Nikon VR, does it really work?

Arkady said:
Copied from Nikon Website

What are the differences between the two VR modes of the Nikon 80-400mm VR lens?

In mode 1, VR system starts when the cameras shutter release button is lightly pressed and allows the user to see that any camera shake is reduced by looking through the viewfinder. Of course, VR operation carries on during exposure

In mode 2, to conserve battery power VR operation is only executed during taking of the image. You cannot see the VR operation through the viewfinder in this mode. This mode is designed for those who feel uncomfortable with compensated viewfinder images caused by the vibration reduction mechanism and conserves battery power. Actual photo results are the same as for mode 1.
Click to expand...
I always smile when I see this. The VR implementation on the 80-400 is eccentric, to say the least. It was probably Nikon's first VR lens and they probably hadn't worked out the practicalities. Mode 2 is a complete waste of space, though I guess it might have sounded clever at the time. And I love the way Nikon have tried to rationalise it - "designed for those who feel uncomfortable with compensated viewfinder images". Yeah, right. I'm sure there must be loads of people out there who prefer a bit of visible shake. :shrug:
 
puddleduck said:
But this doesn't surprise me - at these shutter speeds its clear that VR can only negatively impact IQ as you have an element moving around that doesn't need to be moving.
Click to expand...

Can you (or anyone else) explain this then?

puddleduck said:
I rarely use VR. Certainly on my 70-200 VR it was always off - not only does it impact sharpness, it also impacts bokeh.
Click to expand...

Ok, so you seem to agree with my statement back on page 1 that it sucks, which is nice, but can you give us any idea why this would be the case?
 
Well I wouldn't say "it sucks" but its clear that it needs to be turned off when not needed.

Nikon says 2 stops for VR on the 70-200 VR so I'd suggest they mean that once shutter exceeds 2 x focal length (maybe 2.5x for crop) its time to turn it off.

Have a read here:

http://www.naturfotograf.com/index2.html

Clearly seen also is that VR mode does have a bearing on bokeh. The more active the VR, the less attractive the bokeh...

I imagine that once shutter speeds get so high, VR can no longer compensate, and indeed degrades the image. No surprise.
 
Why would VR have anything to do with shutter speed though? As far as I know it doesn't take that into account at any time.. its just accelerometers detecting unwanted lens movement and shifting a moveable element to account for it.

And I don't get it about the bokeh either!

I agree 100% that VR should not be needed at faster shutter speeds, but equally, why would it fail?
 
I can't answer "why" - I can only report by observation. I don't think the "Why" actually matters tbh.

I imagine if Nikon wanted to, it would be easy to disable VR via "D" information vs. shutter speed. But then they don't tie in AutoISO to lens focal length vs shutter speed, which for a firmware engineeer is a 3 minute job to code.

In the meantime, to maximise IQ, we need to turn off VR while on a tripod, and also turn it off when it exceeds around 3x focal length / shutter speed ratio.
 
Googling around, there is an interesting thread here:

http://community.compuserve.com/n/pfx/forum.aspx?tsn=21&nav=messages&webtag=ws-photography&tid=32830

Reading that the conclusion seems to be:

1) That VR is only useful below 1/60th (although I'd have said that was dependent on focal length...)

2) At "high" shutter speeds VR degrades the image (no idea what "high" means, but presumably around/just above the reciprical ratio)

3) Canon's IS does not exhibit this issue

All of that probably describes exactly what I was talking about in my very first post. I'd never have thought to have turned off VR for my 1/160 - 1/320 motorsport shots with my 70-200VR, why would I when my old Canon gear was more than capable of doing it and yielding positive effects....

I therefore re-conclude, VR sucks, IS works :D

I'll do some Canon tests to further validate this later this week....
 
puddleduck said:
In the meantime, to maximise IQ, we need to turn off VR while on a tripod, and also turn it off when it exceeds around 3x focal length / shutter speed ratio.
Click to expand...

I agree about VR off when using a tripod, the VR lens manuals state that anyway, but turning it off at other times? well thats really only of benefit if we find it affects the images that we alone produce. I for one don't see a detrimental effect on my images at high shutter speeds with VR on, certainly not on any of my VR lenses and to suggest everyone follows this rule of turning VR off without question is wrong.
I think this is a case of personal experimentation and choice, the best answer to the original question is suck it and see. If it works for you, use it, if not, turn it off.
I notice that there is still a lack of images to properly evidence that IQ suffers with VR turned on at high shutter speeds :shrug:

Allan
 
allanm said:
I notice that there is still a lack of images to properly evidence that IQ suffers with VR turned on at high shutter speeds :shrug:
Click to expand...

Yeah I still haven't done that bit of the experiment yet - I'll try to get round to it. However, if you google something like "vr high shutter speeds problem" you do find pages of people saying just that.

I'd have found it quicker had I thought this might have been the case, but it wasn't exactly leaping to the top of the pile as the reason for my dissatisfaction with VR...

I will do some tests though (and some Canon ones just to be complete)
 
I did a test on Sunday and I can see the degradation on a couple of frames (not all of them)- I'm sure the "evidence" won't be acceptable (because its impossible to prove VR was the cause) but in just a few instances, around 1/2000 when really shots should be pin sharp there is an unaccountable loss of sharpness when VR is on. Again this is widely reported, and well known in Nikon circles.
 
My very quick google on vr problems came up with nothing obvious. Can you post any relevant links?

I've just done a test with two Canon IS lenses, 70-200 4 L IS, and 100-400L IS. IS works superbly on both lenses at all times. The former has Canon's latest 4-stops IS (and it costs a freakin fortune more than the non-IS version) and the 100-400 has one of Canon's earliest 2-stops IS incarnations. Camera is 40D. Everything at 200mm.

I shot a bare light bulb from across the room and exposed for the writing on it - only way I can get really high shutter speeds ATM. At 1/4000sec, no difference with IS on or off, ditto 1/2000sec, and 1/1000sec, with either lens.

To see how the IS worked at the other end of the scale, with IS off I was getting sharp images with the 70-200 down to 1/125sec with acceptable results and a decent hit rate (I was leaning against a wall). With IS on, everything was sharp down to 1/30sec, but I didn't try any longer speeds. What was also noticeable was how hard it was to maintain the centre AF point exactly on target with IS off - it dances about too much.

I'm still finding it hard to believe that Nikon can't make as good an IS system as Canon, especially since they appear to have been doing it longer according to Wikipedia :thinking:
 
shakey c said:
Only tested handheld, you can tell the image has moved its in a slightly different place when the mirror returns.
Click to expand...

This line worrys me HOW are you testing it?
It stop small movmet not you moving the camera as you take it?
 
HoppyUK said:
My very quick google on vr problems came up with nothing obvious.
Click to expand...

C'mon, I gave you the keywords further up the thread...

Your tests... how many shots did you take and what was the success rate - like I demonstrated further back up this thread.... just to keep it scientific. Nothing strange about your results btw, you have proven exactly what my several years of experience with Canon have shown - IS does work.

But thanks, one less thing for me to test to get you doubting Thomas's seeing the light :D
 
HoppyUK said:
My very quick google on vr problems came up with nothing obvious. Can you post any relevant links?
Click to expand...

Same here, just speculation in a couple of articles that it could cause softness of images under certain conditions but no evidenced articles anywhere that it actually does. It certainly doesnt seem to be a major problem in any forums I have searched through.

I think panning with high shutter speeds using VR may cause softness as the VR system wouldnt have time to stabilise before firing off the shot but I think this is more a user or usage problem than a fault with VR in general.
Allan
 
puddleduck said:
it doesn't take a rocket scientist genius type to figure that the in-focus areas are going to be affected as well...
Click to expand...

I couldnt find the reference to Bokeh in your link, can you be more specific?

In the post you linked to you said you "imagine that once shutter speeds get so high, VR can no longer compensate, and indeed degrades the image. No surprise."
Well, in my case, I just dont find that is true on any of the comparison images I have produced. I think if we are going to answer the OP question, we need to look at facts and real life situations so the OP can make up his own mind whether VR is worth getting
Allan
 
puddleduck said:
What about the evidence regarding bokeh? (http://www.talkphotography.co.uk/forums/showpost.php?p=1695202&postcount=84) - if VR can have an influence out-of-focus areas, it doesn't take a rocket scientist genius type to figure that the in-focus areas are going to be affected as well...
Click to expand...

Yes, I understand how this technology works, and what we're looking for, and frankly I still find it amazing that it works anything like as well as it does. But it does for me.

I also understand how and why it could affect sharpness at higher shutter speeds, or indeed longer ones, too. The theory is that there are moving elements in the light path at all times, so it should potentially show up at all times. In which case the best way to show it would be on a firmly tripod mounted camera, to actually eliminate the camera shake aspect, but we know this has associated problems.

Having said that, my 70-200 manual makes no reference to this, other than saying to turn it off to conserve battery power. The 100-400 is less clear, but I do use it on with a tripod with IS on (I know others that do the same) and it is very beneficial, however, it occasionaly throws a wobbly so you have to watch it (not hard to spot).

It would maybe be relevant to find out at what frequency/frequencies it moves?

BTW, for the tests I did above I must have taken almost a couple of hundred images, in sequences of six for each setting.

Edit: away from PC now. Won't be back today :( :D
 
This thing is Hoppy, this question is about VR, not IS..... I don't think what your 70-200 IS does or does not do is really relevant. AFAIK there is not a well documented high shutter speed issue with Canon IS...?
 
allanm said:
we need to look at facts and real life situations so the OP can make up his own mind whether VR is worth getting
Click to expand...

You don't get much choice really. If you want any of Nikon's newer/better lenses then you get VR whether you like it or not.

Its only Canon who still make a stabalised and non-stabalised version of some of their lenses.
 
StewartR said:
And I love the way Nikon have tried to rationalise it - "designed for those who feel uncomfortable with compensated viewfinder images". Yeah, right. I'm sure there must be loads of people out there who prefer a bit of visible shake. :shrug:
Click to expand...
there are some people that feel physically ill (like motion sickness) with stabilised viewfinders.
 
puddleduck said:
This thing is Hoppy, this question is about VR, not IS..... I don't think what your 70-200 IS does or does not do is really relevant. AFAIK there is not a well documented high shutter speed issue with Canon IS...?
Click to expand...

I like to check these things for myself, and since Nikon VR works in exactly the same way as Canon IS, it follows that what happens with one system is likely to happen with the other. So there it is. I didn't mean to waste anybody's time.
 
it works, i never use 300mm zooms for dark concerts
last night i thought i would practice before i go to cornbury to shoot in daylight.
my hands are a bit shaky which dosent help

050c372c.jpg



2396a4ee.jpg
 
You must log in or register to reply here.
Back
Top