In body image stabilization technology - why not Nikon?

[ilkka_nissila]

I had a chance to play with a Minolta 7D and some fast primes (28/2,

85/1.4) and was delighted with how the anti-shake seemed to work. I'm

happy with the 70-200/2.8 VR but 70 mm isn't very short and f/2.8

isn't all that fast, especially since it loses some sharpness wide

open. The short VR lenses are all slow zooms, yuck.

 

Since Nikon seems to like the DX format, shouldn't they license

Minolta's anti-shake technology and put it in their bodies? Personally

I could find lots of use for the feature in indoor available light

photography, as I imagine would others who e.g. shoot indoor events,

churches etc. I did not buy the Minolta setup as it was reserved for

another and also some of the lenses didn't have that impressive manual

focus actions. The 7D viewfinder is the best of the

reduced-sensor-size viewfinders I have seen. The VF image was sharp

and the 28/2 was very easy to focus.

 

I might even buy the 7D and the 28/2 and some other lenses to go with

it but one thing holds me back: it might happen that Nikon adopts the

technology and I already have lots of Nikon lenses. What do you guys

think - should Nikon implement it, and would you buy it if they did,

assuming that it costs something like $250 extra in the body.

 

Implementing it in an F-mount body would make sense because there are

so many manual focus lenses (e.g. Nikkors and Voigtlanders) suitable

for this application available on the second hand market, while

specialized Minolta lenses are more rare.

[ilkka_nissila]

Okay, I'll add a qualifier: the viewfinder was among the best; the D2 series also have good viewfinders though I didn't have one for direct comparison. But the Minolta vf is excellent for a body of its price range.

[ShunCheung]

First of all, doesn't Minolta have some patents on their anti-shake technology? Why would they license it to Nikon such that K-Minolta loses one of the few competitive advantage they have? That is like cutting your own throat.

 

My friend Bob Atkins once said that image stabilization is for people who have poor techniques. That statement may be an exaggeration, but a lot of serious photographers use tripod whenever they can, and IS/VR cannot stop subject movement. Therefore, its importance is somewhat limited. Some super teles can take advantage of IS, or in some indoor situations that a tripod is not practical or allowed. Otherwise, it is mainly for those consumers who don't want to bother with a tripod. That is why Canon and Nikon put it in many slow, consumer zooms for that market segment.

[vivek iyer]

I would love to see a Nikon Camera with a in body stabilzation.

 

Take all that VR, G etc out of a lens and let them be just lenses (with a proper aperture ring, focus mount, lens hood and a DECENT tripod collar).

[vivek iyer]

Nikon could (and should) bring out TCs with VR, IMO, just like the AF TC-16A.

[ecarter]

I wouldn't be surprised if Nikon (or Canon) couldn't impliment some form of "anti-shake" in the camera body itself - either licensing KM's or creating some work-alike feature, and may eventually do so, even though I'm sure KM have a patent on their particular brand of this feature. After all, both Canon and Nikon (and others) have Anti-shake or "Image Stabilization" or "Vibration Reduction" technology and both use it.

 

Wouldn't hold my breath though.

 

For technical reasons, though dpreview did say they didn't feel that their test on the 7D's antishake capabilities was by any means definitive, they did do a comparison between the 7D's in-camera antishake and the performance in one of Nikon's VR lenses. The 7D seems to give you about 2 extra stops of hand-holding - so does the in-lens VR, but it seemed to me that the in-lens VR is more consistant - which I'd expect, as it is really adapted to the lens itself, rather than an "every lens" solution.

 

Check their head-to-head comparison at http://www.dpreview.com/reviews/konicaminolta7d/page19.asp - it's an interesting read.

 

To me, though, the technical reason is a doubtful one. More likely, it has to do with profit margins.

 

Both Nikon and Canon are already very well established - I don't think it would be wrong to say that if you were able to poll everyone in the world who has a SLR, you'd find that the majority of people out there have a camera from either Nikon or Canon. Of course, both Nikon and Canon are going to want to sell cameras, but at this point I'd think they make their big money on lenses, with all those people already bought into their systems. They don't need antishake on the camera bodies - the people they are really selling to already are pretty tied to their camera systems - but now, they can sell lenses that do the same thing to all those people who already have Nikon or Canon bodies.

 

The situation for KM is reversed. They're a small player in the dSLR world compared to Canon and Nikon, and really want to attract people into their system. It makes much more sense for them to focus on adding the feature to the camera body itself and then saying "Look at us! All our lenses are stabilized!" If you aren't already dedicated to a camera system and having antishake is important to you, KM starts looking like a pretty good deal (which it is). You get the 7D based on that and you're on board with KM. There just isn't enough incentive to Canon or Nikon to do the same - at least not yet.

[ilkka_nissila]

K-M might license it to get money. There are many companies that actually live off patents. If it works as well as it seems, Nikon could actually charge much more than K-M currently does for the feature, leaving profits for both companies.

 

A 85/1.4 is two stops faster than the 70-200. With a 1.5x crop factor there is plenty of sharpness to be gained using image stabilization even when photographing people. I took a picture at 1/10 s with the 7D+85/1.4 and it seemed to be crisp. With digital you can grab several shots and some of them will be without subject motion.

 

Tripods are great, but in many places where there are lots of people, tripods cause a lot of distress in others. In other cases, for example, in travel photography, image stabilizer useable with every lens would be a great advantage. Yes, if you're a professional hired to shoot a location and can get permissions to shoot in cathedrals (because the owners are hiring you) you use a tripod. For much of professional architecture photography a tripod would be used. But I'm talking about photography for fun, amateur travel photography, indoor concert photography etc. where permits are not necessary because it's done for personal use.

 

I used to look down on the idea of IS/VR on short lenses, but I've gained more experience and I'm convinced that it's something useful as long as the lenses are optically good. You can always play with the odds and avoid subject motion by taking many shots.

 

I've read that photographers are sometimes prohibited from using a tripodin a church. Even if they're not prohibited, having someone moving about with a tripod during a ceremony would be quite a distraction and some people would be watching the photographer and not the ceremony. When you for example take a picture of the audience as a whole, having the image stabilizer would allow you to take the shot in available light and perhaps even stop down a bit for better sharpness. Someone might move in the audience, but mostly the pic will look far better than one taken with a flash. And the photographer disrupts things far less than if they're shooting with a tripod.

 

In concert and performance photography, tripods and flashes are often forbidden, and you're lucky if you can shoot at all (or you may shoot only for a limited amount of time). Again, fast lenses and AS would be very useful. Sometimes you want a shot where the dancers turn and you use a long exposure time to show the motion but nevertheless you would prefer the surroundings to be reasonably sharp.

 

There are tons of applications for the technology. I'm surprised that it hasn't received more attention. I'm afraid that people have forgotten about Konica-Minolta and are afraid to go with it, given their long absence. One thing that they'd need to get market share is new ultrasonic focusing lenses (they have only a few) - even Nikon has most of their zooms with AF-S now.

[joseph_smith3]

I beleive Nikon has such a camera--the Coolpix 990. Joe Smith

[vivek iyer]

Is that a Nikon product?! Darn!

[mattalofs]

"My friend Bob Atkins once said that image stabilization is for people who have poor techniques. "

 

Or for people that shoot in low light without a flash. The 7D allows me to use an SLR like I would a RF; I can handhold the 28 f2 down to 1/8th with suprisingly good results.

 

I get the point about tripods, technique, subject movement, etc, but there's a lot of potential to be explored with fast lenses and in body anti-shake. Flashes and tripods just aren't right for some kinds of photographers.

 

As good as the 7D is, a coarser ground glass screen and better low light af would make it a lot easier to take advantage the antishake system.

 

I'd buy a nikon with a similar system.

[lex_jenkins]

I can think of a scenario in which Nikon might license the anti-shake technology from K-M:

 

1. Technology continues to improve enough to enable further miniaturization so that there's room for AS inside the camera without compromising anything else or making the camera any bigger and heavier. As attractive as the K-M dSLRs in some ways, they won't satisfy many pros until they offer faster framerates, quicker AF and are otherwise comparable to cameras like the D2X and EOS 1D Mark IIN.

 

2. K-M heads toward the toilet financially. I wouldn't wish this on any camera manufacturer because it's bad for competition. But it'd be a shame to see their A-S technology disappear.

 

I've heard a few folks say that the current version of K-M's anti-shake technology isn't quite comparable to IS and VR lenses. They say it swims a bit more and isn't as effective. But I'm sure this can be improved. I recall the first optical and electronic image stabilization on camcorders several years ago tended to induce nausea.

[hash]

>> a lot of serious photographers use tripod whenever they can

 

Because they have to. If they could get as sharp a result without a tripod, I'm sure the only people who would bother with a tripod would be those who want to "feel important" or those in denial, like the vinyl record advocates. Besides, the 200-400VR lens for instance or even the 70-200VR hardly fits my idea of a consumer zoom!

 

I'd buy a Nikon body with built in VR for 250$ extra, but as Shun says it might not be easy for Nikon to license it off Minolta. They would have to develop it themselves. Then to prevent their own VR lenses from losing appeal they would have to make it work even better with lens VR turned on.

[ilkka_nissila]

Eric, I absolutely agree that N/C put it in lenses because they have more money to be made that way, and K-M in the body because they lost their status in the market and are a "newcomer". For the same reason they put a good viewfinder in their mid-priced DSLR, while the established players put crappy viewfinders in them. They want to get back in the market, while N/C want to sell expensive pro bodies to picky customers who want to see what they shoot.

 

Notice also that the 70-200 Nikon lens is a 2nd generation VR system (although not designated VR II) and is reported to work better than the one in the 80-400. I'm sure K-M's technology also improves over time. Even if they don't get it to 3 or 4 stops, it's still better than having only one fast lens with VR like Nikon has (the 70-200).

 

Canon can't do it as they're aiming for full-frame on all but entry-level bodies but Nikon doesn't have a FF DSLR yet and they could take advantage of this opportunity to apply in-camera stablization.I'm sure that some PJs would be very interested if Nikon does it, and they could partially compensate for the lost high ISO image quality which is due to the small pixels in small sensors. So it's an opportunity for Nikon to gain an advantage over Canon. But I guess they're too slow to react.

[ilkka_nissila]

Another possibility would be if K-M goes under, Nikon could buy the rights to the technology (I don't know how this goes in Japan, but I'm sure that the rights and some of the key designer people could be bought). Minolta always had some advanced innovations in their bodies, such as wireless flash ahead of its time. I really believe, with Nikon's push of DX sensors, that this could be an important way to capitalise on the extra image projected by regular 35 mm lenses.

 

The D200 is good, but a D200 with A-S worth two stops at $2k it would be a killer.

[mattalofs]

Since KM has teamed up with sony to produce slr's I'm not sure they going under any time soon. Or maybe they are.

 

If Nikon did implement something similar, it wouldn't work on all those DX lenses. They don't cover a 35mm frame, right?

[akira]

Well, if I remember correctly, K-M said that the CCD unit will move about 5mm max. from the central position when the anti-shake system is activated. You cannot predict the position of CCD when you push the release button. If the CCD is quite a bit off-center, it should exceed the image circle of DX lenses or it should be positioned where the performance of the attached lens is not the best even it is designed for the full frame.

[ilkka_nissila]

Sure, this is true that DX lenses wouldn't work on them but Nikon has made IX Nikkors for APS and they don't work on any other bodies. They made F3 AF lenses which don't AF on other bodies. Experiments are made.

 

In any case DX lenses can be used if you turn of the AS feature in the body. So nothing really is lost. The AS feature would be an extra feature available on FF lenses. Yes, some FF lenses have poor corner performance, but 5 mm is not enough to get to the corners. And good lenses are sharp right to the corners anyway. No one is forced to use or even buy a body with this feature, but I bet a lot of people would.

[mawz]

Nikon and Canon don't for 2 very good reasons.

 

VR and IS work with Full-frame sensors and Film Bodies. AS works with neither (Film transpots are way to big and heavy, full-frame sensors would have image circle issues, especially with Canon Wide-angles, which already have light fall-off problems at wide apertures on FF). KM's film camera business is dead, so is Pentax's (Pentax is likely going to introduce similar technology in their next set of DSLR's, based on the P&S's introduced at CES this week). Nikon and Canon need to support a large number of current film users as well as sub-frame and full-frame digital (Yes, there are full-frame Nikons, the Kodak's)

[hash]

>> You cannot predict the position of CCD when you push the release button

 

We need a better understanding of this. As far as I understand, the CCD follows the image in the frame to keep the framing constant. Now 5mm is a lot of travel for a DX sized sensor - almost one third its height. If your camera is shaking that much in the fraction of a second that the exposure happens - well, I'd say it's not the typical shot.

 

Even if it does, I'd think it would show up as light fall-off near the edge. big deal!

[ilkka_nissila]

Yeah, 5 mm motion with e.g. a 50 mm lens would be a really massive shake! In practice this is something that only happens with long lenses like a 400mm. In that case a lens stabilizer would be better.

 

Canon and Nikon make whatever they get the most profits out of. Bad viewfinders, poor af sensors, new lenses with incrementally improved features, and so on. In-lens stabilizers are fine but only a tiny fraction of lenses support this feature and most of them are hopelessly slow. Think about all the money that is wasted as people upgrade their lenses. I sold my 80-200 and got a 70-200. I suspect I will do the same when/if the 105/2.8 AF-S VR Micro comes out. This is a never-ending swamp of cash flow. An in-body stabilizer would conserve natural resources and because it supports fast lenses, hand-held photography would be possible in almost any light that you can see in.

[ShunCheung]

Any business should try to make money, or they would become a non-profit organization. Several people have mention that K-Minolta might go under and therefore may need to sell their technology. At least that is not at all a concern for Canon or Nikon at this time.

 

Another issue is that high-ISO results are improving quickly in DSLRs. In the film days ISO 800 was very iffy. People are now getting fairly good results from ISO 800, 1600 or perhaps even higher. That solves problems from subject movement as well as camera/lens shake and also makes IS/VR far less important.

[ilkka_nissila]

Shun, high iso compromises dynamic range and if you really go high, noise and muted colors are common. If you shoot in a high-contrast low-light indoor situation, you need all the dynamic range you can get.

Right now I get best results with black and white film in these conditions. In the future I expect to be able to get similar dynamic range from digital, and that means shooting at a really slow ISO setting. There is no free lunch.

[ilkka_nissila]

Just as an example, I shot with my D70 in downtown Boston and I went iso 800 I think, and the skin of black people was extremely dark and noisy in the images, yet their white shirts were overexposed. I also shot some 400UC in this case and that came out beautifully, although I did have to expose a bit longer. I'm sure the D2X and D200 are better in this respect but I've lived long enough to see incremental improvements often, true revolutions very rarely.

[akira]

Ilkka,

 

I don't necessarily oppose your idea, but I don't think it very practical or cost effective to design a camera that virtually requires another (larger) format. The flexibility of Nikon system is well kept because Nikon is concetrating on the DX format. Canon makes DSLRs in three different format which makes its system rather tricky. Also, 10mm addition of image circle for AS should require significant increase of size, weight and cost, which seems to be well proved by the lenses for Olympus DSLR system.

 

Now that Nikon has released D200, we can enjoy backward compatibility of the Nikkor system and usability of a bunch of strange lenses in metered modes at lower cost, which should be more important.

[ralf_strandell]

In-camera image stabilization would help me a lot when I'm doing available light photography "dynamically" in a crowd. But so would wide and fast VR lenses...

 

I've tried to use a tripod, but as soon as the camera was locked to the tripod, I could not take a shot. Tripods are incredibly restricting and probably only good for architecture/landscape/product photography - and studio portraits. Try to photograph people using a tripod... You have to ask them to turn, stay, smile... Try to capture the moment. It will fail. One could use a monopod, of course. Much simpler. But you are still going to be "the man with the stick" and stand out in the crowd. And it's still not as flexible and comfortable as a plain hand held camera. I don't love flash photography either. I find it too distracting to my taste.

 

We must remember that in-camera stabilization doesn't stabilize the viewfinder. Any sports photographer using a long non-stabilized lens would certainly greatly suffer. Hence I can accept the present situation where VR belongs to the lens. Stabalized viewfinder is worth it...

 

I want an AF-S DX 30mm f/1.4 VR desperatly. My 35mm f/2 is both too slow and too shaky (and just a tad too narrow, maybe - but that's a minor head ache). Sigma almost has the lens: 30mm f/1.4 but without VR.