Accuracy of ScrewDriver Vs AF-S Lenses

[arun_seetharam]

<p>Yea, Focusing speed is the primary issue being addressed, understood. I was wondering about the accuracy/sharpness in the focus between an AF-S and a non AF-S lens. Is this the same across these two lenses? Which could err more so to say.......lets say in a situation where I am shooting with a standard lens, hand held. <br>

1. Shooting a bird that is jumping around but is not a fast moving object - Static focussing<br>

2. Shooting a speeding car on a race track with a gallery background. - Continuous focussing<br>

Assuming that in both cases the lighting is pretty good. <br>

Any experiences or feelings regarding this? <br>

Camera driving the lens Vs. the Lens driving itself. :-) </p>

[Peter_in_PA]

<p>I compared my 35mm AF-S to my 50mm AF-D screwdriver lens recently. Both are f1.8 lenses, and although they aren't the same focal length, I thought it was a reasonable test.</p>

<p>I found no practical difference between them. I didn't clock them or anything, but the AF-S lens focused only a teeny bit faster, and neither lens hunted very much. Both seemed just as accurate. There was no obvious advantage to either imho.</p>

<p>I then compared the long end of my Tokina 11-16 f2.8 screwdriver lens with my 18-200 AF-S lens. Only a half stop difference, between them (I was testing the 18-200 at the short 18mm end). Both were just as accurate, but that Tokina focused perhaps a teeny tiny bit faster, even though it was screwdrive lens. I only used the center (cross) sensor on my D50. Interestingly enough, I'm finding I like the "clutch" mechanism on the 11-16 more than the over-ride on AF-S, as once I pull the ring back to focus, it stays in manual till I tell it otherwise. I can't do that without taking my eye from the finder on the 18-200 (too many switches, never can remember which are which).</p>

<p>years ago, I remember the 17-35 AF-S being WAY faster than the 20mm f2.8 screwdriver lens we had where I used to work, but I never compared them critically.</p>

<p>I don't think accuracy and speed differences are really due to AF-S vs. screwdriver, but are more dependent on the lens speed, focus throw, stuff like that. I'm interested in others' experiences, though, as mine are limited.</p>

[anthony_bez]

<p>I bought a AF-S 50mm f1.4 and was very disappointed that it was slower focusing than my old 50mm f1.4 AFD (mounted on D300 & D700).</p>

<p> </p>

[breogan_gomez]

<p>This just an asumption but It is logical to think that AF-S would make real difference on long telephoto lens.</p>

[ilkka_nissila]

<p>Accuracy depends on the specific lenses you're comparing. Inherently the AF-S system is more precise and hunt less but accuracy (how close you get to the best focus) is not the same as precision (which means how finely the control can be made). In my testing with static test subject I found the 50 AF-S to be less accurate than the 50/1.8D; the 1.4G had to be manually focused to yield accurate results at wide apertures. However in practical use with moving subjects I've found the 1.4G to work very well.</p>

<p>There is some discussion about 105 and 135 DC and their autofocus accuracy, many users report systematic focus errors and I've mostly observed random focus errors but also with some bodies systematic ones. Manual focus accuracy is very good with these. It looks like the control of focus distances is implemented with insufficient precision, or both. It also depends on how detailed the subject area corresponding to the autofocus sensor area is. This lack of accuracy is mostly an issue at f/2 and to lesser extent f/2.8. On the other hand the optical quality of the 105 DC is wonderful. The 85/1.4D is autofocused <em>very</em> accurately on the FX bodies; even sufficient for shooting at 1.4. The autofocus accuracy with the 50/1.8D is very good and so is it with the 180/2.8D on a relatively static subject; with moving subjects AF-S type lenses tend to be better. Just a few examples.</p>

[andylynn]

<p>I've been hard pressed to find problems with accuracy of either type of lens that weren't caused by movement, low light or user error. As for the speed, with AFS it varies by lens (some have faster motors or shorter travels than others) and with screwdriver it depends on the lens travel and the body - more expensive bodies sometimes have faster motors. But the issues you're talking about have more to do with the AF electronics than anything else - 3d motion tracking is implemented better in the pro grade cameras (D300 and up).</p>

[alex_lofquist]

<p>It is good to see that there are people out there that know the difference between precision and accuracy! Keep up the good work Ilkka!</p>

[joseph_leotta]

<p>Correct me if I'm wrong, but isn't the focus sensor in the camera and not the lens. <br>

Whether its AFs or D the same sensor makes the decision as to when the image is in focus.<br>

Doesn't that mean that the camera is making the decision and not the lens. So both types of lenes would have the same accuracy on the same camera?</p>

[ilkka_nissila]

<p>Joseph, there is a tolerance to what the camera considers acceptable focus and this is typically too large for wide open use. The AF and AF D lenses are focused by the screwdrive type mechanism which has some play which slows things down when the system needs to change direction. AF-S lenses typically start focusing until they're in focus. No jitter or play. AF and AF D lenses often focus until close, then its starts to refine the focus and there is some play in this process due to the mechanical linkage, so the refinement takes longer and is more erratic. It's also possible that different lenses have different "minimum steps" of adjustment that the motor can implement and their relationship to the depth of field wide open may be different.</p>

[arun_seetharam]

<p>Yea, the last microsteps of focus adjustments could matter. </p>

[rarmstrong]

<p>I have found that the accuracy of my Nikkor 80-400 mm on my D300 is excellent. For very fine adjustements, like a bird in flight, it is very quick as long as the lens is already focused near the object of interest. Rapid adjustments are tough, though, as you would expect with a screw drive. I'm also sure that the AF module in the D300 helps. Once the lens finds focus...it is tack sharp. Good light also helps.</p>