Does VR degrade image quality? « Nikon Rumors Forum

Nikon Rumors Forum » Nikon Lenses

Does VR degrade image quality?

(49 posts) (25 voices)

Tags:

12 Next »

Floyd

member
Joined: Apr '09
Posts: 36
offline

I've seen this alluded to in passing and also stated plainly as a fact. But I have not seen anything to suggest this, besides what has been said.

Does VR really affect image quality? That is, if camera shake is not an issue, does switching off VR improve image quality?

Does a VR lens with VR switched off produce the same quality image as a non-VR equivalent lens (maybe this would only be a theoretical test)?

Obviously, at alower shutter speeds (relative to focal length), VR will improve image quality by stabilising the image for the duration of the exposure if the lens (or, more particularly, the sensor) moves relative to the subject. My question is really about whether or not VR degrades an image, with all other factors being equal. I think you know what I mean. Excuse me: I'm not a rocket scientist. I just make pictures.

?

Posted 3 years ago #
NikoDoby

The Terminator
Joined: May '09
Posts: 6,598
offline

There was a discussion on this topic before. It depends on some conditions but for the most part image quality benefits from having VR.

Posted 3 years ago #
Floyd

member
Joined: Apr '09
Posts: 36
offline

Hmmmm. Must have missed that discussion. "search" sounds good.

Posted 3 years ago #
NikoDoby

The Terminator
Joined: May '09
Posts: 6,598
offline

The search doesn't really help since I'm pretty sure the discussion came up within another topic. I can't remember which though and I don't feel like digging through several pages of search results to find it :^(

So feel free to just discuss it here. I guess this is an example of why we mods should keep threads on topic ;^)

Posted 3 years ago #
Floyd

member
Joined: Apr '09
Posts: 36
offline

I couldn't find it with the search facility. Clearly, the moderators are not doing their job !!!! They are too busy posting, probably. Having said that, I personally think that free range forums are better, as opposed to those caged types.

I hear what you say Nikodoby, but what I want to know is - beyond the image stabilising benefits, does VR otherwise affect the quality of an image, if it is not adding to it by stabilising the image. That is, if VR is on, but is not needed to stabilise an image (say, for example, for an exposure shot at 2000th of a second), does it have any impact on the image?

I am curious because I have seen it talked about authoritatively as a fact, but some 'facts' originate as urban myths.

Posted 3 years ago #
NikoDoby

The Terminator
Joined: May '09
Posts: 6,598
offline

Yeah the mods really suck around this place! Anyway there is a right and wrong why to use VR. For example Active versus Normal modes. I thought you meant image quality with a lens with VR turned off versus a lens with no VR at all.

Here's Nikon's VR learning center
http://imaging.nikon.com/products/imaging/technology/vr/index.htm

Posted 3 years ago #
Floyd

member
Joined: Apr '09
Posts: 36
offline

Thanks for the link, Nikodoby.

"I thought you meant image quality with a lens with VR turned off versus a lens with no VR at all." - I kind of did mean this: The question I have is as follows:

If, for example, an image of a stationary subject in bright light is captured at 1/2000th of a second with a lens such as a 70-200 VR, at 70 mm, I assume that the VR mechanism is doing nothing at all, since it would not need to. Is it safe to assume that the same lens without a VR facility/option (with exactly the same optics, etc.)would produce exactly the same image? To put it another way, does a lens with VR turned off, when taking a picture of an subject which does not need to be stabilised, produce an image which is the same as the same lens (which would be a hypothetical lens) without a VR facility/option?

Man, this is kind of hard to explain.

Or maybe I'm just splitting hairs and it doesn't really matter that much ! Feel free to "moderate" me out (:-] if this seems to be more about philosophy than photography.

Posted 3 years ago #
NikoDoby

The Terminator
Joined: May '09
Posts: 6,598
offline

Floyd I think what you're really asking is if a tree falls in the forest and no one is there to hear it does it make a sound? ;^)

Posted 3 years ago #
Floyd

member
Joined: Apr '09
Posts: 36
offline

NikoDoby said:
Floyd I think what you're really asking is if a tree falls in the forest and no one is there to hear it does it make a sound? ;^)

Probably. It's late and I've had a long day. I'll come back later.

Posted 3 years ago #
NikoDoby

The Terminator
Joined: May '09
Posts: 6,598
offline

My opinion is that image quality isn't worst with a VR lens versus a non VR lens. The same lens elements are still there it's just that on one lens the element moves. I think the confusion comes from using Active VR versus Normal VR. Tripod versus hand held on a moving boat etc.

Posted 3 years ago #
soap

preferred member
Joined: Jul '09
Posts: 606
offline

Actually, I don't see a single Nikon VR lens out there which is an identical optical design to the non-VR variant.

In all cases I looked at the VR variants have extra elements, with the rear group being the ones in motion - strongly suggesting that the VR is not utilizing the original lens elements, but rather additional ones.

Clearly the addition of additional elements means, in theory, reduced image quality from some hypothetical ideal lens, but practice is soooo far from theory that this is not a safe conclusion to draw. Many of Nikon's best lenses are complicated beasts with more elements than I have fingers and they do quite well - including control of flare and ghosts.

In a nutshell, the performance of a lens is dependent on the effort and money the designers wish to employ. VR vs no-VR is not a safe measure of that.

PS - "Active" vs "Normal" is often described as "Yaw correction" vs "No yaw correction" (yaw = panning for sake of argument). Pitch correction is what is almost always needed as the "up and down" motion is the most common from hand-holding, and roll correction can't be done with (current?) in-lens solutions.

Posted 3 years ago #
bmxdad

preferred member
Joined: May '09
Posts: 1,864
offline

soap said:
Actually, I don't see a single Nikon VR lens out there which is an identical optical design to the non-VR variant.

In all cases I looked at the VR variants have extra elements, with the rear group being the ones in motion - strongly suggesting that the VR is not utilizing the original lens elements, but rather additional ones.

Clearly the addition of additional elements means, in theory, reduced image quality from some hypothetical ideal lens, but practice is soooo far from theory that this is not a safe conclusion to draw. Many of Nikon's best lenses are complicated beasts with more elements than I have fingers and they do quite well - including control of flare and ghosts.

In a nutshell, the performance of a lens is dependent on the effort and money the designers wish to employ. VR vs no-VR is not a safe measure of that.

PS - "Active" vs "Normal" is often described as "Yaw correction" vs "No yaw correction" (yaw = panning for sake of argument). Pitch correction is what is almost always needed as the "up and down" motion is the most common from hand-holding, and roll correction can't be done with (current?) in-lens solutions.

I have not seen any IQ difference between VR being turned ON or Off, but one area that you should be careful with is shutterspeed, VR will allow slower shutter speed and moving objects can now become blurred, So if I am shooting anything moving I usually turn the VR off or set my shutter speed to a specific speed

Pete

Posted 3 years ago #
jonnyapple

Goldfingers
Joined: May '09
Posts: 3,400
offline

Floyd said:
I personally think that free range forums are better, as opposed to those caged types.

I love the analogy, and I agree.

I'd listen to soap. I think he's right that -if- VR degrades image quality, it can be designed for in high-end lenses and they'll still outperform low-end lenses (not to mention the VR can help with the slow shutter-speed shots like pete and niko mentioned). I think it's a step forward.

Posted 3 years ago #
soap

preferred member
Joined: Jul '09
Posts: 606
offline

PPS - The is much ambiguity over whether or not "Active" VR actually changes how a shot is taken, or only your view in the viewfinder pre-exposure. One can read the Nikon documents to strongly imply that in "Normal" mode panning (yaw) correction is disabled _pre shot_ only, as Nikon's system always recenters the VR element (upon shutter release) and then, and only then, applies the final correction. This would mean that the difference between "Normal" and "Active" is only in what corrections are made while the shutter button is half depressed, an aid to composition, and has nothing to do with what corrections are made once the button is fully pressed.

None of the other informed discussions on Nikon's docs (Thom's included) appear to address this specific question.

This would be a difficult question to answer through field experimentation, as you would need to repeatably reproduce consistent panning motion between shots, and is probably one I should just email Nikon on (and pray for a respondent with half a brain).

Posted 3 years ago #
optimaforever

senior member
Joined: Nov '09
Posts: 85
offline

I have seen that VR on my 70-200 produces white and/or black contrasted halo around moving people when shooting at 200mm , 1/125s under a cloudy weather, making the people difficult to cutout. So I prefer shooting at 1/500s or 1/750s and no VR for moving people. VR is nice for shooting landscapes when you're tripodless, or perhaps immobile wildlife.

Posted 3 years ago #
heartyfisher

preferred member
Joined: Apr '09
Posts: 1,701
offline

By my undersatnding, VR is like a little Movable TC at the end of the lens. So it does Theoretically add some small degradation to the image. However, Lens design technology has improved so much that the degradation of a VR lens elements compared to the previous version that does not have VR is more than made up for by the design improvements.

Difference between VR off and VR? I can see 2 or 3 scenarios here.
1) If the shutter speed is high then VR off is likely to be better because VR may "shake" the image while trying to compensate while the picture is being taken. as the shutter actuation actually shakes the camera abit.
2) if the shutter is low then VR ON will of course be better as thats what it is designed to do.
3) Due to the movement of the VR elements, the image may not be the best due to the VR elements not being optimally alligned. So if you have a very high shutter speed or have the lens on a very sturdy tripod. Its best to have VR off.

But what do I know... I may be totally wrong. :-)

Posted 3 years ago #
tai

preferred member
Joined: Mar '09
Posts: 315
offline

I've heard VR lenses tend to have worse boke than their Non-VR counterparts, but it's pretty hard to tell whether that's the VR causing that or just that the new lens has worse boke regardless of VR.

Posted 3 years ago #
warprints

preferred member
Joined: Mar '09
Posts: 769
offline

soap said:
Clearly the addition of additional elements means, in theory, reduced image quality from some hypothetical ideal lens, but practice is soooo far from theory that this is not a safe conclusion to draw. Many of Nikon's best lenses are complicated beasts with more elements than I have fingers and they do quite well - including control of flare and ghosts.

I agree. If fewer lenses mean better image quality, then all lenses would have one perfect element. I realize that what you're saying is that by adding more lenses, there is greater propensity/likelihood of IQ degradation, as each lense element adds the opportunity for problems.

On the flip side, by adding extra elements, maybe the manufacturer is able to better correct or control IQ -- and charge for the extra effort. (Not likely, as Pro lenses should already be the best they can do anyewhere near a given price point.)

Posted 3 years ago #
Panamon_Creel

senior member
Joined: Jun '09
Posts: 95
offline

heartyfisher said:
....
1) If the shutter speed is high then VR off is likely to be better because VR may "shake" the image while trying to compensate while the picture is being taken. as the shutter actuation actually shakes the camera abit.
....

If it has a positive effect with slow shutter where mirror flap is also present why would it have a negative effect with fast shutter speeds?
confused :~

Posted 3 years ago #
soap

preferred member
Joined: Jul '09
Posts: 606
offline

I also don't understand why it would have a negative effect with high shutter speeds. It (VR) "resets to zero" as soon as you press the shutter all the way, and then and only then applies the final correction. Maybe I haven't thought it all the way through - but I would think a faster shutter speed wouldn't matter.

Posted 3 years ago #
Willis

preferred member
Joined: Mar '09
Posts: 1,123
offline

I have both the 18-55 and the 18-55 VR. Truth be told, I believe they are optically identical with the exception of some nano coating on the non-VR version to prevent ghosting. I never have noticed the difference, but I suppose I could shoot the two side by side with VR turned off if anyone is curious.

Posted 3 years ago #
heartyfisher

preferred member
Joined: Apr '09
Posts: 1,701
offline

Panamon_Creel said:
If it has a positive effect with slow shutter where mirror flap is also present why would it have a negative effect with fast shutter speeds?
confused :~

Again, I may be blowing hot air through places we cannot talk about! :-) but, I think this is the case because the vr system moves the image WHILE the shutter is open to compensate for the camera movement. The shutter causes a very quick spike in movement and vr kicks in to compensate for it as quickly as posible. this causes the image degradation. At relatively longer expotures this small degradtion is compensated for by the longer exposure of a "good" steady image. So at high shutter speeds there is double the degradation of the image, once by the shutter itself and then by the VR trying to compensate for that spike and no "clean" image to overwhelm the badness :-) lol sorry running out of words..!! hehe ...

Posted 3 years ago #
Johndbr

preferred member
Joined: Sep '09
Posts: 102
offline

What is best for me is to use VR at shutter speeds of 1/350 or slower with my 500mm on the tripod of course, but not have the head locked down. It seems that most of the time I get sharper photos at lets say 1/750 sec with VR off. When it's off, you don't have to wait that little bit to let VR settle down. This takes a little longer when hand holding a lens like the 70- 300mm. It depends on the subject and how fast you have to move to get the shot. Like frmming a bird in flight and getting the shot before it's gone . You might only have a few seconds to get it. Anyway I make a judgement from what I've learned the last few years and about 50,000 photos when I use it or not. This may or may not be what others have figured out for there type of shooting. Everyone shoots a little different and different subjects.

Posted 3 years ago #
Floyd

member
Joined: Apr '09
Posts: 36
offline

Johndbr said:
What is best for me is to use VR at shutter speeds of 1/350 or slower with my 500mm on the tripod of course, but not have the head locked down. It seems that most of the time I get sharper photos at lets say 1/750 sec with VR off. When it's off, you don't have to wait that little bit to let VR settle down. This takes a little longer when hand holding a lens like the 70- 300mm.

Hey John That has been my expereince and the reason for seeking the opinions of others by starting this thread. I conclude that VR is useful and does not degrade an image when it has the opportunity to do its job - that is, the job of stabilising an image. Clearly, if the wrong, or inappropriate VR mode is selcted, the image may not benefit either.

Posted 3 years ago #
Bland

preferred member
Joined: Nov '09
Posts: 883
offline

All I know is my pictures are much better when using the VR.

Posted 3 years ago #