Camera Labs

An interesting article. I did not see discussion distortion introduced by lens based image stabilization. Looking at diagrams of how the lens elements move with a fixed sensor position there must be some degree of distortion introduced and likely it would be much greater than the moving sensor method. The downside is of course that image stabilization using the moving lens element is apparent in the viewfinder of an slr. I also was surprised to see that the Pentax K10d did not perform better. Pentax touts their image stabilization as operating in 3 axis vs 2 for most other moving sensor image stabilization slrs. An earlier comment about the type of vibration introduced is also relevant. I wonder what the performance differences would have been if vibrations of different frequencies and axes would have shown. I didn’t see whether Pop Photo did any analysis of how a typical human actually moves. I would guess this varies greatly across people and shooting conditions i.e. outside the wind is certainly a factor, or in other occasions shooting from an unstable platform such as a boat or moving auto. .. so I don’t know what the true verdict shows… other than there is typically 2 stops of improvement. If the image stabilized glass is 2 stops slower to begin with why go the expense of image stabilized lenses?

_________________
Japhy

Well Japhy, I can't see a connection between IS and aperture. There is hugh glass out there with or w/o IS and there is normal glass out there with or w/o IS.
But normally the price to pay for 3 aperture larger glass is much higher than for a IS-system doing the same. Can you imagine what a 18-200mm F1.4-2.8 would have cost
So I wouldn't complain...

_________________
Thomas (beware: Nikon-fanboy and moderator!) My Lens Reviews, My Pictures, My Photography Blog
D800+assorted lenses

Or, indeed, when you are attending a concert at which the audience around you is moshing, while you are trying to take a picture of the band. This could be helpful to some of our fellow forum members who are into concert photography. (hint, hint)

Anyway, on a more serious note... Nikon claims its vibration reduction system, which as we know is lens-based, is superior to that of some of the other brands, which is sensor-based.

The rationale for this being that, for each individual lens, there is a tailor-made vibration reduction device, and thus optimisation is achieved, unlike the system that relies on correcting shake at the sensor and must therefore become a "jack of all trades" (as it must correct every lens you want to attach to the body).

Is there any merit to this claim, or is it just marketing?

_________________
Nikon D90 + MB-D80 + SB-600 + AF-S VR 18-200mm f/3.5-5.6 + AF-S 35mm f/1.8 + AF 85mm f/1.8 | My photos on Flickr.

If the lab report from pop photo is telling the truth then this is just marketing.

_________________
Minolta primes: 2.8/20 2.8/24 1.7/50 2.8/135
Minolta zooms: 28-135 35-70 70-210(beercan) kit
Minolta cameras: Dynax 5D 7000
M42: Pentax Asahi S-M-C Takumar 1.4/50 and 2.8/105 Carl Zeiss Jena Flektogon 2.4/35 Industar-61L/Z Kiron 2.8/105 Macro

I dont know but I want one! IS or not ... probably better a constant F1.2 .. so it wont vignette at 1.8 while zoomed all the way out
.. Nikon, Leica, Zeiss are you there??

_________________
Japhy

Yeah, from Leica please with IS

What I'm currently wondering is the result of a tele-converter (say 2x)?!
On a lens-based system, the lens is dumb and will not alter the IS-programm so the residual shake (after IS does its work) will be magnified by the tele-converter (in this case x2) but also the "rule-of-thumb" would dictate that you use a shutter-speed that is twice as fast (as the focal length is doubled). So that should be "wash".
On a body-based system is the body "knowing" that the focal length has now doubled (with the 2x tele-converter)? If not, same as above: shake doubles and the correction stays the same thus you should use shutter-speeds twice as fast as without tele-converter. But if the body knows that the focal length is now double it could double it's reaction to shake and thus increase the effectiveness of IS by one stop!

Well, anybody here who knows about tele-converters and IS???

_________________
Thomas (beware: Nikon-fanboy and moderator!) My Lens Reviews, My Pictures, My Photography Blog
D800+assorted lenses

The use of tele-converters may be an argument against lens based stabilization. Although from what I have read about stabilization systems the motion is sensed by an accelerometer or rate gyro (both now are available mems (micro – electromechanical systems) and resemble an integrated circuit. Analog Devices currently manufactures several versions of mems gyro and accelerometer devices. The signal to actuate the movable lens element is computed by integrating the rate signal (gyro) 1x to compute position or integrating the accelerometer input 2x to get the new position as a result of camera shake. These computations must be done fast enough to sense the motion, compute the solution, and actuate the movable lens element during the motion and before the shutter closes. The same is true for the movable sensor type of motion stabilization. With the tele-converter in place it makes a difference in that the position of the moveable lens element further from the sensor and the stabilization compensation computed will not be correct. In the body stabilized version there is nothing to move in the lens assembly and all the accelerations and motion could be compensated by moving the sensor as Sony, Olympus, and Pentax. Olympus is actually promoting tele-converterers on their web page. That indicates to me that their image stabilization design is fully compatible. There are other arguments against tele-converters but the question you pose is indeed worth a test. Perhaps Gordon will do a piece on tele-converters with both types of stabilization???

_________________
Japhy

Next time I'm testing an IS lens, I'll try it with a teleconverter and report back. I bet someone else out there has a teleconverter though!

Japhy, you may be correct as to the benefits of body-based IS when using tele-converters. But your argument that the sensor gets put further away from the corrective lens-element does not hold, as in lens-based systems the sensor is also in the lens together with the corrective elements. So putting a tele-converter between the lens and the body, does not change the lens-based IS.

_________________
Thomas (beware: Nikon-fanboy and moderator!) My Lens Reviews, My Pictures, My Photography Blog
D800+assorted lenses

I think you will find Japhy is quite correct, the image sensor is further away from the lens stabilization mechanism by a factor of the tele-converters thickness and that is yet another algorithm that may or may not have been applied to the lens stabilization parameters, from all that I have heard, VR lenses DO NOT correct as well with a converter in place as they do when used with the lens alone. I think that we may have been diverted from the true course of this discussion by reference to two different sensors, one being the image sensor and the other being the stabilization sensor

Yep Andrew, you're right: I assumed Japhy meant the gyro-sensor.

But I'm not sure whether a different distance from the photo-sensor with lens based IS-systems makes a difference. The tele-converter works like bringing the photo-sensor forward (otherwise the lens would not focus correctly and just makes it virually smaller (viewed from the lens). Thus the sensitivity against redisual shake is amplified by the factor of the tele-converter.
But speculation is enough. Someone should have/buy a TC and make some tests. I hope Gordon will make good on his promise...

_________________
Thomas (beware: Nikon-fanboy and moderator!) My Lens Reviews, My Pictures, My Photography Blog
D800+assorted lenses

Andrew ..
Thanks for the clarification.. its exactly what i meant.

next question :

will lens or body stabilized cameras work when panning while in the portrait mode?

or are the algorithms specific to the landscape camera orientation?

will they pan vertically and remain stabil?

I would try all these experiments .. but the topic is buying a new camera.
Im leaning toward Olympus unless the Nikon D80 replacement is announced before early next year.

_________________
Japhy