D700 and autofocus

I wish I could share the experience, if for no reason other than to see if I could help. My D700 focuses just fine, but I have the sneaking suspicion that in your case, you are not aware which of the 51 points are cross-focusing points (I'm not sure about the term, but you get my drift). Of the 51, only around 16 or something have the capability to focus in both directions, not all 51.

But what you say about the D200 is not entirely true - it had less cross-focusing points than the D700 - chances are you were simply more accustomed to using those. I know cause this happened to me as well the first time I used my D700 and I know for a fact a friend of mine (who also bought a D700 after his D80) managed to shoot an entire event with EVERY single shot being completely out of focus because he failed to learn to use the new equipment properly.

Personally, whenever I shoot portraits I always use the center point, focus and then recompose my shot...

Overall speaking, the D700's Multi-CAM 3500 AF module is far superior to the D200's Multi-CAM 1000. I have both cameras. The D3 and D300 also use the same Multi-CAM 3500, but it is optimized for sports photography. The Multi-CAM 3500 has 51 AF points and 15 of those 51 are cross type. Those are the more sensitive ones. Please see the attached image showing where those 15 cross-type AF points are and I would try to use those as much as you can.

The problem on the D700 (and D3) is that all 51 AF points are concentrated in the center, especially the 15 cross type ones. Therefore, when you shoot portraits, there are no good AF points near the top of the frame, especially where the subject's eye tend to be. Therefore, you need to experiment more with you new D700 and work around this issue.

Can I say that I am a portrait photographer, have had a D700 for a month now, and share exactly the same focussing issues as Steven. I came from a `D2x which would acquire the focus almost instantly, and find that the D700 hunts for a second or two or four, even using good studio lighting (ie not dim light). I always focus on an eye, and use the centre AF point (focus and recompose - I am an old lag) and for me this is my one big problem with the D700. In the last two jobs I have become more and more frustrated at losing good poses because the AF does not find the focus fast enough. I have done loads of comparisons with my old D2x and I am becoming increasingly irritated with the D700. To be honest I thought it was just me, because I have not heard anyone else complain till now. In fact I was considering selling the camera because of this issue.

I have this with my new 24-70, and my well used 85 1.4.

If anyone has any thoughts on this - maybe there is a problem with my (and Steven's) camera...

David, the big difference is that the D2X is a DX body and therefore the AF points cover a much higher percentage of the entire frame. Additionally, the D2X has 9 cross-type AF points that are very well positioned for the DX sensor.

AF point position for portrait photographers is certainly a weakness on both the D3 and D700. I have clearly pointed that out in the up-coming D700 review here in photo.net. Generally speaking, on full-35mm-frame DLSRs, the AF points are too concentrated in the center of the frame. The Canon 5D and 5D Mark II as well as the Sony A900 all have the same problem.

Hopefully Nikon will redesign the AF module in the future, but there seems to be some technical difficulties when the frame size is much bigger than DX.

Hi Shun and thanks for the reply.

I understand what you are saying, but I presume the centre AF point (which I use almost exclusively) should be perfectly usable for my needs. The fact that horizontally or vertically it is very slow to 'find' focus is disturbing, irritating and ultimately loses me shots.

I own a D700, a D300 and have used two different D3 bodies extensively. At least in my experience, on any one of those bodies, all 15 cross-type AF points work very well under indoor, dim-light conditions.

I was partially aware that not all focus points were created equal, but the magnitude of the inequality is more striking then I thought. Shun's diagram is very useful as it is those focus points that function extremely well no matter what I am looking at. The reminder have variable focus capabilities depending on the content of the image at the focal position.

It appears that lines running perpendicular to the long axis of the focal point box are difficult for the camera to see and in when the camera is held in a vertical position, the dominate eye structures....such as eyelid in fact run along the axis more difficult to detect.

I suspect some of my puzzlement is caused by my expectations of more equal performance (particular on faces and more complex objects) across the 51 points.

Below are two images. One taken in the vertical orientation and one taken in the horizontal position with the actual focal point shown. With absolute certainty, the 15 centered focal points worked perfectly and instantaneously in either camera orientation. Again with absolute certainty, in the horizontal position, none of the focal points outside Shun's red box would focus..period...no way. But place the camera in the vertical position and all 51 points focused instantaneously.

The focal point is very simple so I might expect some problems, but I would not expect it to affect taking portraits.

Shun, I remember you commenting about the points all being the center of the sensor field... but the constraint of the cross hair focus point to the center of the 51 makes it even worse.

In reflection, i suspect that with the D200, I addressed this problem by simply using the center focus point, locking focus and then reframe the composition as desired and when I got 51 points in the d700, I loved it and it worked great..so I used it more and more...and had not realized how different the points really were.

Anyway, thanks for the help..I will need to find a comfortable work around or go back to old ways!! Steven

Second try to upload the image mentioned above...

I would add one last observation. If the camera is held horizontal, the outer points will not focus as described above and that is absolute. If one rotates camera about 15-20 degrees in either direction off horizontal, then it begins to find the focus and the more one rotates to the veritical position, the 'tighter/faster' the focus finding becomes.....