Image stabilization in lens or camera body

[albertjohnmichael]

Image stabilization in lens or camera body; which is best?
And in what camera combination is best?

[Sushigaijin]

In-camera IS at long focal lengths

:thumbsup:

[Ron Kruger]

Good idea, Al. I'd like to see this subjected discussed further.
A lot of people think the camera they chose is the best (especially if they paid quite a bit for it), without having actuall experience with others, but that's not the way I feel.
I shoot an in-camera IS, but I'm familiar with the work a friend of mine (a professional that owns a local photography studio) gets with his Canon IS lenses.
Like most everything else, there are advantages and disadvantages to the mechanics and technology of each, and I would like to see these explored.
The quality of the lens is also a factor. IS lenses are relatively expensive to begin with, so they come with good glass. The advantage of having an internal IS camera, is the technology is there for whatever lens one uses, but if you put cheap glass on it, you're not going to get nearly as good results as with the expensive IS lenses.
One advantage with in-camera IS, is that one can buy high quality lenses for considerably less than what comparable IS lenses cost.
But which technology is better?
Even though I shoot a good in-camera IS, I have suspected that the IS in lenses is a slightly better technology. But I don't really know.
Another factor in all of this is the camera's auto-focus capabilities and how those affect in-camera and in-lens IS functions? Start-up delays, ect?
We have plenty of things to investigate.

[freygr]

It depends on your style and type of photography you shoot. It real does not mater if it's in the lens or camera body the image will be stabilized. The amount of movement that either system can compensate for will very by camera model or lens depending on the manufacturer.

With telephoto lenses a image stabilizer would make the most difference with free hand use.

[Wild Wassa]

I've been wondering for a while, whether my Canon Ultrasonic EFS 17-85 is better with the IS turned off. If the image is taken with the camera's functions set to their optimum sharpest recording settings, while the camera is placed at its optimum stability then the lens is set to its optimum Nyquist settings (something that I can do, the real world stuff, not that optical bench quality stuff).

The lens is described here ... have a play with the moving scales. This will draw your attention to basic factors that affect zoom lens quality (to varying degrees) ... http://www.dpreview.com/lensreviews/..._c16/page3.asp

I took these four shots several minutes ago after reading this thread and they are my starting point for my test. The shots are not an overall range of image quality tests, just easy hand held situations to do firstly, and shot at 1/5th of a second ... the shots are straight out of the camera.

With the IS off ... enlargement to 804 pixels wide. Nearly all of my images are 804 wide that I put on this Forum.





With the IS on.





Full size enlargement with IS off from image #1.





Full size with IS on from image #2.





I don't think that I'll question whether IS works at 1/5 of a second. Most of my hand held shots are taken at the slower range of speeds, below 1/25th sec, with many as slow as 1/5th or a 1/6th of a second when inside.

The next two images are both shot on a tripod.

With IS on.





With IS off.





My conclusion from this simple test, using this lens is; the best overall image quality and contrast (by a fraction of a degree) was achieved with the camera on a tripod, the mirror locked up, the self timer set and with the IS turned off ... by employing good camera handling techniques consistent with the subject and available lighting.

As a note: in photograph #1, I can't account for the slight magenta colour shift. I've left it as is, because it could be an artefact caused by the system, which could be a problem ... but I can't say, with so few test images.

Warren.

[Ron Kruger]

Thanks, Wassa. The book case is a good test subject. Pretty much what one would expect. There's no doubt IS is a great aid when the shutter speed is less than the focal length. I'll be waiting for more.
Please be sure to shoot a series at higher shutter speeds. That's where I suspect IS not only doesn't help, but might be a deterent.
I was going to do a similar thing with my camera and 200mm, but had to send my camera back for repairs.

[Wild Wassa]

Ron, I was also wondering if IS degrades the image at the higher shutter speeds. At the slower speeds employing IS is (meant to be) equivalent to 3 - 4 stops improvement in shutter speed in reducing camera shake. ... but there is often a sting in the end of the tail ... as you have just hinted.

I also subscribe to the recipcal of the lens's focal length as the minimum shutter speed ... or I try to, in daylight if hand holding. It is a good rule of thumb whenever possible.

This thread is a perfect trigger to now find out where the degrading point is, if one exists. I'll also shoot the bookcase with the camera on a tripod using the self timer, to see what the optimum sharpness possible is with my setup.

Warren.

[SmartWombat]

On my 100-400 L IS there's an interesting side effect of the IS, random colour fringing.
If it was the lens itself, then the colour fringe adjustments in LR would be the same in each shot (like my EF-S 10-22).

[Wild Wassa]

Ron and Paul, I've added two more images to the bottom my first post, if they are of interest to you and to others of course.

I hope that Forumites with IS in their cameras, might submit a similar IS test.

Paul, I'll see if I can duplicate the problem that you have mentioned, I'll try that this afternoon with one of Canons L teles if I can borrow one. My kids have the cool lenses in our house ... :17: .

Doing this test has been a good gap to fill in my understanding of using my camera system and what to expect. I learnt how to use Canon's Live View today, also to go into the programme and enable the Autofocus for the live view mode and even use the self timer.

I should also mention that the sound of the IS turned on in the lens is an extraordinary sound of microbes playing with Lego and that the images above were shot in the 'Standard' picture mode with the functions set to default. If I had noticed variations with IS in either the off/on functions, this 'User Settings ' feature could have been perfect to change and correct the variables. Lesser quality lenses with IS could benefit from this function, if problems arise.

'Thin Living in Fat World' ... a photo manipulation that I wish that I had the IS on for. I can't see in the meta data if I did or hadn't. A meta data issue that Canon could address.





Warren.

[albertjohnmichael]

Thanks to all for this informative tread. AJM

[Ron Kruger]

On my 100-400 L IS there's an interesting side effect of the IS, random colour fringing.
If it was the lens itself, then the colour fringe adjustments in LR would be the same in each shot (like my EF-S 10-22).

Wombat: I have noticed a slight difference in IQ between shots taken with IS on and IS off, which is the reason I started questioning it and decided to stop using IS as a cure-all and just use it when needed. Those taken with IS off, when a higher shutter speed warrants, seemed crisper, but I haven't taken the same shot with IS on and off to make a real comparison. When I get my camera back, I'll do that with my 200mm, trying to keep PP as identical as possible, and post them here.

[Wild Wassa]

Sorry to take so long to post further results of the IS tests. Sometimes work must take a precedence over doing (this) more serious stuff.

The Canon 100-400L lens used for this part of the test really has a focal length 105-400mm (so the metadata says). With the IS switched to IS 1. I didn't use or test the IS 2. The shots are again, straight out of the camera and in (about) the worst lighting that I could find.

105mm taken on a tripod with the IS off. 1/10sec @ f18.0.





105mm on tripod with the IS on. 1/10sec @ f18.0.





Zoomed to 400mm on tripod with the IS off. 1/60 @ f5.6.





400mm on tripod IS on. 1/60 @ f5.6.





400mm hand held with IS off. 1/640 @ f5.3. By this stage the setting sun was flairing badly and there was plenty of bush fire smoke around, as always. An on-going problem when living in the Bush Fire Capital of the World.





400mm hand held with IS on. 1/640 @ f5.3





Detail from 400mm hand held with IS off. The enlargements are at 'actual view' size.





Detail from 400mm hand held with IS on. Even under worsening conditions that the final image was taken under, having the IS on shows a real benefit when hand holding the camera.





My conclusion is turn the IS on when handholding the camera if sharpness is a priority. I also found no real downside to leaving IS switched on but it is up to each photographer to test their equipment.


Warren.