In shooting HDR on my tripod, with 28-300 Nikkor, I found some of my shots not to be tack sharp as Nikon glass should be. I turned off VR and they significantly improved. Just did a hand held shoot at a party and found the same thing. Inadvertantly, I had left the VR ON and got some blur. Turned it off, and things sharpend up.
Is this my imagination or has anyone else found this?
Not your imagination, VR is supposed to be shut off when on a tripod or when shooting at shutter speeds over 1/500. It can introduce image degradation in some circumstances if left on in those circumstances.
Further, it does take a half second or so for the VR to settle, if you just snap a photo without first half pressing the shutter and pausing ever so briefly before snapping the photo it can also cause image degradation.
That list is now 10 years old; in the meantime several new and updated VR lenses have been released.
However, the principle is unchanged - lenses like the 28-300mm referenced in the original question should have VR turned off when the camera is mounted on a solid, locked tripod. The advice given above, to give time for VR to engage and settle, is also good - for any VR lens.
As Dave says, this feature doesn't exist today, and Nikon may never implement it. Personally, I don't like the idea as I prefer to decide myself when to use VR.
But... if it were to be implemented, I'm not sure that it would require a new channel of communication between camera and lens. It's the camera that initiates VR in the lens (when the shutter button or AF-ON button is pressed, depending on your settings). The camera also knows the shutter speed selected, and could perhaps decide by itself to not transmit the "VR ON" signal to the lens when the shutter speed is higher than a certain value.
Mon 09-Dec-13 01:26 PM | edited Mon 09-Dec-13 02:15 PM by Chris Platt
This topic is becoming as controversial as the topic of protective filters with adherents on both sides of the issue, and nobody is wrong.
With the exception of tripod mounting, I don't know of any manufacturer that provides guidance on turning the VR on or off at shutter speeds above 1/500 - they are silent on the issue, so we are left to speculate.
The speculation is based on the sampling rate said by Nikon to be 1000 times per second. At speeds above 1/500th the system, needing two samples to calculate a correction and shift the lens elements, can't respond quickly enough to make a correction - so VR is presumed to provide no value above 1/500th. But can it actually have a negative impact? I haven't seen any evidence of negative impacts either in field use or in tests I've done to look for problems.
So - is there a negative impact? I doubt it and I haven't seen any evidence of it either in my shots or others. Yes you can detect very slight variations from shot to shot - very slight - that you would see whether you had VR on or off in my opinion.
Pros above 1/500?
Steadying the view finder image for tracking and focusing
Cons above 1/500?
Accelerated battery depletion with no benefit to the image.
Recommendation? Whatever you're comfortable with. I tend to leave VR on, not because I think it is helping, but because it isn't in my mental checklist to turn it off at speeds above 1/500. I usually shoot in aperture priority and don't want to concern myself with clicking it on an off as I move back and forth across the 1/500 line.
I edited the reply above because I failed to consider how the shutter actually operates at exposure times above 1/500th and speculated that VR wouldn't even attempt to shift above 1/1000. That speculation failed to consider that exposure time is not the same as the time required for the shutter to complete it's travel, i.e. that it takes about 1/250 for the shutter to complete travel even at exposure times above 1/500. So there will always be as many as 4 shifts during shutter operation regardless of the exposure time? But will those shifts have a negative impact? As I said above, I haven't seen any.
Thom Hogan had done tests and wrote an article about it. I think he also had some sample images but I don't recall for sure. The essence was that VR does degrade the image at high shutter speeds, albeit slightly.
Now the topic the OP brought up - using VR on the tripod - that is a different issue. If the lens is steady, no movements, as can be the case on a tripod, then the VR system will struggle with that, as it is designed to counteract movement.
For my personal use, I typically shoot flying birds and other moving wildlife at about 1/1250s, to freeze any subject movement. At those speeds I always turn VR off. The issue with VR in that scenario is the extra second or two it takes to stabilize the image, from the time I press AF-ON (or half-press the shutter). I may not have that second when following a bird. That is the main downside for me, not the battery depletion!
>Thom Hogan had done tests and wrote an article about it. I >think he also had some sample images but I don't recall for >sure. The essence was that VR does degrade the image at high >shutter speeds, albeit slightly.
I thought there was a thread on Nikonians as well a couple years ago where someone posted photos demonstrating this phenomenon - I did a quick search and I couldn't find it though so I might be thinking of somewhere else. Personally I have left VR on at higher shutter speeds and not noticed anything - but as it appears to potentially have an impact in some situations, best practices may be to turn it off. I always try and remember to turn it off now at speeds over 1/500 but YMMV.
I read Thom's article a couple of years ago. I don't believe he ever posted images demonstrating what he was talking about - at least not with the article. Much of his commentary was also speculation IMO - not that I'm taking issue with it - I just haven't observed the effects he predicted with the possible exception of bokeh. I believe some of the harsh bokeh I've observed occasionally with my 50-500 OS may be attributable to OS, but I don't believe that is shutter speed related - just the effects that shifting lens elements may have on anything outside the plane of focus.
We hashed this topic over pretty well in this thread which has some test images I posted: