In-camera IS at long focal lengths

[Loupey]

... But at higher shutter speeds, or when using a tripod, where IS is not really needed, it not only doesn't add anything, it detracts slightly...

That's something that I would have to prove to myself under highly controlled conditions before I believed it - at least with really long telephotos. A tripod stout enough to render the IS unusable would be far too bulky/heavy for my kind of shooting.

In my mind, I have a picture of sliced bread... right next to that I have a picture of IS

[Anbesol]

I wasn't saying IS was useless at telephoto ranges, just not as helpful.

[OldClicker]

I wasn't saying IS was useless at telephoto ranges, just not as helpful.

I see it as the opposite and for the reasons you stated. At telephoto lengths is where it is needed. The 'shake' is more pronounced and the DOF is more critical. - TF

[Anbesol]

I see it as the opposite and for the reasons you stated. At telephoto lengths is where it is needed. The 'shake' is more pronounced and the DOF is more critical. - TF

Replace the word 'helpful' with 'effective' then, helpful wasn't the right word.

Loupey - I too would be very interested in your experiment.

[Loupey]

Replace the word 'helpful' with 'effective' then, helpful wasn't the right word.

Like OldClicker, I too was a bit confused by what you initially stated. Even if you said, "I wasn't saying IS was useless at telephoto ranges, just not as 'effective'", it still might not be clear to others. But I think what you mean is that, at the same low shutter speed, the IS will help remove the effect of camera shake more with shorter focal length lenses?

But since telephotos magnify everything, it is unreasonable to use the same shutter speed analogy at increasing focal lengths. So longer telephotos, in addition to requiring higher shutter speeds, do indeed benefit greatly with IS technology.


As for the IS experiment, I will do this in the spring. It won't apply directly for in-camera systems, but it is something I'm curious about.

[Ron Kruger]

Well, I'm not real sure about it. I'm just beginning to experiment with this, but so far, I've noticed more sharpness with the IS off. What I have observed is just with a 200mm, and it's a f/2.8, so I can often get more than 1/200th shutter speed. When hand-held, I'm going to turn it on under that and turn it off over that. As Clicker pointed out, the longer the focal length the more pronounced the shake. If I was shooting a 400mm, I wouldn't turn it off unless I had 1/500 or higher, and with a 600mm, I don't image I'd ever turn it off.
You're excellent at these kinds of experiments under controlled conditions, Loupey, and I for one would be very interested in your results.

[Sushigaijin]

I find that mirror lock-up is only useful in the 1/2 to 1/8 range. Any slower shutter, and the mirror slap isn't long enough to matter. Any shorter, and the exposure is too fast to be effected. Composing with the LCD is ungainly, for now. I'd imagine the newest crop of DSLRs (olympus, at least) with p&s style live-view make it a lot easier.

[Ron Kruger]

I'll be waiting for your experiment, Loupey.
But this discussion leads into a natural question about the difference between lens IS and camera IS.
The way lens IS technology was explained to me by a friend who owns a photography studio, is that some kind of computerized system in the lens remembers what was in sharpest focus during a certain time before and up to the moment the shutter is pressed and then chooses that best image to send to the sensor. (Not sure I'm explaining this well.)
What I understand about in-camera IS technology, is that the sensor itself detects tiny movements when the shutter is pressed and compensates by moving within the camera to compensate.
They're both pretty amazing, but if I'm correct, they are totally different technological appraches to the same problem.
I'm not sure which is best, but even though I own a Pentax, it seems the lens IS technology would be slightly more reliable and accurate. I knew about this and considered it before buying my K20D, and chose it because of the larger megapixels for the money and the ability to have IS for every lens I use (again a cost factor because IS lenses of equal speed and quality are much more expensive.)
Nor am I sure Loupey's experiments with IS lenses would be applicable to in-camera IS.
As Pentax suggest, does Canon also recommend turning the IS off when using a tripod?
(I'm thinking outloud with a keyboard here.)
It would also seem to me that with in-camera IS on, the sensor is sort of floating, ready to move, while an IS lens, because of the time-lapse way it opperates, would work equally on a tripod or hand-held, and if that's the case, Loupey's experiment wouldn't apply to in-camera IS.
Whew

[Loupey]

The way lens IS technology was explained to me by a friend who owns a photography studio, is that some kind of computerized system in the lens remembers what was in sharpest focus during a certain time before and up to the moment the shutter is pressed and then chooses that best image to send to the sensor. (Not sure I'm explaining this well.)

The lenses aren't that smart They don't really send anything - they employ servo motors to compensate for small movements along two axes. That does bring up a question of start-up time as in-lens IS requires a half second or so to spin up. I would imagine that the in-camera systems are quicker and use less energy.

As for Canon recommending that IS be turned off when tripods are used, yes and no. The latest generation IS (third or fourth) doesn't require the IS be turned off. I'm not sure if the IS is actually working when mounted to a tripod or if the servos can detect that a tripod is being used (by the lack of movements) and simply shut down.

[Ron Kruger]

Thanks, Loupey. Then the technology is much more similar than I thought, and your experiments should be at least somewhat applicable to both systems. Please test it with different focal lengths.
I don't know about "start-up time." When I turn IS on, I hear a click of some kind instantly, but I can't imagine how a half-second would make any difference.

[OldClicker]

Thanks, Loupey. Then the technology is much more similar than I thought, and your experiments should be at least somewhat applicable to both systems. Please test it with different focal lengths.
I don't know about "start-up time." When I turn IS on, I hear a click of some kind instantly, but I can't imagine how a half-second would make any difference.

When you press the shutter button half way and the IS comes out of 'sleep' mode, it takes some time for it to get in sync (or I gues out of sync) with the camera shake. - TF

[Loupey]

...but I can't imagine how a half-second would make any difference.

As a fellow nature photog, you know how long one must sometimes be poised to get the shot. Others might not realize that we have our cameras up to the eye and ready to shoot for minutes on end. Doesn't sound hard until you point a camera and hold it steady for even just a minute. Now I'll stand there and keep the shutter button pressed half-way so that the IS motors are keeping the image steady - but only for about a minute or two at a time (to keep down the battery drain and to minimize wear-and-tear on the servos). But if the action suddenly erupts while the servos are resting, that 1/2 second can cost a shot. I shoot anyway without waiting - sometimes the stabilization takes effect by the 2nd or 3rd shot. Sometimes it doesn't.

A tripod, lens with IS switched OFF, and live-view enabled would definitely be ideal in these types of cases, but it doesn't happen regularly enough for me to carry a tripod anymore. I'll follow a hunting dragonfly from here to there all day with the camera handheld or on a monopod where it would be nearly impossible to chase with a tripod. But then that dragonfly suddenly comes to rest on that perfect stalk and I have to wait, and wait, and wait for it to take off/land to get the action shot.

[Ron Kruger]

I'm not sure I understand, Loupey. Once I turn IS on, it's on and ready. I don't have to activitate it in any other manner, such as depressing the shutter half way. I presumed lens IS worked the same way.
If it has to be reactivated every time you want to take a shot, that half-second would be a big deal, indeed.
Or, did I misunderstand what you said?

[Loupey]

In-lens IS must always be activated before shooting. Even with the camera on, the shutter release button (or whatever camera switch you designate) must be pressed to activate the IS servos. These servos turn off after about 2 seconds after releasing the switch. It has to be this way - the servo mechanisms on the larger lenses draw a lot of power and imagine all the wear-and-tear if they were always on when one is walking through the woods.

I would think that in-camera IS works the same way for the same reasons. You just can't see it or feel it.

[A.M.D.A.]

In-lens IS must always be activated before shooting. Even with the camera on, the shutter release button (or whatever camera switch you designate) must be pressed to activate the IS servos. These servos turn off after about 2 seconds after releasing the switch. It has to be this way - the servo mechanisms on the larger lenses draw a lot of power and imagine all the wear-and-tear if they were always on when one is walking through the woods.

I would think that in-camera IS works the same way for the same reasons. You just can't see it or feel it.

Hmm... with my FZ7K there are two MEGA O.I.S. modes, Mode 1 and Mode 2.

Mode 2 is activated as you describe, only when the shutter button is tapped. IS disengages right after you take the picture or when you release your finger off the shutter button.

Mode 1 is activated all the time, whether your finger is on the shutter-button or not. You get a stabilized image in the VF/LCD, so long as the camera is turned on and is in any shooting mode. (e.g. aperture-priority, program, manual, etc.). Yes, I have noted that my battery power goes down faster than usual, and in quiet conditions the IS noises are quite audible. And I would imagine that Mode 1 puts a lot of strain on the camera/lens.

[Ron Kruger]

Come to think of it, when I was setting my camera up, I was given these choices as well. That must be why sometimes my batteries don't last very long and other times much longer. When I have the IS on, it drains it much more quickly, I guess because it is on all the time. But I've never noticed any hesitation or noises during IS use, or any noises when I depress the shutter half way.
I also have a dial on the camera to turn the IS on or off, so I guess my IS is either on all the time, or off all the time. From what Loupey says, I would prefer it the way it is.

[Greg McCary]

I think on my E3 the IS is only working during the exposure and not when the shutter is pressed half way. At times I can hear it working. I have also forgot and had the IS on while mounted to a tripod and I didn't notice any difference in IQ, even on some long exposures.

[A.M.D.A.]

Perhaps I should've titled this thread "In-camera IS versus in-lens IS"...

[Telecorder]

FWIW, DPR's evaluation of super zoom systems came to the conclusion that...
"...It's also obvious that you can't expect any image stabilization system to work miracles now that these big zooms have started to stretch well beyond 400mm (equiv.). Our experiences on that score showed a clear advantage to those cameras using lens stabilization over sensor-shift, and that the design of the camera itself plays a big role in how still you can hand-hold at long focal lengths (the lightest cameras being the least stable, obviously)."

http://www.dpreview.com/reviews/q109...oup/page16.asp

Yes, in-lens IS/VR/OS are hampered by the need for initiation and speed up when implemented during action shots but apparently offer the greatest stabilization benefits at the expense of either battery drain waiting on action to commence or loss of initial action sequence(s).

It's similar to the old adage of lens lust...

Choose 2 and only 2 of the following -->

Long Reach
Good-Great IQ
Low Price

Maybe we need to add Great & Effective image stabilization to the list but you only get to choose 3 of the 4:mad2: