Nikon 300mm f/4E PF ED VR artifacts

Just bought this lens, works fine except I'm getting a very strange artifact in certain light (not the well-documented problem when shooting into the sun). The problem is a series of slightly arcing waves in the background of certain images. The problem is illustrated in the attached photo (Nikon D7100, raw, minor post processing in Lightroom). I've always used this lens is used with a Nikon 1.4TC (TC-14E). The problem is the same shooting wide open or at a smaller aperture. I'm not sure of the VR setting when I took these pics, I generally turn it off when shooting faster than a 500th, and as I have images shot both faster and slower than that with the same problem, I assume this is not VR related.

 

This only occurs in specific light -- I've shot a few hundred pics so far as this example is the worst. Note the attached image is cropped, the arcing behavior of the distortion is more apparent when the entire frame is observed.

 

Is this a known problem, or might there be a problem with my lens or setup? To my uneducated eye, looks like it could be a fresnel-related issue.

 

Thanks, Jim

 

PS If any birders out there, the birds are relatively uncommon00 red-billed pigeons, Rio Grande Valley, TX.

 

THis is a known effect whith strong backlight subjects , it is caused by the use of a fresnell element in the lens construction .

A comparable effect can be seen when taking pictures at night and having street lighting in the frame. The effect is stronger for light which contains more parts in the red spectrum .

It is not a lens fault, it is a "feature" of fresnell lenses which cannot be avoided totally .

 

More on this in the following article :

 

http://cameralabs.com/reviews/Nikon_Nikkor_AF-S_300mm_f4E_PF_ED_VR/index.shtml

 

 

"But there is a caveat (quote from Nikon's website): "[...] when there is a strong light source within the frame or when light enters the lens from outside of the frame, ring-shaped colored flare may occur according to shooting conditions".

 

And here :

https://petapixel.com/2015/01/10/closer-look-nikons-new-phase-fresnel-pf-lens-technology/

Edited February 17, 2017 by c.p.m._van_het_kaar

Thanks. I find it pretty annoying.

Not having a PF lens, I can't speak from experience, but that particular behaviour did look oddly directional rather than radial (which is how I'd expect PF artifacts to behave). I wondered about VR (or camera shake), but it does seem unlikely for a fast shutter speed. Just noting the oddity, not disputing the conclusion.

The out of focus rendering of the 300 PF isn't the best, but it's not terrible, either, just sometimes a little odd. I guess you are likely to run into artifacts with many lenses if there are branches in the background. Trying with VR OFF may be a good idea, and use a fast shutter speed.

 

Personally I don't usually turn off VR with this lens for fast speeds, I shoot with VR in SPORT mode most of the time and turn the VR OFF when using a tripod. I haven't noticed any problems at fast speeds with this VR mode; there may be some but it hasn't bothered me, and I find it easier to focus with the AF sensor squarely on the main subject when the lens is stabilized. There may be slight loss of sharpness with VR on in some cases but I generally find the steadying effect to be more important for getting a higher percentage of precisely focused shots. I do occasionally try VR OFF to see if it is working for me or against me. If you get a particular artifact, it makes sense to try without VR to see if it helps alleviate the problem, or not. I think the main issue is likely the optical design of the lens (use of the PF element). If you want the finest out of focus rendering at 300mm focal length, go to one of the 300/2.8's (used copes may be available for a reasonable price). However, the 300/2.8 of course doesn't have the supreme convenience of the light weight 300/4 PF. I don't have the 300/2.8 myself, as I am sort of waiting for the FL version to appear as it is likely to be lighter.

To be clear, this effect was the same with shutter speeds up to 1600th. To Andrew's point, it is only slightly radial. For me this moire-like bokeh is very unsettling.

 

Ilkka, tx for the VR on/off comment, I turn it off at fast shutter speeds only in response to Thom Hogan, et al, comments regarding VR at > 500th, but I'll try both.

Turning VR OFF may help in some cases with gaining the last bit of sharpness, for me it is just that I think the better control over composition and focus point stability are usually more important than the slight loss of sharpness that VR may cause at fast speeds (which is usually not very easy to see). I keep VR in SPORT mode at least up to 1/800s and have felt it beneficial to my focus keeper results with this lens. For faster speeds than 1/800s, if I know I will stay there and have no reason to shoot at lower speed (i.e. if there is plenty of light) then I may consider turning VR OFF but it's such a hassle to remember to play with the switch, I don't worry about it too much. I usually know what the print size is which I am aiming for and since I normally use all the frame or make a slight crop only, I am not too worried about a slight loss of sharpness if it is to appear because of use of VR. I guess it involves a compromise between the highest percentage of focus keepers (VR SPORT) and the highest sharpness of the best shots (possibly VR OFF). Perhaps Nikon can one day implement a mode where VR SPORT is available during focusing and composition and VR is OFF during the exposure. But I guess VR OFF involves a recentering of the VR element to a fixed position which would potentially throw composition off from what it was when VR was ON during composition and focusing.

Yes, I was only really wondering whether the apparent division of single subjects into two (looking at some branches in the background) might resemble some behaviour I've seen with camera shake, where most of the exposure happens in two stable positions with a jump between them (as in a mirror slap). If the VR tried to do something clever, like deal with the lens moving in one direction, keep the focal point in the same place (by rotating the field of view), but have the background offset by the same amount that the camera moved, I'd kind of expect to see something like this, But probably not at a high shutter speed, and not consistently. I've (accurately) looked like an idiot before when trying to analyse bokeh behaviour, so I'm not saying that what's happening here isn't caused by the PF element, I'm just curious about the behaviour and trying to get my head around what's going on.

What is looks like to me is the result of the PF element combined with camerashake though. Wondering : was this on a tripod while using a remote control of some sorts ? Otherwise 300mm with a 1.5 crop factor is a lot to hold still for a lot of ppl ( including me..) , more so because this 300mm lens is relatively lightweight compared with the older types 300mm F4, so the mass of the glass does not help slow down camera movement ..

Read the above and more from other sources, but this problem does not seem to pertain to the sample image, as there is no backlit or lighted items, especially no ring-shaped flares on the image. Actually I am most disturbed by how unsharp the image is with the focus points on the birds @800/s.

 

<br><br>Nothing in particular seems to be in focus. Maybe the problem is elsewhere as far this image is concerned.

Edited February 26, 2017 by Mary Doo

In my experience, the 300 PF does well when there is high contrast, colourful detail in the main subject, and/or the lighting contrast is good, especially if the subject is backlit. The optics create a bit bland image if the lighting and subject are not colourful/contrasty. I would not be overly concerned about technical issues in this image, but try to seek better light to take advantage of the lens's properties.

Read the above and more from other sources, but this problem does not seem to pertain to the sample image, as there is no backlit or lighted items, especially no ring-shaped flares on the image. Actually I am most disturbed by how unsharp the image is with the focus points on the birds @800/s.

 

<br><br>Nothing in particular seems to be in focus. Maybe the problem is elsewhere as far this image is concerned.

Ok, maybe something is off with this lens then.

Maybe do some tests both with normal focussing (from a tripod) and focussing through liveview , and compare the results ..

Not sure if you can adjust focussing on a D7100 , but maybe you have a heavy case of front or back focus on this one ?

Jim, if every shot is like that I'd get Nikon service to look at that lens. The directional waviness might indicate that the PF or some other element is de-centred. Lenses should always show axial symmetry, and yours appears not to. Even the parts that are in focus don't look too sharp either, but it's hard to tell from a single tiny sample image. Edited February 28, 2017 by rodeo_joe|1

Thanks for the comments. As mentioned in my original post I've taken hundreds of pics with this lens and this was the worst example, many of the images I've taken with this lens are sharp and exhibit no bizarre bokeh. Looking over the images I've taken, the effect is magnified by specific backlight, the lack of focus probably due to the lack of sharp contrast.

 

An example of a better image with the same setup is attached.

 

If you look carefully in the background above the bird's (Kingfisher?) head, you can see double imaging in the OOF highlights and foliage. Is this normal for the PF lens?

The bird is a Green Heron. The bokeh outlining isn't what I've seen with my 300 except in the critical range of shutter speeds where some of these lenses behave less than optimum if VR is active (although that's supposedly limited to D8x0 bodies). Was VR on? What's the shutter speed? With TC?

 

To me, the image does not look particularly sharp. Mine had to go back to Nikon to fix the VR behavior despite the SN being outside the range where he should have been necessary.

Edited March 13, 2017 by Dieter Schaefer

The heron pic was shot at 1/250 f/13 with the 1.4TC, VR was on. However the bokeh issue exists both when VR is on or off, and at shutter speeds greater than 1/1000.

 

Sounds like I should send it in for repair.

Sounds like I should send it in for repair.

I certainly would.

 

Here is a 100% crop of an image taken with the D810 and the 300/4E PF VR, 1/640s, f/4.5, ISO 560, VR off; not sharpened in processing, processed with latest version of photoshop CC, all parameters left at default. I see none of the outlining that's in your image. Person was in the shade but the scene is backlit.

Jim. You might want to eliminate the TC from the equation before sending the 300 off for repair.

I like the Green Heron image. I shot this lens quite a bit on a wildlife trip and was very pleased with the results. Neither I nor anyone else noticed any fresnel artifacts.

I certainly would.

 

Here is a 100% crop of an image taken with the D810 and the 300/4E PF VR, 1/640s, f/4.5, ISO 560, VR off; not sharpened in processing, processed with latest version of photoshop CC, all parameters left at default. I see none of the outlining that's in your image. Person was in the shade but the scene is backlit.

 

No circular artifacts in this top-hat photo, despite the backlighting.

 

Bjorn Rorslett did some tests, said the fresnel artifacts are hard to find, and posted an example shot (link)