How good is manual focus with the green dot?

<p>Some people seem to get on OK with manual focus using the little green dot; others less so. I've always imagined that the green dot was driven by the autofocus system and should be as accurate as autofocus - neither better nor worse.</p>

<p>I decided to try an experiment. I put the D40 on a tripod with my only AFS lens - the plastic-fantastic 18-55 at 55mm and f/5.6. I focused on some text about 2.5 metres away using the green dot manual focus and took a couple of pictures. I then repeated the experiment with autofocus and compared. They looked just the same when zoomed in. What's more when I switched from manual focus to autofocus the green dot was still lit and the autofocus motor didn't run - presumably because the system judged focus to be OK.</p>

<p>I then got out my F90x (N90S) and fitted my only FX autofocus lens - the 28-105 f/3.5-4.5. I focused on the text with the lens at 105mm using manual focus (a dot and arrows this time - much better) and then switched to autofocus. Again the motor didn't run.</p>

<p>My conclusion is that manual focus using the green dot is much like using autofocus - neither better nor worse. Anyone with a much wider aperture autofocus lens care try to try the same experiment, mine are all manual focus?</p>

 

<p>At least on my D2H the green dot focus confirmation is only approximate. It's accurate enough for casual use, especially if I stop down a bit so that DOF compensates for slight error. But it's not accurate enough for critical focus with macrophotography or telephotos wide open. I still rely on visual confirmation for that. Fortunately the D2H has a bright, crisp screen that facilitates accurate manual focus.</p>

<p>I was "playing with" my Nikon F4s and Nikkor 24-120mm f/3.5-5.6 AF-D lens last night, and found that the "green dot" in manual focus and in autofocus to be pretty much exactly the same. The things that the camera/lens couldn't lock focus in on with autofocus (low light, low contrast, etc.) also wouldn't lock in using manual focus and the green dot.</p>

<p>Most of the time when I need to manual focus is when I am using a wide aperture lens wide open in dim light, for example, a f1.4 lens shooting an object that is just 1-2 feet away. The green dot does not work reliably enough under such demanding situations so I now use the live view feature on the D90 with magnification. Under good light with your object relatively far away and with lots of DOF, the green dots will work fine.</p>

<p>I think at least with wide angles the dot indicates correct focus at a range of focus distances that is way too large for operation at wide apertures. I get the feeling that this is done in order to make manual focus "possible" with autofocus lenses that typically have a very fast turn. If the dot required precise focusing, people wouldn't be able to find it easily as the turn is to quick. This is unfortunate for people using manual focus lenses. I think the only good solution is the use of a focusing screen that is made with manual focusing in mind. Thankfully, these are widely available from 3rd party suppliers for a number of DSLR cameras. </p>

<p>For me on my D50 it's pretty darn good...</p>

<p>Green dot accuracy with small aperture lenses seems to be fine - for static subjects. Trying to use the green dot confirmation for something that's moving - even just a person for a portrait - and frustrations mount quickly.</p>

<p>Ilkka,<br>

I did wonder if what you say might be true but I don't think it can be, at least on my two cameras and lenses. I found that switching from manual to autofocus with the green dot on failed to initiate any response from the autofocus motor, presumably because the autofocus system found the error was within bounds.<br>

In fact on the two lenses I tried getting the green dot on reliably is very difficult!</p>

 

<p>While I used to have 20/15 vision, I now have 20/40 vision. Whatever its limitations, the green dot is more accurate than my own eyesight, even when corrected to better than 20/20. That said, as Ilkka indicates, you need to be acutely aware of which focus sensors are in use and what portion of the image the camera's focus sensors are looking at.</p>

<p>I have to ask this because I have always wanted to know: Is there a split screen focusing screen available for the N90s, and D100? If so, does this affect the Autofocus capabilities? I was always told a split screen focusing screen would screw up the AF.<br>

Also of note with my N90s and the Nikkor AF 70-210: In continuous autofocus mode the lens will continually switch focus with small objects that exist in two different vertical planes. I've never had this problem with my other Nikkors, but the 70-210 is very sensitive to this.<br>

In manual focus mode on my N90s, I've become pretty adjusted to using the 'focus dot' with my peripheral vision. It's not great for fast scenes but for typical shots, it gets the job done. And for landscape photography, it is almost always necessary. At least in what I shoot.</p>

<p>Richard, perhaps you may find of interest the threads below:<br>

In this one my experience with the green dot on the D200 with AiS and AFS lens: <br>

<a href="00M3Pp">http://www.photo.net/nikon-camera-forum/00M3Pp</a><br>

In this one, check Rorslett` comment about a manual focus testing procedure (18th post):<br>

<a href="00M2MI">http://www.photo.net/nikon-camera-forum/00M2MI</a></p>

<p>When I had a D60 (which has a green dot and a little +- bar to give you direction) it was accurate enough to use at f/1.4 with a 50mm prime. It was also very demanding - I had get it riht on to get the D60's green dot of approval. I didn't get the impression it was doing anything different from the normal AF system.</p>

<p>Now maybe because it has the +- bar they made the constraints for the dot tighter...</p>

<p>Jose,<br>

Thanks for that it's interesting. I wonder if the 20mm f/2.8 AIS you tried has some focus shift on stopping down? I think that might explain the asymmetric behaviour you observed since as I understand it the AF detection system is using a portion of the lens corresponding to about f/5.6. Some variability with AF is to be expected; there has to be some hysteresis, or 'dead zone' to prevent, or at least minimise, hunting. With manual focus using the green dot I'd expect the behaviour to be symmetric, that is the same whether approached from near or far focus.</p>

<p> </p>

<p> </p>

<p>Generally speaking, it's good down to f/2.8. I believe the D60 actually uses the EC graph as a focus indicator in manual mode, which I imagine is more accurate.</p>

<p>I have a Katz Eye screen installed and the green dot is usually right on when I focus manually...and I use both to confirm each other.</p>

<p>From my experience with a D50 and Fuji S3 Pro, focusing wide open (1.4 & 1.8) produces unacceptable results. I never trust it, especially with the tiny tunnel-vision viewfinders of both cameras.<br>

Manual focus on my FM however, is a joy. I don't even need the focus aids (split screen, microprism etc)</p>

<p>When my D3 is mated with the Zeiss 21mm Distagon and it is focused really closed, the green dot is not trustworthy. The lens is focusing short (front focusing) and so, I had to compensate it manually. I tested it with a lens chart on tripod a few times. I have also discussed it with Zeiss. They suggested using the LCD to focus for critical projects.</p>

<p>I used a D90 with a 50 f/1.2 AIS and the green dot and could not get focus accurate below about f/4. My eye sight is not good enough to rely on, so I had to get rid of the lens. If the viewfinder had the arrows like a Leica I think that would greatly help. I had an M2 and my pictures were sharp at f/2 consistently.</p>

<p>It is very good on the D2X, but I trust my eyes just as much. For sports I do not have time to check the green light. I have been manual focusing for 25 years!</p>