Technical aspects of the AFS 300mm f/4 E PF lens

[Ilkka Nissilä]

The D500 is a DX camera, with smaller mirror and shutter. In a pro DX camera the mirror box is larger relative to the shutter window thus the shutter can be designed to accelerate slower (more room for the curtains to pick up speed). Thus the effect of the shutter vibration is smaller and high frequency components, which the VR system cannot handle, are less pronounced. If a grip is attached to increase weight this attenuates the high frequencies which is probably why some find it helps with the 300/4 PF (extra weight in the body in general to reduce vibrations at slow shutter speeds, not only with this lens).

The VR system probably can't deal with the high frequencies that the shutter causes with the gripless FX body and such a light weight high magnification lens. Also the way the lens is supported varies from user to user. I think it is not at all surprising that there is no easy solution. If there was one, I am sure they would have given to us with the lens and body FW updates. However, recalling the lens a second time should be done in such a way that it is guaranteed to be successful. It must cost Nikon hugely to issue such recalls. They reported a 25% increase in budget reserved for warranty repair reportedly. 25% increase may have been a result of D750, 300 PF etc fixes. I would imagine the D750 in particular must have been very expensive for them. And the D600 and D800 in the past but I'm giving them some slack since they did lose their factories and the society infrastructure due to the natural disasters in 2011 in particular. However the 300 PF problems have nothing to do with that I am sure.

If there are body sample and lens sample dependent variations in VR performance with the same FW installed, then it is very curious. Component weight differences should be very small and I would imagine the variation in material strength to be very small as well. Electrical components may have slightly different characteristics (analog components). But they should have been able to account for the effects of such variations in the design. Assembly?

[bjornthun]

A Nikon D500 probably has a smaller mirror than an FX format Nikon, which also contributes to less vibrations in a D500.

The D500 was released a year after the 300/4PF, so the design of the D500 may account for the 300/4PF at the outset.

[Erik Lund]

Chris, I'm not stating it's just the weight - It's of course much more complex than that!

And most likely it's also a combination of different little details that make a difference,,, MY advice is; Relax and let the Nikon techs do their thing.

[chris dees]

Chris, I'm not stating it's just the weight - It's of course much more complex than that!

And most likely it's also a combination of different little details that make a difference,,, MY advice is; Relax and let the Nikon techs do their thing.

I'll wait (already a year).

[simsurace]

Well stated Illka - On a general note, please just enjoy your cameras and lenses - Learn how to work around any issues or return/sell the equipment if you don't like how they work... I have a 300mm f/4 PF it does do nice images at any speed on D810. Even with TC's

On the specific issue with VR - My guess is that the cameras in question have too little mass or a mass at a critical low value - The issue could be the VR is designed around more or less stable camera (heavy camera) and lens going/swinging up/down/sideways - With a light weight camera/lens the whole thing is moving lens and camera, sometimes camera more than lens,,,

Think how the VR lens element compensates, how it actually moves when for instance panning, the element should/can only compensate for up/down shake of the lens and not the rotation of lens/camera... unless it's set for Action VR then it tries to compensate for both camera and lens movements...

My D600 does not show the problem, or at least much less, than my D750. Their weight is almost identical.
Some users experience issues on the D810 even with the firmware fix.
Our experiments suggest that not only are there good/bad samples, but good/bad combinations, so the problem is quite complicated it seems.

[simsurace]

One more thing: I also tested with MUP and exposure delay. Both did not remove the blur issues. So I don't think the blur can be attributed to the mirror slap; if anything, it is the shutter-induced vibration, or simply a bug in the control algorithm (which I think is much more likely; the algorithms there are quite complex, a lot can go wrong). Anyways, I don't know whether the VR is designed to work with MUP and exposure delay. Since the viewfinder is black for an extended period, it is hard to keep the subject centered, so this is not really a practical solution either.

Martin suggested that it might be an issue related to power supply. So the battery grip might just help because it provides a more steady voltage to the VR motors.

Could someone (maybe Chris) with a battery grip test this? E.g. tape the contacts of the battery grip batteries to see whether it is the mass of the battery grip that leads to less blur, or the power supply coming from the battery grip? I don't possess any battery grips.

[Akira]

Does the direction of the blur changes from vertical to horizontal, when you hold the camera in the portrait orientation?  With the mirror lockup, of course.  If so, you could be sure that it is shutter related problem.

[simsurace]

Thanks, Akira.
The blur is always more or less in the up-down direction in a horizontal image. But I haven't tested with the exposure delay/MUP.

[bjornthun]

Is it possible to test it with mirror-up and electronic first curtain shutter (EFCS) enabled at the same time? This is one more way to get a hint, if the shutter movement is causing the problem. Afaik, the D810 has got EFCS.

[Øivind Tøien]

Lots of activity in this thread; out on work travel I have not been able to follow up.
Now to the question: Did anyone on the trip test with/without a tripod collar attached? I cannot see any mention of this.

[Erik Lund]

My D600 does not show the problem, or at least much less, than my D750. Their weight is almost identical.
Some users experience issues on the D810 even with the firmware fix.
Our experiments suggest that not only are there good/bad samples, but good/bad combinations, so the problem is quite complicated it seems.

I have heard and read that.

I still believe my reasoning is correct.

[simsurace]

Lots of activity in this thread; out on work travel I have not been able to follow up.
Now to the question: Did anyone on the trip test with/without a tripod collar attached? I cannot see any mention of this.

I see no difference when using the collar vs. no collar attached. But I'm not using the Nikon collar, but a rather inexpensive one off of ebay I got for the 70-200/4. It does a good job as a tripod collar, but not to dampen the vibrations that are causing the blur when handholding.

Does the direction of the blur changes from vertical to horizontal, when you hold the camera in the portrait orientation?  With the mirror lockup, of course.  If so, you could be sure that it is shutter related problem.

I just tested with one second exposure delay (to avoid having to press the shutter release twice). The blur direction relative to the camera is conserved, i.e. it is up-down for the horizontal image (image 1 below) and left-right for the vertical image (image 2 below). Look at the bright dots just below the edge in the shade.

I'm not a 100% sure about your reasoning though. It could also be that the VR unit has a defect that manifests itself in a up-down blur (relative to the camera). When turning the camera to portrait orientation, the lens turns also. It would be less likely to have a defect which is always up-down irrespective of the camera orientation, unless it is gravity-related. But these are all wild speculations anyway, the Nikon engineers will hopefully find the real problem.

[chambeshi]

Chris, I'm not stating it's just the weight - It's of course much more complex than that!

And most likely it's also a combination of different little details that make a difference,,, MY advice is; Relax and let the Nikon techs do their thing.

Interesting observations. Obviously, it is Nikon's responsibility to solve this problem(s), but i suggest it will be enlightening run off coordinated empirical tests. Although the logistics appear  challenging to get together enough copies of the offending lens in this case and  say 4-5 bodies (Df, D750, D800, D810 and DX's etc). The ideal would be to get together several 300 f4 lenses in the 2 categories (i.e well behaved/ no issues Versus Shaky) and test them on the same standardized images (this test sheet can be circulated electronically). With standard settings (e.g 1/30 >>> 1/1000) with VR On vs Off). Cameras with mirrors up vs usual function, and tripod collar on vs off, and of course on different DX and FX bodies.

The internet can take this process a long way forward. Many of us are positioned to run off such tests using a std test image that we can pool then the images to a willing analyst.....Suitably one or two Nikon owners who have bought a troublesome copy of this 300 PF can then summarize and bring the results to Nikon's attention.

At the moment I don't own a 300 f4 PF [been put off so far!] and only have a D7200 - Df scheduled for later this year :-) but in my case, I can ask around in the Cape for willing collaborators. Indeed I know at least willing 3 friends, and we could arrange to test together with a D7100 and D5100 etc.

Basically this is doing heavy lifting on behalf of Nikon,  but purely a suggestion. Where some Clients have been waiting for too long - and there is yet to be informative Feedback from Nikon - such a coordinated test will provide fairly statistically solid empirical evidence....

[Erik Lund]

Is there any report of issues on a D3 D4 or D5?,,,

[Ilkka Nissilä]

If you plan on carrying out a set of tests using multiple lens and camera samples, you also need to rotate the photographers taking the shots since photographer variability is likely to be much greater than lens sample variability in hand held shooting. And the number of shots made with each combination of body+lens+photographer+shutter speed should be at least 20 to get meaningful data and to reduce the effect of randomness in the hand holding on the mean and other statistics. The image should be focused using live view and the distance to target such that the shake of the lens forward and back does not affect image sharpness appreciably. The photographer should be unaware of which sample of lens they are using. This is not an easy thing to do and there are many ways to do it wrong.

If Nikon can make EFCS work without M-UP mode, and put it in more cameras, this problem would be solved for all lenses, not only the 300mm PF.