PDA

View Full Version : Image Stabilization in camera Vs Lenses....


senthil
19th June 2008, 07:01 PM
Last month i saw one small text content from the leading indian photography magazine..


That is Image Stabilization in camera Vs Lenses....


In-camera IS works with all lenses compatible with the camera, thus reducing the overall cost of your lens system. Smaller focal length lenses require extremely small sensor deflection making the IS quite effective. However, larger zoom and tele lensees require an extreme amount of sensor deflection because the image coming in through the lens moves around a lot. In these cases, in-camera IS is not as effective as IS built into the lens, which would show marginally better performance. Lenses with IS built in tend to be rather expensive , especially if you plan to buy many of them. you need to weigh these against each other and make a choice.

regards, senthil

Who's_E
19th June 2008, 10:39 PM
Interesting learnings on the telephoto side of things.

I attended the London Camera Exchange E-System demo day and bought one of those tiny E-System beanbags for 10. It is a phenomenally effective piece of IS and works with all lenses... :)

250swb
19th June 2008, 11:41 PM
However, larger zoom and tele lensees require an extreme amount of sensor deflection because the image coming in through the lens moves around a lot

Yes, but consider how big the Canon and Nikon large zooms are in comparison to Olympus of the equivalent focal length. The Olympus zooms and fixed lenses are both ligther and shorter in length than those big IS stabilised Canikon lenes. So the parameters of what is good for 'best' IS change again, and what is good for one system is a ridiculous over engineered and expensive solution for another.

Canon and Nikon need lenses with IS ability because they are such big things that only a system that can work around the nominal point of axis will work. Shorten the physical length of the lens and this diminishes, so an in body system can keep up an effective IS.

Steve

Zuiko
20th June 2008, 02:01 AM
Yes, but consider how big the Canon and Nikon large zooms are in comparison to Olympus of the equivalent focal length. The Olympus zooms and fixed lenses are both ligther and shorter in length than those big IS stabilised Canikon lenes. So the parameters of what is good for 'best' IS change again, and what is good for one system is a ridiculous over engineered and expensive solution for another.

Canon and Nikon need lenses with IS ability because they are such big things that only a system that can work around the nominal point of axis will work. Shorten the physical length of the lens and this diminishes, so an in body system can keep up an effective IS.

Steve

Not strictly true as it's the magnification and angle of view of the lens that really determines the effect of camera shake, rather than focal length. Thus a 300mm Four Thirds lens will be just as prone to shake as a 600mm full frame lens. Without any form of IS both lenses in this example should be used with a shutter speed of 1/500th sec maximum and ideally 1/000th sec or less.

Having said that, other factors of course come into play. Some photographers naturally have much steadier hands than average and can get away with much slower speeds. Others, like myself (I have Parkinson's Disease), need to use higher shutter speeds than most other photographers - with or without IS.

There will also come a point, as you say, where the sheer weight and bulk of a large lens will additionally add to the risk of shake caused by the magnification and angle of view. This is where the smaller and lighter Four Thirds lenses start to score. However, not all Four Thirds lenses can be considered lightweight. The 90-250mm for example is an excellent lens but a real beast. I would never even attempt to hand hold it.

Which takes me to my next point. The most effective image stabilization of all is a sturdy tripod (the heavier the better) or a bean bag placed on a firm surface. As regards to which IS system is best I accept the arguement that lens based stabilisation is slightly more effective than systems built into the camera body on the basis that it can be optimized for each particular lens, but nevertheless consider that a camera body system is more convenient, cost effective, and ensures that all lenses are covered. I must say the Olympus IS system has made a tremendous difference to me and I cannot believe that othe systems could be anything more than marginally better at the price of being much less flexible.

Finally, one other point in favour of the Four Thirds format that is often overlooked is the fact that wider apertures achieve the same depth of field as smaller apertures (at least 2 stops over full frame) on larger formats. This applies more to standard and wide angle lenses as telephotos on all formats will often be used wide open. But then again this also gives Four Thirds lenses an advantage as, being of shorter focal length, they often have larger maximum apertures than equivalent lenses on larger formats, allowing faster shutter speeds or lower ISO ratings.

Hmmmm....... earlier I said that I accepted that lens based IS systems were marginally more effective. Maybe I've just convinced myself that a body integral system is in fact the best when combined with the Four Thirds advantage. :D

But just one final word of caution. Lenses such as the Oly 70-300 are deceptively small and light but in this example when zoomed to maximum you are still using a 600mm equivalent focal length. That's some lens and, even with IS, should be treated with respect!

DTD
20th June 2008, 06:28 AM
I can never hold a camera steady (or level for that matter).

I've been impressed with the IS in the E-3 and as I seem to be using longer focal lengths than I used to, switch it on fairly often.

When using film I invariably use ISO 400, but in digital try to keep it at ISO 100, so IS is a nice feature.

As to which is best… for me it's an academic question – Olympus have it in-body, so that's what I use. The old saying about the best camera being the one you've got in your hands at the time holds true.

Ian
20th June 2008, 06:57 AM
Both types of IS have their advantages and disadvantages, but the nice thing about Four Thirds is that you have the choice of both :) I found that in-lens IS found in the Panasonic Leica stabilised lenses was similar in effectiveness to the E-510's moving sensor IS. In theory, AF will benefit from in-lens IS as it stabilises the view to the AF sensor. However, moving sensor IS now works with all lenses, regardless of whether they are stabilised or not.

Ian

Ian
20th June 2008, 07:45 AM
Both types of IS have their advantages and disadvantages, but the nice thing about Four Thirds is that you have the choice of both :) I found that in-lens IS found in the Panasonic Leica stabilised lenses was similar in effectiveness to the E-510's moving sensor IS. In theory, AF will benefit from in-lens IS as it stabilises the view to the AF sensor. However, moving sensor IS now works with all lenses, regardless of whether they are stabilised or not.

Ian

I forgot to mention - with in-lens IS, in theory, moving a corrective element off-axis inside the lens is an optical compromise you don't get with moving sensor IS, though of course if the sensor is shifted extremely at the time of exposure, the side or corner of the image will be more into the boundaries of the image circle where there is less sharpness and more corner shading, usually. In practice, both these concerns are less problematic than you might think.

Ian

HughofBardfield
20th June 2008, 12:08 PM
From my PoV, one of the big advantages of in-body IS is the ability to use it with legacy lenses. The firmware update to the E510 that added the legacy lens stabilisation finally produced some nearly usable images from a knackered old Sirius 500mm I've had for years - so much so that I went out and bought a Tamron SP 500mm to play with. Unfortunately, weather and work have stopped me from trying it out seriously yet... However, I am looking around for an OM 300mm f4.5 if I can find one for "the right price".

theMusicMan
20th June 2008, 12:57 PM
Hi Hugh - can I ask which OM-E-Series adapter do you use...?

HughofBardfield
20th June 2008, 02:11 PM
Hi Hugh - can I ask which OM-E-Series adapter do you use...?

Hi John - I was a mug punter and wasted my money on the genuine Olympus adaptor (at least it came from Hong Kong, so it was a bit cheaper!) :o

At the time, I knew no better! There are, I'm told, numerous Fleabay and other adaptors that are just as well made. The Fotodiox ones have a good rep (http://www.fotodiox.com/shop/index.php?cPath=27&sort=2a&page=3 ), but I have been told that most of the "sensible" price ones are OK. They are, after all, a fairly simple machining job if you have the right equipment.

Someone recommended this brand recently: http://cgi.ebay.com/OM-lens-to-Olympus-E-500-E-330-E-300-E-1-Adapter-NEW_W0QQitemZ150014625010QQihZ005QQcategoryZ30059Q QcmdZViewItemQQ_trksidZp1742.m153.l1262

The third-party ones come in numerous flavours and, apart from Canon, I think almost any kind of legacy 35mm - and some medium format - lens can be used.

senthil
20th June 2008, 05:45 PM
Nice discusssions...

About the IS quality in camera...

Sure the focal length is a matter....
But i feel can we consider the sensor size?

Like 4/3 sensor (it's very small compare to full frame)
Small sensor deflection is enough for the stabilization...


In Olympus, the short focal length lenses and small size sensor is a major adventages for Image Stabilization..

It's just funny & confusing imagination...

Can we imaginate, if we use in built IS lens in the in built IS camera...

R MacE
20th June 2008, 08:17 PM
Lenses with IS have 4 or thereabouts extra elements compared to the same lens without IS (I'm thinking of Canon lenses here) That must have a detrimental effect on IQ.

Interestingly when Canon introduced their Anti-Dust measures they claimed that systems that used a filter in front of the sensor were inferior as they required an extra 'Piece of glass' in front of the sensor which reduced IQ :rolleyes:

Don't forget that IS lenses were developed to suit film cameras where moving the sensor (film in this case) wasn't an option regardless of which system was better.

Fluffy
20th June 2008, 08:47 PM
It may well be true that lenses with IS have lower image quality generally than lenses w/o IS. However you couldn't prove that statement by using the Panleica 14-150. It is razor sharp. If your thesis is true then it scares me to think what it would be like with the extra built-in elements for IS.

Chrs, Stv

250swb
20th June 2008, 09:12 PM
Zuiko wrote "Not strictly true as it's the magnification and angle of view of the lens that really determines the effect of camera shake, rather than focal length. Thus a 300mm Four Thirds lens will be just as prone to shake as a 600mm full frame lens. "


Well of course, but I think you missed the point.

If an Oly lens is shorter than a Canon lens, for instance, the deflection caused by movement is less. So if both are fixed to a body on a tripod and rotated 1 degree upwards around the sensor plain the end of the longer Canon lens will move further upwards than that of the Olympus lens. This is easy to prove using some graph paper. This is why it is more efficient to have the IS in the longer Canon lens, because it reduces the effect of that extra deflection.

Come the day that Canon fit IS into their equivalent of a 25mm pancake lens then I think it would be fair to call the technology equal, until then Olympus have a better system for day to day work precisely because all lenses are image stabilised.

Steve

Zuiko
20th June 2008, 10:28 PM
Zuiko wrote "Not strictly true as it's the magnification and angle of view of the lens that really determines the effect of camera shake, rather than focal length. Thus a 300mm Four Thirds lens will be just as prone to shake as a 600mm full frame lens. "


Well of course, but I think you missed the point.

If an Oly lens is shorter than a Canon lens, for instance, the deflection caused by movement is less. So if both are fixed to a body on a tripod and rotated 1 degree upwards around the sensor plain the end of the longer Canon lens will move further upwards than that of the Olympus lens. This is easy to prove using some graph paper. This is why it is more efficient to have the IS in the longer Canon lens, because it reduces the effect of that extra deflection.

Come the day that Canon fit IS into their equivalent of a 25mm pancake lens then I think it would be fair to call the technology equal, until then Olympus have a better system for day to day work precisely because all lenses are image stabilised.

Steve

Hi Steve, I'm not sure I agree. I did conceed that the sheer weight and bulk of a Canon (or Nikon) lens is certainly a contributory factor towards shake but, within reason, I think the length of the lens is immaterial. Surely it is the movement of the sensor in relation to the path of light striking it that creates blur, rather than the point at which the light enters the front element. Also, the focal length is the true optical length of the lens regardless of how efficiently that lens has been shortened in physical length by the telephoto (as opposed to telescopic) design.

But I may be wrong. I freely admit I'm straying on to thin ice here, as far as my knowledge of physics is concerned!

What really matters is that the Olympus in body IS works remarkably well and, with the choice of certain Panleica IS lenses and support for legacy lenses, is certainly the most versatile system available. In this respect I'm in full agreement with you. Olympus do have the best system. :)

Cheers,

Ian
21st June 2008, 07:47 AM
Hi Steve, I'm not sure I agree. I did conceed that the sheer weight and bulk of a Canon (or Nikon) lens is certainly a contributory factor towards shake but, within reason, I think the length of the lens is immaterial. Surely it is the movement of the sensor in relation to the path of light striking it that creates blur, rather than the point at which the light enters the front element. Also, the focal length is the true optical length of the lens regardless of how efficiently that lens has been shortened in physical length by the telephoto (as opposed to telescopic) design.

But I may be wrong. I freely admit I'm straying on to thin ice here, as far as my knowledge of physics is concerned!

What really matters is that the Olympus in body IS works remarkably well and, with the choice of certain Panleica IS lenses and support for legacy lenses, is certainly the most versatile system available. In this respect I'm in full agreement with you. Olympus do have the best system. :)

Cheers,

My understanding of the issue concurs with John's - the physical size of the lens isn't known by the camera, only its focal length. Steve is correct that a focal length is a focal length, but it's the effective magnification in relation to the sensor size that matters. That magnification is simply a cropping factor - this applies to all lenses and all frame size standards. So it is perfectly reasonable to say that a Four Thirds lens of 300mm has the same angle of view as a 600mm lens on a 135 format (so-called full frame) camera.

Although the 300mm four thirds lens should see less image shift on the sensor in terms of millimetres for a given degree of movement of the body compared to a 600mm lens on a 135 format camera, the proportion of the sensor shifted would be the same and so the correction would be the same 'power' relative to the sensor size. Put that 600mm lens on a Four Thirds camera and the corrective power would need to be doubled, of course.

Ian