Defects in D800

[AutumnCurl]

I got mine last night - the serial starts in the 8008XXX it has come with the new version of firmware on it. so far all seems ok, but i will endeavour to do some testing today with all my lenses. so fingers crossed its all good

I hope the turn around time for you Mongo is a fast one.

[rellik666]

As for singling out Nikon as being any better or any worse in respect to "new model problems" than any other maker I still reckon they are ahead by a short nose on the right side of the post to most.

Other manufacturers may have problems like this also from time to time but they do not charge the sort of money Nikon does - so Nikon have a far bigger responsibility to get it right. It is not good enough to for Nikon or for us to think that because it happens to others its OK to happen to one of the most expensive brands as often or as much. That's just what Mongo thinks.

You should try buying a certain brand new car when they bring out models and let the customers be the testing regime!

Nikon aren't the first and won't be the last to rush a product out to only update it as problems arise!

Frustrating I know, but we do push for these things to be out sooner rather than later!

[mongo]

A thanks and quick reply to those last comments made by most of you.


Jim, the answer to your question varies but you need to show it is a lemon either incapable of being fixed by the manufacturer or has been attempted to be repaired so many times that it is unreasonable to have it fixed again. This would no doubt help establish it is not of merchantable quality or fit for the purpose.


Wayne, it is true there are some slightly dearer than the nikon D800 but my point was nikon has SLRs that cost nearly $10,000 and if it wants to be in that category of manufacturer, it needs to demonstrate it makes products of a superior quality . Canon does not even come close to this sort of expense for its top of the line.


Rick, even nikon have advised Mongo against doing the firmware upgrade himself and have insisted they are happy to do it for free when calibrating the AF sensors. So, Mongo is going to let them do it all and therefore remain responsible for it all.


Sar, Mongo is very glad you missed out on this problem but make sure you get the firmware update. It will apparently stop the freezing problem

[I @ M]

but i will endeavour to do some testing today with all my lenses.

Testing or observing the l/h AF point issue is not rocket science and fails to even reach pixel peeping levels.

As a clear indication of how it is observable simply set up 3 objects with the same area under the centre, right and left AF sensor points.
Have the camera level on a tripod close to on a flat plane with the targets.

With 51 point AF points set.
Focus, defocus and refocus on the centre object with the centre AF point and take a picture, move the toggle pad to select the far r/h AF point, focus, defocus and refocus and take a picture. Repeat step #2 with the far l/h AF point selected.

These images show it fairly clearly.

Centre AF point selected.


and cropped.


R/H AF point selected.


and cropped.


L/H AF point selected.


and cropped.

[Sar NOP]

Andrew, so yours has the same problem with the l/h AF point (according to these images test).

[I @ M]

Andrew, so yours has the same problem with the l/h AF point (according to these images test).

Yes Sar, on two bodies with serial numbers 3 units apart.

[Wayne]

A good test as well is to do a similar setup to Andrew, and also use live view, so that contrast detect AF is used directly from the sensor. Compare that to the images made through the viewfinder using phase detect AF, it will quickly show up the focus screen anomalies. Are you guys using CNX2 do show the focus point on your images? I don't have it installed, but will for the purpose of testing, just to reference later.

[I @ M]

The above post of mine relates to checking the focus accuracy of the camera via the viewfinder / mirror / focus screen as that seems to be where the problems lie with most cameras, live view is a different fettle of kish and is not to be relied on as a definitive check of conditions.
Wayne, re the focus point showing, Nikon View NX2 will show the focus point in the full size images but for clarification with those shots the screen cap below shows where the focus point was achieved in these shots. In all 3 cases the point is on almost identical areas of the face in the match box.
Sorry about the red lines against the red box, was the last thing I was thinking about when I did it.

[Lance B]

I have been reading reports of the focus issue on the various forums ever since the D800 went on sale and so I am well aware of the issue and how it occurs. Generally, it occurs on the very wide aperture, or fast lenses, like the 24 f1.4G, 35 f1.4G, 50 f1.4G, 85 f1.4G and the misfocus is most prevelent with the left outer AF point manifesting itself as back focus and can also occur as slight back focus on the right outer most AF point. Basically, all the other AF points are spot on for accuracy.

I have checked my D800 using all my lenses and they all have perfect focus using all the various focus points. I have taken about 4,500 images so far and as you can imagine, I use just about every AF focus point in my bird photography and I have never missed a shot due to incorrect focus. I am consistantly using wider apertures as well, so AF inaccuracies would not be masked by DOF. Also, using all my other lenses (see below), the AF seems spot on and again, I am using all the various AF focus points.

However, when using the 85 f1.4G I occassionally get misfocus when using the left AF point and very wide apertures like f2, it is not consistant and most times it is spot on.

I have been in touch with Nikon service and I am waiting for them to advise of whether they know there is a fix. At the moment, the AF issue has not affected any of my photos as it only occurs on the one lens and I can just work around it, so, I am in no real hurry to do anyhting about it at this stage. I would be interested in whether yours is actually fixed or not.

[mongo]

sorry if Mongo misunderstands what you are saying Andrew. Are you saying your cameras exhibit the same problem Mongo is having but that you do not think it significant enough to worry about ???

Maybe Mongo cannot see it as clearly as others can but on the examples you first posted, look roughly the same to Mongo - except that in the last image of that set, the middle matchbox looked less in focus than the left side matchbox and that the left was roughly the same as the right side i.e left and right were roughly the same while the middle was less in focus.

Whichever way Mongo has read you examples, given the very obvious and visible results from Mongo's tests of his camera (particularly at such a short distance with a 200mm supposedly super sharp lens), there would be no way that Mongo would not insist it be fixed or get his money back.

Mongo dropped it off to Nikon today and has been assured it will come back super fixed in all respects. Will let you know Lance and others.

One quick note to those of you who may have dealt with Nikon in Sydney, Steve has been there for years and from Mongo's dealing with him on a number of occasions, Mongo can honestly say he is Nikon's best asset ! Mongo has found him knowledgeable, interested, prepared to listen and do whatever is possible to assist Nikon's customers. A big thanks to Steve.

[I @ M]

Mongo, you are misunderstanding.

Yes, the problem is real.
As I stated originally, the l/h AF point simply isn't that.
It doesn't work, it is useless.
No it is not acceptable.
Yes I will, along with many others be sending the bodies back to have them fixed.

As for the images I posted, the centre AF point and r/h AF point shots are good ( not perfect ) as far as focus goes, the l/h AF point shot is woeful.
There could be some advantage gained from an AF fine tune being done on the centre point image but until the body focusses consistently across all points it is not worth doing.
The issue is not confined to any one lens, I can replicate it with many different lenses and focal lengths. The posted images are taken at F/2.8 with an 85mm F/1.4 lens. I can produce much the same images with a 200mm or 300mm lens at F/4. I can produce the same at 16mm at F/5.6.

Lance, your observations on "back focus" simply don't sit well with me. As far as I know if you are getting a "good" focus on the centre point and substandard performance on both the outer focus points, then the lens either has plenty of field curvature attached to it or is faulty. In the images presented above, the centre point and r/h shot show pretty normal evenness across the image, if there were lens factors at play we would see problems in all images. It is only when the outer l/h AF point is used that we see ugly looks across the entire image. Remember, we are only talking about the area under the focus points which with the Nikon FX bodies is still quite a long way short of the full image so edge softness in lenses etc hardly has any impact.
The most obvious thing to look at with this problem would be lens mount deformity except that we are seeing repeated examples of the problem occurring only at the l/h AF point and no softness at the edges using the centre point.

[Ezookiel]

It's not just Nikon that can have problems. My Canon had focus issues and had to be sent back for some tweaking.
Hope it gets sorted, and that it doesn't take too long. I was told it would be up to 6 weeks to get the Canon back, but it was less than a week. Hopefully yours is back just as quick or quicker. Bit disappointing to say the least when you spend this kind of money and end up with something not perfect.

[Lance B]

Lance, your observations on "back focus" simply don't sit well with me. As far as I know if you are getting a "good" focus on the centre point and substandard performance on both the outer focus points, then the lens either has plenty of field curvature attached to it or is faulty. In the images presented above, the centre point and r/h shot show pretty normal evenness across the image, if there were lens factors at play we would see problems in all images. It is only when the outer l/h AF point is used that we see ugly looks across the entire image. Remember, we are only talking about the area under the focus points which with the Nikon FX bodies is still quite a long way short of the full image so edge softness in lenses etc hardly has any impact.
The most obvious thing to look at with this problem would be lens mount deformity except that we are seeing repeated examples of the problem occurring only at the l/h AF point and no softness at the edges using the centre point.

No, lens curvature has nothing to do with it. It is an issue with the left focus points not being accurate and to a lesser degree the right focus point. So, when I use the left most outer point, there is back focus (sometimes) and when I use the centre points it doesn't BF. This never occured on the D700 or D7000. There is no lens mount deformity, it is a left focus issue, just like those that everybody else has observed on other fora. It is an issue with fast lenses as that is what others have also reported and I believe has something to do with lens focus shift (a common phenomena with wide aperture lenses) and an issue with the left focus point. I know how to test lenses and I know what can and can't happen.

[mongo]

...Hope it gets sorted, and that it doesn't take too long.

they promised expedited attention to this problem - literally within days. Mongo clearly got the impression Nikon's reputation is hurting over this and the staff have a directive to give this sort of "repair" top priority.

[AutumnCurl]

I meant endeavour to find enough time in the day to fully test the camera, not work out how to test it.

I have had no problems with my 50mm at 1.4 on the 51 points.

but to confirm its 100% i will get my uni lecturer to give it a once over and also test it in studio conditions with a focus lens alignment chart.

[mongo]

I meant endeavour to find enough time in the day to fully test the camera, not work out how to test it.

I have had no problems with my 50mm at 1.4 on the 51 points.

but to confirm its 100% i will get my uni lecturer to give it a once over and also test it in studio conditions with a focus lens alignment chart.

that's an idea. However, both MOngo and Andrew have shown some simple ways of testing the lens above if you have the time.

[MajorPanic]

Mongo your not the only one disappointed I have much less trust in my D800 than my D700 (after AF Fine Tune) to give razor sharp images.

My expectation of what constitutes sharp images comes from years of shooting mainly 10X8, 4X5 & medium format cameras.

I have done some testing & thought that specific lenses were just not up to standard (24-70mm for example) but when I pushed it to the limit shooting tiger cubs in low light the lens produced sharp images. Others have reported the 14-24 to be good except in the deep corners but I find the lens soft in the D800. It's MUCH sharper on the D700 so I'm thinking that the AF array is not in critical alignment.
I'm glad to hear NikonAU is across the problem & I'll send the body in for adjustment soon.

Please let us know how you get on.

[arthurking83]

.....

but to confirm its 100% i will get my uni lecturer to give it a once over and also test it in studio conditions with a focus lens alignment chart.

You don't need a lens alignment chart to confirm if the camera itself is misfocusing.

The tests already shown and explained will confirm if any of the camera's AF points are not working properly.

Live View is your best friend (either in AF mode or manually focused)

[Wayne]

One thing that seems to be a trend in the testing many have done, is to shoot wide angle <50mm and wide open at less than f/4 WITH a focus distance of several metres at least, not near MFD. I have been reading lots of posts from USA owners on USA boards, who consistently test with this method and it shows the issues very well.

I'm going to do some more testing in coming days, just haven't got around to testing all my glass yet.

[arthurking83]

If there is a problem with AF module alignment, neither the focal length used nor the focus distance set will make any difference, with respect to the outcome of the issue described about the left AF point being so far out of whack.

A wide open or mid range aperture setting will haven an impact as to the severity of the problem and stopping down to a certain degree may mask the problem due to a deeper DOF.
Although in saying that, as the camera has such a high level of resolution, even when stopped down should still highlight the issue but closer to the pixel level when viewing.