> for attaching a telephoto adapter or a wide-angle adapter.
>
> If the camera has optical image stabilization, as do those
> two cameras, will that stabilization still be effective with
> the added zoom adapter?


Yes.

David


 
Reply With Quote
 
 
 
 
David J Taylor
Guest
Posts: n/a
 
      05-18-2008
Paul Allen wrote:
> On Thu, 15 May 2008 10:02:48 +0000, David J Taylor wrote:
>
>> Bob Cunningham wrote:
>>> Some cameras, like Canon G9 and Canon S5 IS, have provision for
>>> attaching a telephoto adapter or a wide-angle adapter.
>>>
>>> If the camera has optical image stabilization, as do those two
>>> cameras, will that stabilization still be effective with the added
>>> zoom adapter?

>>
>> Yes.

>
> What David meant was that a zoom adapter will have no effect on the
> proper functioning of image stabilization. He should have added that
> any shaking that gets past the stabilizer will be magnified by the
> added zoom.
>
> I use a tripod when I care about my pictures, but that's just me.
>
> Paul Allen


I don't carry a tripod around in my pocket, or on a field trip or visit to
the races. I find that Image Stabilised lenses suit my needs admirably.

When I want a much longer exposure I just find some natural object against
which I can brace the camera.

David


 
Reply With Quote
 
 
 
 
Dave Martindale
Guest
Posts: n/a
 
      05-18-2008
"David J Taylor" <(E-Mail Removed)-this-bit.nor-this-bit.co.uk> writes:
>Bob Cunningham wrote:
>> Some cameras, like Canon G9 and Canon S5 IS, have provision
>> for attaching a telephoto adapter or a wide-angle adapter.


>> If the camera has optical image stabilization, as do those
>> two cameras, will that stabilization still be effective with
>> the added zoom adapter?


>Yes.


Are you certain of that? It's not what I'd expect, though I don't have
such a camera to test.

Here's my reasoning: Suppose the adapter is a 2X telephoto adapter.
Suppose your hand holding the camera shakes +- 1 degree in one axis.
Because of the angular magnification of the telephoto adapter, the main
lens of the camera, looking through the adapter, sees +- 2 degrees of
angular motion, not 1 degree. The camera's acceleration sensors
detect the camera shake and move an internal lens element (or prism) by
the appropriate amount to cancel 1 degree of shake - not 2 degrees. To
properly cancel the larger apparent shake, the stabilization system
needs to double the gain of the system, in terms of compensation per
degree of shake motion.

The anti-shake system will automatically adjust its gain as required as
the main lens changes focal length via zooming, but it doesn't know that
the tele or wideangle converter is there.

Now, if the camera has a menu item to tell the camera that the tele or
wideangle adapter is currently mounted, ignore what I said above - the
camera does have the required information to set the antishake system
properly.

Also, if the magnifiying device is rigidly mounted and only the camera
shakes, then the stabilization would still work correctly. This might
be the case if you are handholding the camera at the eyepiece of a
telescope or microscope, shooting through the eyepiece. But not when
the camera and other optics are both shaking.

Dave
 
Reply With Quote
 
David J Taylor
Guest
Posts: n/a
 
      05-18-2008
Dave Martindale wrote:
> "David J Taylor"
> <(E-Mail Removed)-this-bit.nor-this-bit.co.uk> writes:
>> Bob Cunningham wrote:
>>> Some cameras, like Canon G9 and Canon S5 IS, have provision
>>> for attaching a telephoto adapter or a wide-angle adapter.

>
>>> If the camera has optical image stabilization, as do those
>>> two cameras, will that stabilization still be effective with
>>> the added zoom adapter?

>
>> Yes.

>
> Are you certain of that? It's not what I'd expect, though I don't
> have
> such a camera to test.
>
> Here's my reasoning: Suppose the adapter is a 2X telephoto adapter.
> Suppose your hand holding the camera shakes +- 1 degree in one axis.
> Because of the angular magnification of the telephoto adapter, the
> main
> lens of the camera, looking through the adapter, sees +- 2 degrees of
> angular motion, not 1 degree. The camera's acceleration sensors
> detect the camera shake and move an internal lens element (or prism)
> by
> the appropriate amount to cancel 1 degree of shake - not 2 degrees.
> To
> properly cancel the larger apparent shake, the stabilization system
> needs to double the gain of the system, in terms of compensation per
> degree of shake motion.
>
> The anti-shake system will automatically adjust its gain as required
> as
> the main lens changes focal length via zooming, but it doesn't know
> that
> the tele or wideangle converter is there.
>
> Now, if the camera has a menu item to tell the camera that the tele or
> wideangle adapter is currently mounted, ignore what I said above - the
> camera does have the required information to set the antishake system
> properly.
>
> Also, if the magnifiying device is rigidly mounted and only the camera
> shakes, then the stabilization would still work correctly. This might
> be the case if you are handholding the camera at the eyepiece of a
> telescope or microscope, shooting through the eyepiece. But not when
> the camera and other optics are both shaking.
>
> Dave


Dave,

Thanks for your input. I had been assuming that the magnification of the
adaptor would not affect the stabilisation, and whilst I am still unsure
about exactly what happens (in that isn't the stabilisation magnified by
as much as the image?), I do accept that for best results you would need
to tell the camera that the optical chain had changed via a menu. I note
that the Panasonic FZ20, for example, has just such a menu item.

Cheers,
David


 
Reply With Quote
 
Archibald
Guest
Posts: n/a
 
      05-19-2008
On Sun, 18 May 2008 15:52:42 -0700, Bob Cunningham
<(E-Mail Removed)> wrote:

>We need to keep in mind that the stabilization technique
>used on a given camera may react to vertical and horizontal
>movements only, so that angular movement of the longitudinal
>lens axis could not be compensated. In a Wikipedia article
>about image stabilization at http://tinyurl.com/5jm42k *, it
>says in part
>
> In Nikon and Canon's implementation, it works by
> using a floating lens element that is moved
> orthogonally to the optical axis of the lens, using
> electromagnets. The vibration signal which is
> compensated for by the stabilizing lens element is
> typically acquired using two piezoelectric angular
> velocity sensors (often called gyroscopic sensors),
> one to detect horizontal movement and the other to
> detect vertical movement.
>
>In a review of the Canon G9 at
>it says
>
> Lens-Shift type optical image stabilization system
>
>About the S5IS, it says at
>http://www.dpreview.com/news/0705/07050703canons5is.asp
>
> Image stabilization Yes (Lens shift-type)
>
>To counter the angular motion that DM discussed, wouldn't
>the system have to rotate the lens rather than just shift
>it.


Horizontal and vertical movement of the camera isn't going to blur
images much, except for stuff that is close to the camera. So IS must
be able to deal with angular movement, I think.

Archibald
 
Reply With Quote
 
David J Taylor
Guest
Posts: n/a
 
      05-19-2008
Bob Cunningham wrote:
[]
> To counter the angular motion that DM discussed, wouldn't
> the system have to rotate the lens rather than just shift
> it.


Bob,

The lens senses an angular motion, and then uses a shift of elements
within the lens to produce an angular displacement of the image. If you
like to look at it this way, a linear shift, at a finite distance, becomes
an angular shift.

Cheers,
David


 
Reply With Quote
 
Chris Malcolm
Guest
Posts: n/a
 
      05-19-2008
David J Taylor <(E-Mail Removed)-this-bit.nor-this-bit.co.uk> wrote:
> Dave Martindale wrote:
>> "David J Taylor"
>> <(E-Mail Removed)-this-bit.nor-this-bit.co.uk> writes:
>>> Bob Cunningham wrote:
>>>> Some cameras, like Canon G9 and Canon S5 IS, have provision
>>>> for attaching a telephoto adapter or a wide-angle adapter.

>>
>>>> If the camera has optical image stabilization, as do those
>>>> two cameras, will that stabilization still be effective with
>>>> the added zoom adapter?

>>
>>> Yes.

>>
>> Are you certain of that? It's not what I'd expect, though I don't
>> have
>> such a camera to test.
>>
>> Here's my reasoning: Suppose the adapter is a 2X telephoto adapter.
>> Suppose your hand holding the camera shakes +- 1 degree in one axis.
>> Because of the angular magnification of the telephoto adapter, the
>> main
>> lens of the camera, looking through the adapter, sees +- 2 degrees of
>> angular motion, not 1 degree. The camera's acceleration sensors
>> detect the camera shake and move an internal lens element (or prism)
>> by
>> the appropriate amount to cancel 1 degree of shake - not 2 degrees.
>> To
>> properly cancel the larger apparent shake, the stabilization system
>> needs to double the gain of the system, in terms of compensation per
>> degree of shake motion.
>>
>> The anti-shake system will automatically adjust its gain as required
>> as
>> the main lens changes focal length via zooming, but it doesn't know
>> that
>> the tele or wideangle converter is there.
>>
>> Now, if the camera has a menu item to tell the camera that the tele or
>> wideangle adapter is currently mounted, ignore what I said above - the
>> camera does have the required information to set the antishake system
>> properly.
>>
>> Also, if the magnifiying device is rigidly mounted and only the camera
>> shakes, then the stabilization would still work correctly. This might
>> be the case if you are handholding the camera at the eyepiece of a
>> telescope or microscope, shooting through the eyepiece. But not when
>> the camera and other optics are both shaking.
>>
>> Dave


> Dave,


> Thanks for your input. I had been assuming that the magnification of the
> adaptor would not affect the stabilisation, and whilst I am still unsure
> about exactly what happens (in that isn't the stabilisation magnified by
> as much as the image?), I do accept that for best results you would need
> to tell the camera that the optical chain had changed via a menu.


Unless the IS system was adaptive, noting its over- and underestimates
and adapting to cancel them out. I don't know if any camera maker uses
such a system, but it's well understood how to do that well, and it
automatically avoids the problems of being improperly informed.

--
Chris Malcolm http://www.velocityreviews.com/forums/(E-Mail Removed) DoD #205
IPAB, Informatics, JCMB, King's Buildings, Edinburgh, EH9 3JZ, UK
[http://www.dai.ed.ac.uk/homes/cam/]

 
Reply With Quote
 
Dave Martindale
Guest
Posts: n/a
 
      05-20-2008
"David J Taylor" <(E-Mail Removed)-this-bit.nor-this-bit.co.uk> writes:

>Dave,


>Thanks for your input. I had been assuming that the magnification of the
>adaptor would not affect the stabilisation, and whilst I am still unsure
>about exactly what happens (in that isn't the stabilisation magnified by
>as much as the image?), I do accept that for best results you would need
>to tell the camera that the optical chain had changed via a menu. I note
>that the Panasonic FZ20, for example, has just such a menu item.


The stabilization happens behind the adaptor, so its effect is *not*
magnified by the adaptor. But the effect of shake is magnified by a
tele adaptor (and reduced by a wideangle one) because, to the camera, it
looks like the world is shaking and the adaptor magnifies the apparent
movement of the world.

Dave
 
Reply With Quote
 
Dave Martindale
Guest
Posts: n/a
 
      05-20-2008
(E-Mail Removed) writes:

>We need to keep in mind that the stabilization technique
>used on a given camera may react to vertical and horizontal
>movements only, so that angular movement of the longitudinal
>lens axis could not be compensated.


No, you're thinking of the wrong motions. Shake can cause the camera to
move in any combination of 6 directions. There are 3 translation
directions (left-right, up-down, near-far) and three angular motion
diretions (pitch, yaw, and roll).

Now, when you're using a long focal length lens (the usual place where
stabilization is used), none of the translation motions are significant
sources of blur. When the subject is much further away from the camera
than the lens focal length, the small translational movements due to
shake produce *much smaller* motions of the image on the film plane, not
enough to worry about.

It is the pitch and yaw rotations that matter most. The longer the
focal length, the more apparent motion you get at the image plane for a
small amount of rotation. And that's what lens stabilization systems
correct for. They measure pitch and yaw *rotations* of the lens or
camera body, and move a lens element or prism to shift the image an
amount that approximately compensates for the shake-caused shift.

In theory, you could also cause blur by roll around the camera optical
axis, but (a) the amount of blur is not magnified by the lens focal
length, and (b) there's no simple way to compensate for this anyway.
So stabilized lenses do not correct for roll motion, to the best of my
knowledge.

>In a Wikipedia article
>about image stabilization at http://tinyurl.com/5jm42k *, it
>says in part


> In Nikon and Canon's implementation, it works by
> using a floating lens element that is moved
> orthogonally to the optical axis of the lens, using
> electromagnets. The vibration signal which is
> compensated for by the stabilizing lens element is
> typically acquired using two piezoelectric angular
> velocity sensors (often called gyroscopic sensors),
> one to detect horizontal movement and the other to
> detect vertical movement.


Right. It measures rotation around pitch and yaw axes, and compensates
by Y and X-direction image shift.

>In a review of the Canon G9 at
>it says


> Lens-Shift type optical image stabilization system


Ditto. They say "lens shift" to distinguish it from the sensor-shifting
systems in the bodies of some other manufacturers.

>About the S5IS, it says at
>http://www.dpreview.com/news/0705/07050703canons5is.asp


> Image stabilization Yes (Lens shift-type)


>To counter the angular motion that DM discussed, wouldn't
>the system have to rotate the lens rather than just shift
>it.


But I wasn't talking about roll, I mean pitch and yaw.

Dave
 
Reply With Quote
 
Dave Martindale
Guest
Posts: n/a
 
      05-20-2008
Chris Malcolm <(E-Mail Removed)> writes:

>Unless the IS system was adaptive, noting its over- and underestimates
>and adapting to cancel them out. I don't know if any camera maker uses
>such a system, but it's well understood how to do that well, and it
>automatically avoids the problems of being improperly informed.


To do this, you'd need images fed back from the sensor to the
stabilization system, plus some moderately powerful hardware to do the
correlations. Doing this in a video camera would be relatively simple,
since it's already producing 60 images per second. A P&S camera could
do the same, likely at a somewhat slower frame rate.

But a *film* camera cannot provide any image feedback at all, and most
DSLRs provide only one image. A stabilization system for use on these
cameras has to operate open-loop, without feedback, so that's what all
the Canon and Nikon stabilized lenses must use. P&S cameras, and DSLRs
with live LCD preview, *could* have IS with image feedback for
self-tuning, but so far I haven't heard of one using it.

Dave
 
Reply With Quote
 
 
 
Reply

Thread Tools

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are On
Pingbacks are On
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
Long telephoto lenses with image stabilization -- any manufacturers other than Canon and Nikon? Wally Digital Photography 9 10-20-2012 06:19 PM
Point and Shoot with wide angle lens (28mm min), optical or electronicviewfinder, image-stabilization, and support for external flash (hot shoeor wireless) SMS 斯蒂文• 夏 Digital Photography 4 11-21-2007 09:00 PM
Lens stabilization vs Camera stabilization Al Clark Digital Photography 119 12-09-2006 01:30 PM
Swivel lcd + lens stabilization + high optical zoom Norm Digital Photography 7 09-20-2004 04:34 PM
Image stabilization in high optical zoom digicams swingman Digital Photography 6 07-15-2004 11:46 AM



Advertisments