What factors determine focus speed, lens, camera body, or both?

Msmoto

As the discussions of the new Sigma Art 24mm f/1.4 have gone to the question of what determines focus speed, I thought a new thread could be helpful. And, this has not been discussed for several years as best I can tell. SO, go for it!

funtagraph

Is there any reason you didn't mention "accuracy"? I mean, what good is high speed focussing when it's just missing the point? Or did you just imply, AF without accuracy is not what we're talking about? I think I get your point, but sooner or later somebody will show up and says, his/her slowmo AF is every time spot on. And it's a valid point.

snakebunk

I thought we where talking about ability to focus in low light. Not sure where to write now, but I try here

My point is that a lense cannot have a low light auto focus quality compared to other lenses with the same focal length and aperture. It is a quality of the camera to find focus in low light. If I am wrong, please tell me why, I am eager to get rid of any misunderstandings I still might have.

Msmoto

This thread is about all issues which may be related to the general discussions of AF, and just how fast accurate focus can be obtained under any conditions. If one wants to discuss other manufacturers as well, that is certainly permissible as there are some folks who believe Nikon is not the fastest focusing camera.....

We have not had a thread about this topic since the previous NRF iteration according to my search.

funtagraph

@snakebunk I see it the same way. If we take as granted the acceleration behavior of focus drive is the same in two compared lenses. Third party manufacturers will have to find a compromise between those acc. behaviors between their supported systems.

And we also need to know if the moved mass - glass and mechanics - is about the same. Sigma lenses introduce themselves to the camera as some other similar lens. I found accidentally, depending which one I AF fineadjust first, the 24-105/4 is a 24-105/4 - or a 105/2.8 micro Nikkor. I don't know about the 50/1.4 Art, as it is introducing as 50/1.4 and I don't have another one. But the 50/1.4 from Nikon is smaller, lighter and therefore faster to focus if the drive has the same power. If Sigma uses a more powerful drive in their Art lens, I see a possibility of continuously overrunning the right spot. It doesn't happen - or not often enough to be noticed.

Msmoto

Oh, a note about my experiences.... one of the critical factors IMO when shooting rapidly moving targets.....i.e, racing venues, bikes and cars, I have found it is very important to begin the AF process a moment before actual exposures. I think this allows the Predictive Focus Tracking to gain info and this has been true for the D90, D800E, and D4 with lenses from the old 80-400mm f/4.5-5.6, up to the "old" 400mm f/2.8.

Thus, my conclusion is some operator input can effect how effectively the camera AF will actually work.

However, I must admit, my experience, once again, suggests that all the Nikkors, Nikons focus so quickly I really have difficulty seeing much difference. Some of the more experienced on NRF probably have different observations.

funtagraph

The use of focus limiter is also helpful, but only tele-lenses or macros do have this switch.

And getting a good target with enough structure for phase detective AF also helps a lot, as well as using monopod/tripod for the longer FL.

Replying to your question, I think "both" is the appropriate answer. If the camera's AF module has troubles to detect focus, what could the lens do about? If the lens is too slow, the camera can't make it faster. A simple lens construction or light weight glass and tube materials will also help. The fastest mass to move is the one you don't have to move - like a Lytro.

sevencrossing

Replying to your question, I think "both" is the appropriate answer.

+10

This also applies to IQ and in fact make most of the image making process
IMHO it will always be a combination of the lens and the camera. You are not going to the best results with "good glass " and camera with a first generation sensor or visa versa

needles to say. The really important factor is the person behind the camera

Vipmediastar_JZ

Disclosure: I have very little knowledge of the technicals of camera and lenses. Below are just my observations and opinion.

My 70-200 2.8 recently tends to be slower to acquire focus with the Df and D810 shooting ice skaters in overcast or partyly sunny days. It hunted in a dark room with little ambient light by switching to the 105 the problem went away. Prior to winter it was very fast acquiring BIF with minimal hunt if any.

The 24-70 2.8 is fantastic in aquiring focus in dark rooms.
The 105mm 2.8 vr aquires focus before I even press the af/on button (exegarating but that's how fast it is)

For my Sigmas from memory I can't comment to specifics because I haven't tested them recently only when I first purchased them to do the tests. If I would say that they are slower to acquire focus I would say that it is with the Df.

So then that leaves does the camera have anything to do with focus? I would say yes unless my af/on button is getting sticky.

I shot recently with the df with the sigma 35 and the d810 and the sigma 50mm at the auto show @msmoto was there too so she can relate to the light.
I had very little issues focusing with both lens. I have had issues with the 35mm in the past where it hunts for focus in a diffrent location for an auto show with the d800. With the df I had no issues but aquiring focus was not as fast as expected.

Im going to another auto show this weeekend and I will have the same setup. I will switch from af/on to just the shutter button and give my observations (granted i do go to the show).

heartyfisher

Of course the subject and the amount of light are also factors. also as MSMOTO says technique helps the type of focus used for example phase detect and contrast perform differently and technique plays a part in the performance.

Also of course the "body" and "lense" components can be broken up into several factors and sub factors as well.
1) for AF-D lenses. the AF motor power and the lense steping/gearing ratio.
2) for AFS lenses. The focus algorithm and settings in the lense. The AFS lens motor and of course the gearing ratio.
3) The camera AF mode . the actual AF sensor used. eg the left edge sensor may have problem focusing vs the middle cross type one. The predictive algorithm. etc ..
4) Some even mentions a difference in AF performance due to the metering mode used.

I am sure there must be a few other factors I have missed out.. so the answer is much more than both the body and lense.