Attached are the FoCal reports from two different D800's. One purchased in March, and one Purchased today (Dec. 30th). The serial number of the one purchased in March is 30054XX, the one purchased today is 30559XX.

Am I doing something wrong???



Purchased March 2012





Purchased December 2012




Paul

"life is a lesson, you learn it when you're through" f. durst

Attachment #1, (jpg file)
Attachment #2, (jpg file)

This shows that Nikon has managed to perfect the production of D800's with poor outer focus points.
Dave Jolley

Visit my Nikonians gallery.

Yes it sucks


BTW: The D800 purchased today went back, the D800 purchased in March is back in Nikon's hands (for the 3rd time in Melville).

Not sure what to do, or where I could purchase a D800 that has reliable focus across the 51 focus points.


Paul

"life is a lesson, you learn it when you're through" f. durst

"Nikon has managed to perfect the production of D800's with poor outer focus points."

Of course there is no question about the lens, procedures, software, reliability of test, etc. How do photographs of real world subjects look and do they validate the FoCal test?

From my shots I'd say Nikon has not perfected the production of poor outer focus points.

Visit my Nikonians gallery.

Visit my Nikonians gallery.

Yes, real world examples show a lot of softness when using the outer focus points when compared to using the center focus point.



Paul

"life is a lesson, you learn it when you're through" f. durst

Yes, I agree. This is the same result that I found in my former D800e, which I purchased in early Dec.

I tested on a 35mm F/1.4 and a 24-70mm F/2.8

I also use my camera for street work, so outer focus points are critical.

So I returned it and purchased the D4. Focus on my D4 is stellar.

~Dan
www.danielstainer.com

I was trying to save on some weight. I recently had some spinal surgery and found carrying my D3 around gave me lots of problems. It was my wife who suggested getting a smaller camera after the surgery (right around March 15th, just in time for the D800). I jumped on the chance and it has been unreliable when shooting hand held using edge focus points (and like you, I compose using all focus points, quite often edge to edge) It is AMAZING on a tripod using LiveView.

The D4 was not an option because of the size and my D3 I find to still be very reliable, but now I use in mainly for sports photography on a mono pod.



Paul

"life is a lesson, you learn it when you're through" f. durst

Since the tests indicate consistency, it suggests other than missaligned focus point have all other variables been addressed, for example a known flat field lens used. What lenses repeated this pattern?
When you tried Nikon service before, did you include your current test lens? Did they adjust it or just report that is was within tolerance?
Stan
St Petersburg Russia

Visit my Nikonians gallery.

>. . . have all other variables been
>addressed, for example a known flat field lens used.

That's the thing that has me confused about some suggestions on this site that a wide angle lens should be used to test.

WA lenses are (generally) more likely to have curved fields, which it seems to me would contribute to "mis-focusing" by outlying focus points, unless one understands exactly what's going on in the process and ensures that the test isn't impacted by that fact.

I have a severe left-focus issue, and I choose to use a 105 Micro (considered very flat-focus) to test.

AviSys

AviSys,

I... like you, use an 85MM prime to do my testing... I just think that Nikon has a lot of bad cameras (I have only really tried two, but I see a lot of other reports) in the market place and I don't know if many people are just putting up with badness and don't know it.


Paul

"life is a lesson, you learn it when you're through" f. durst

Stan,

I don't see consistency across the focus points, they vary from 273/.64 to 429/1.00 and 335/.59 to 564/.99 (Left to Center, March and December respectively).

The first two times I included my primary lens and they were not able to get it right.



Paul

"life is a lesson, you learn it when you're through" f. durst

>I don't see consistency across the focus points, they vary
>from 273/.64 to 429/1.00 and 335/.59 to 564/.99 (Left to
>Center, March and December respectively).

My recollection of FoCal's instructions is that they say "you may have a problem if any of the numbers are below 0.50". Although your numbers are not uniform across the frame, you may not actually have a problem. The outer AF points on all the Nikon SLR's I've owned have been somewhat less reliable than the centre ones.

I'm sure FoCal and similar tools can be very useful, but I'm equally sure they are causing us to worry about stuff that never concerned us in the past

Brian
Welsh Nikonian
Check out the Nikonians Team pages

Good advice Brian. I sometimes wonder if there aren't more people buying Nikons to test them than to make photographs.

Russ Lewis
www.FineArtSnaps.com

I love the 85 1.4 but it sure is not equal at the edges to the center, which is all I care about since I use it for portraits. My worst Fx lens is the Sigma 50 in the corners but its center is so nice for what I got it for. I am sure both of those well regarded lenses would prove to be very poor in these tests. My 24 1.4 is the same way, and lens tests show a distinct fall off wide open at the edges in resolution.
How does it look in real world photos? Pretty darn good.
Here is the charts for the well respected 85 1.4D which I have. I would never pay attention to resolution or focusing tests at the corners unless on my D7000 Dx. I am not seeing the major problem you do in your results, can you post an otherwise compelling image that was negatively impacted less focus resolution in the corners? When the lens has much less resolution, the AF system has less ability to resolve the optimum focus plane. MF focus is best in those cases because phase information that is distorted to the AF system can still be determined by eye and contrast AF.
Stan
St Petersburg Russia

Visit my Nikonians gallery.

Attachment #1, (jpg file)

Paul,

FoCal is a great tool but like Brian says it creates a whole new category for us to worry about.

My $0.02...

Center AF points and center frame sharpness should be the best. Outer AF points and edges of the frame simply will not be as good, but on a high precision machine they do need to be repeatable and perform at an acceptable level.

Lock the camera down and shoot some real world images with your favorite lens. Review your test data from your FoCal reports and see what those numbers actually look like on screen. Try contrast and phase detect focus, try AF-C and AF-S modes as well.

I have FoCal as well and used it on my D800, its a useful tool but when I take my camera out I'm not shooting test charts.

Good luck, let us know what you find.

Those test charts look about like the ones I got with my D800 and Sigma 85mm lens. My Nikkor 50mm f1.8G also gave similar results. They are actually what I hoped that I would see - corresponding focus points on each side of center gave about the same QOF. My real concern was the possibility that the left and right focus points would give drastically different results. Sure, I wish that every focus point was as good as the center one. But I am very happy with my D800 and my current lenses because the center of the images are amazingly sharp. That's how I shoot in everyday life...

Visit my Nikonians gallery.

Paul

Have you tested earlier cameras with this lens and gotten different results? I'm looking for something that rules out a lens problem or testing problem.

Which model of the Nikon 85mm lens are you using? The newest 85mm f/1.4 AF-S tests remarkably well across the entire frame but there is still some drop off toward the edges wide open. Other 85mm lenses show more pronounced drop off from center to edge. And if a lens element is very slightly misaligned, the edges can become much softer.

Eric Bowles
Nikonians Team
My Gallery
Workshops

Nikonians membership — my most important photographic investment, after the camera

Eric,

Makes sense. I use the 85 f/1.8, and I've tested my 24-70 f/2.8. Both have similar results. I have also tested both wide open (not testing any other apertures).

I get where you are going and when I get the camera back from Melville, Ill try some more variability with lenses to see if the issue is consistent.

In the past, with my D3, across all my lenses, I have dialed in micro focus adjustments with each lens, and found no issues at any of the 51 focus points. With the D800, with real world examples, I see I difference from center to edge focus points. Now with the 36MP of the D800, it may be more pronounced than the D3, but that will be one of the other tests I run, comparing the D3 vs. D800 with each lens to see if the issue existed before and I just didn't notice.

Thanks for the advice.


Paul

"life is a lesson, you learn it when you're through" f. durst

I believe one of the other members using FoCal found that the increased resolution of the D800 allowed it to show more focus error - the center was so sharp that it allowed a higher level of resolution - and normal performance at the edges was worse in comparison to the center.

I don't think FoCal is ready to support using the numerical values to compare cameras or lenses. But it should give you a good indication of performance and how fine tuning should be adjusted.

There is definitely an art to interpreting the results. FoCal produces a lot of information and is a good tool. Having a bit of context from different lenses and camera bodies should be helpful.

Eric Bowles
Nikonians Team
My Gallery
Workshops

Nikonians membership — my most important photographic investment, after the camera

Here is what set my mind at ease. I took some test images of real life subjects using the extreme left, right hand and center focus points at distances I would likely use . Then had 24" x 36" prints made and compared the prints.

I make sure my subjects have very hard lines/points so there is no ambiguity.

Visit my Nikonians gallery.

Visit my Nikonians gallery.