Does Anyone Know when Nikon will start shipping the D800 with the focus and LCD problems solved?

pippigurl

I need glasses after looking at these!! They're alive.

roombarobot

You're right, pippigurl, the full images are too small. Here is a smaller set where I cropped out the center target:
http://www.flickr.com/photos/99475088@N00/sets/72157632788661099/with/8483158403/

What you are looking at are images of the center target when focused on the far-left or far-right using LiveView or the standard AF. Note that the LiveView-focused images are sharp and the ones focused using the standard AF are not at all sharp. This seems like a defective AF to me, what do you all think?

Msmoto

For this to make any sense to me, the full size image needs to be loaded onto Flickr. And, for this, I suspect you will have to by a "pro" membership at Flickr.

roombarobot

I'm sorry, Msmoto, this is very confusing to me and I am running these tests. I don't have a Pro account, so let me try to explain what I am doing. It is also covered in this video = https://www.youtube.com/watch?v=lkP1bGRQMdM.

I have three star targets on my wall, one under each of the left, center, and right AF points of my D800E on a tripod. For each star, I snap a picture using LiveView to focus, then a couple using the standard viewfinder AF.

For me the center star pictures come out sharp using either LiveView or standard AF. However, both the left and right stars are sharp when using LiveView, but much blurrier when using the standard left and right AF points. Here are those pictures..

I hope that makes a little bit more sense and I greatly appreciate you all talking this through with me. I need the help, thank you!

Symphotic

I saw the video and set up the same test here using a D800 and an 50 mm f/1.8. I can reproduce your results at a distance of 2.3 meters and the center of the targets 40 cm apart. When I get results similar to yours, I get no consistent AF lock confirmation. A slight adjustment of my camera position gives me consistent AF lock and true focus, so please verify you are getting a reproduceable lock in AF-S mode. If you are, I cannot reproduce your problem with my camera.
May I suggest a different test:
Try left focusing, right focusing and center focusing on the eye of a cooperative model. If the eye is in focus, the camera works. I ran this test with a real person in low light at f1.8, at 2 meters and I had no problems locking in focus and the focus is true. We make industrial cameras and often find that contrived tests give artifacts unique to the test conditions and not representative of a true photographic scene.

roombarobot

Thanks Symphonic, I'll try that. By "focus lock" you just mean the green dot?
What "slight adjustment" did you do that gave you consistent AF lock and true focus?

Symphotic

Well, I misspoke. It is obvious that neither of us were getting focus lock, as the camera was not in focus. By shifting the left or right focus square to a different feature on the test chart I could get consistent focus. But I didn't like the test, so I asked a young lady to sit for focus tests of several different focus points and had no problem. Could you try a similar, simpler test? Anyway, I don't know if you will see the left focus issue unless you use a wide angle lens.

Msmoto

The setup requires extremely careful measuring so as to make certain the camera axis is precisely perpendicular to the surface where the targets are located. This can be done with a mirror and a laser pointer. The laser is on the camera, and when it is lined up so as point to the middle of the field of view, it is on the camera correctly. Then, the mirror is attached to the surface where the middle target is located. and the camera is lined up so the laser hits the center of the lens. Now attach your targets, and shoot you photos. It is best to use electronic flash with a short duration so as to eliminate any potential for camera movement.

Once you have tested in this way, and have good data, it is possible to state unequivocally the issue is present or not on your D800.

Now, a note, with a bit of tongue in cheek...if you can afford a D800, why not get a "Pro" account at Flickr so you can show off the results? The full size images on Flickr really demonstrate the capabilities of your camera.

roombarobot

Yes, I think the setup and measuring needs to be careful, but given that I see a huge difference between LiveView AF and standard AF with this setup, I think it is precise enough. If my setup was flawed, LiveView AF should look just as bad.

Here are the full resolution images, named with which target I focused on (therefore which AF sensor) and whether I used LiveView AF or not.

I guess all I want to know is if I could expect the outer standard AF points to focus the camera well. Do others' D800/E do that? If not, then I guess I would have to focus using the center point and re-compose or focus using LiveView AF. If yes, then I think my D800E is not functioning properly.

Msmoto

@ roombarobot

I can not find my resource, but I think the second focus point from the left on the lens central axis is one of the problem point. And, I would suggest, as you have very nicely posted images in full res, and the one labeled Left Standard AF -2 seems to be out of focus. I this is the one I have described, you might have the problem. However, he definitive way to prove this is to repeat the process, five examples for each test point. If the results all are consistent and the focus for the second from the left or any of the left focus points shows a soft focus then it would suggest the problem exists and I believe Nikon will fix this.

I think they want to see the images before authorizing a repair.

You are doing a good job at demonstrating this. Just change from point to point as you cycle thorough five repetitions.

roombarobot

These are only shot with the outermost AF points, the far-left and far-right. I did not do any testing of any other AF points, except for the center. I spent all of a day just testing those three!

So what you see is 1 shot using the LiveView AF from far-right, center, far-left, all of which are in focus. Then two shots each using the standard AF from far-right, center, far-left AF points. There, when focused on the far-left and far-right, no targets are in focus. When focused on the center all targets are in focus. Does that make sense?

Thank you all so very much for helping me think through this and test! I am trying to learn as I do this. After my terrible experience with my spot-filled D600, I first want to make absolutely sure that my camera is functioning well while I canstill return/exchange it, then I'll get on to photographing. I did the opposite with the D600 and almost couldn't get rid of it.

roombarobot

I just heard back from Nikon Support, I had sent them the full-resolution test images too. They said to send it in for warranty service. There is no way I am going down that path again, after my D600 went through three service attempts. I guess it is time to exchange it and hope for a better copy.

roombarobot

OK, I re-did my entire setup and testing tonight and I think I got a different result. Please help me think through my findings and help me figure out of my D800E has a problem or not

Here is my new setup. I focused on the horizontal bars and used the stars to gauge focus sharpness. Testing Setup 2 The horizontal bars were suggested on dpreview as the outer phase detection auto focus (PDAF) points are only sensitive to vertical changes.

Here is a poster with all of my results. The top crop of the star is from a LiveView focused image, then the next four rows are from PDAF/viewfinder AF focused images. The columns are for far-left focus point, center focus point, and far-right focus point = LiveView and PDAF/Viewfinder images at Left/Right/Center poster

With this much more precise setup and a lot more testing, I think I am seeing the PDAF focused images for the far-left and far-right are not too much worse than the LiveView-focused. I think the far-right looks better than the far-left, is that true? The center is very sharp all the way through. The far-left and far-right are not nearly as good as the center. Is this due to the falloff of the 50mm 1.8G?

So, is there a problem with my camera? I really appreciate your help in this. It is a complicated test to prove if one's camera is functioning properly, but the pressure is on to do so before the return window closes. I am so very thankful for your help!


P.S. Is the star too far away from the horizontal focusing square? Did I mess up the testing?

Msmoto

I really do not see any difference in the images, which would suggest the points are all working. What I would do is to obtain a model, use the wide focus points and AF, Single Servo, Wide open, shutter speed of 1/250 or 1/500 sec. Shoot some photos using the eyelash on the near eye for the point of focus...Then see what you get.

If the camera consistently front or back focuses on the left side points, yet is sharp with the right side points...then you have a problem.

Symphotic

... What I would do is to obtain a model, use the wide focus points and AF, Single Servo, Wide open, shutter speed of 1/250 or 1/500 sec. Shoot some photos using the eyelash on the near eye for the point of focus......

Yes. I tried his setup with my recently serviced D800 and saw his results, which varied depending on where on the target my focus point was set. But if I stopped using black and white targets and used the eye of a model (the lovely Mrs. Symphotic), and then a row of books on a bookshelf, I saw consistently good focus on all focus points. The fault appears not to be in the camera but in the particular test used.

roombarobot

I wanted to do one more round of testing, just to prove to myself in an experimental setting. This time I put some text in and around the AF target and put a small siemens star next to it. Here are my results for LiveView and standard AF for the center, far-left, and far-right AF points.

I think this shows that I do not have a problem with this D800E. Anyone disagree? I think I've also figured out a better experimental setup that I hope benefits others that are trying to test their cameras during the return period. Thank you all for your help on this, I greatly appreciate it!

I assume that the far-left/far-right focus can be improved if I work on fine-tuning my lenses. I hope to get it closer to the LiveView results.

D900E

Roombarobot: I am not an expert on these tests and they are the reason why I am still hesitant to buy a D800E, so don’t value much my opinion – I am just preparing myself for the day I will decide to have no more peace, testing a new camera.

In the PDAF tests I do see a difference between the center and the sides; IMHO the left seems to be a little worse than the right, compared to the center, which looks sharp.

It is hard for me to buy any excuses like the lens curvature or the camera not aligned/perpendicular because if it was so shouldn’t the tests in LiveView also be soft?
I might be wrong but would like someone to explain this.

The amount of light is another thing that I don’t know if and how much may affect these testing. Your targets don’t look very bright and don’t know if the results will be better with a stronger light. I bought for myself two umbrellas with two 800W florescent bulbs, as with my luck I will get for sure a defective camera.

Any opinion on the light, folks? Thanks.
lorenzo

D900E

I am on the same boat of the beginner of this thread and the more I read here the worse I feel :-(
To summarize these are the defects found on the D800E by members of NRF at 01/24/2013
1. LT AF fails with PDAF
2. Lose 10-pin connector
3. Oil and dust on sensor
4. Green LCD
5. Viewfinder out of focus
6. Pop-up flash lose screws

I must be crazy but I will order one anyway before the end of March. My major concern is going through the risk of dreaming Siemens Stars every night for two months... Here is my question for all:

Any positive experience with Focal? If so, how difficult is to use it? It is a s/w made by Reikan Technology Ltd in the UK and comes in 3 versions; the Pro version is L69.95 (Pounds, equiv to $105) maybe plus VAT, See http://www.reikan.co.uk/focalweb/

Thanks!

Msmoto

I cannot see how at that price the FoCal can be automated. I have the LensAlign MkII Focus Calibration System which is nicely constructed and seems to have a good logic about it. I have not used it as of yet, but will be doing this soon. The LensAlign MkII Focus Calibration System is available from Amazon for about USD $80.

Ironheart

Lensrentals also rent the LensAlign and maybe the Focal, so you can try before you buy.