I suspect there if very little if any difference in the Canon or
Nikon form of stabilization. They just call it different names.
Patent seaches reveal your suspicions to appear unfounded.
No, "VR" and "IS" are
not identical technologies.
Nikon, Canon, and others all hold various patents relating to this
issue. You have to look into details of each to see how
they differ. Nikon made a big deal about how it alone provided
final centering of the optical axis immediately before exposure.
Apparently this is important (or they want you to think it is
.
For VR, start with Nikon's own publication about this, noting in
particular the last line in my citation below:
http://www.nikon.co.jp/main/eng/portfolio/about/technology/nikon_technology/vr_e/index.htm
Nikon's VR System changes operation algorithms when the shutter
release button is lightly pressed and during exposure. Therefore,
when the shutter release button is lightly pressed, the VR lens
gives you a smooth viewfinder image. During the exposure,
however, the algorithm changes to compensate for every slight
movement. What's more, just before exposure, the VR lens will
reset to central position (optical axis) from an off-centered
position which is a result of VR operation during the shutter
release button is lightly pressed. Since the shift amount of
the VR lens is limited, this operation maximizes VR effects as
well as optical performance.
--> Only Nikon has this "Centering Before Exposure" feature. (Fig. 3)
To more understand just what they're talking through, you need to
read through Nikon's patent for this technology, a portion of which
I cite in detail here:
http://www.patentstorm.us/patents/5805937-description.html
After the ON states of the switches S1 and S2 are confirmed (S107:
Yes, S108: Yes), a centering operation of the VR lens 21 is
performed (S109) to continue the first VR control (S110). In the
centering operation the center of the VR lens 21 is positioned at a
predetermined initial position so as to assure the maximum driving
range of the VR lens 21. In this embodiment, the VR lens 21 is
positioned, so that its center agrees with the optical axis of the
image taking optical system.
Immediately after the first VR control is started, exposure is
started by driving a quick return mirror, stop, shutter, and the
like (S111). After an elapse of a predetermined period of time, the
exposure is ended by driving the shutter, stop, quick return mirror,
and the like again (S112). After the end of the exposure, the VR
operation is stopped (S113), and power supply to the respective
sensors is also stopped (S114). In addition, the VR lens is reset
(S115), thus ending a series of image taking operations.
[...]
In general, since the amplitude of vibrations is small in the high
frequency range and is large in the intermediate and low frequency
ranges, even when the second VR control that does not reduce
vibrations in the high frequency range is performed, vibration
reduction errors during the second VR control are small. Therefore,
in this embodiment, when the vibration reduction control method is
switched to the first VR control mode thereafter, normal vibration
reduction with high precision can be performed again within a
sufficiently very short period of time in practice.
Note that the recognition performance, upon observing an object
image on the finder, of the photographer is normally lower than that
upon observing a still picture. Therefore, like in this embodiment,
if vibrations of only intermediate and low frequency components are
to be reduced in a non-exposure state, i.e., when the photographer
observes an image on the finder, the photographer does not
practically feel it.
So IR and VR are not identical technologies. But do differing technologies
provide identical results?
That's perhaps the more important question,
and now I don't know. Maybe they do, but quite possibly they do not. You
wouldn't
think they would, given how convergence and equivalence are much
less common than their absence; different inputs provide different outputs.
I know no reliable, 3rd-party testing standard that can tell how "good"
each one is at what they are trying to do and in which circumstances.
If you expect 4 stops at 300mm, can you really expect the same at 20mm?
I don't think so. The same tech may not do the same thing across all
focal lengths.
All accounts do suggest that Nikon's second-generation VR system, dubbed VR
II, is more effective than its older models, by at least 1 and perhaps 2
stops. However, accounts vary.
Probably it depends at least in part on how good the sniper/marksman skills
of the tester are. What all changed between generations of Nikon's VR?
Probably several things. There may well be additional patents involved.
And this is just talking about optical measures in the lens proper, not
sensor alignment as non-optical and/or non-analogue companies sometimes
provide in their bodies (although Nikon did it first). This would seem
especially important for predictive AF tracking (birds in flight; sports;
etc) in long glass.
VR and IS both attempt to effect the same result using somewhat different
technologies. Only speaking vaguely can you say they're same sort of thing.
Speaking with precision, they are not. Also, Nikon is using the more
honest term here; it's VR, not IS. You'd think no harm would come of this,
but it has. Toy-camera makers who apply high shutter-speeds with negligent
ISO-pushes and call it (digital) image stabilization, pretending it's all
the same when it's anything but.
--tom
