Mirrorless Focusing Ability Field Tests....

Thom Hogan has an interesting field test of mirrorless cameras and focus on birds in flight, among other things....

"In particular, I was interested in the difference between performance of the Panasonic G3, the latest Olympus Pens (with claimed "world's fastest AF"), and the Nikon V1. Simple answer: it wasn't even a contest, as the V1 won hands down every time. Indeed, for many bird passes it was the only camera that got a critically sharp bird (and often more than one or two frames of them). More often than not the Panasonic got no sharp birds in close in the passes by the swiftest birds. When it did manage a sharp bird it was usually much further away from me and it was "one and done." The Olympus E-PL3 and E-P3 performed slightly better, but not by a lot. (Lenses in use: mostly Nikon 30-110mm, Panasonic 45-175mm X.) "

http://www.sansmirror.com/newsviews/galapagos-focus-testing.html

I for one, agree that the Panny focusing on birds, or planes, in flight, sucks...not that I do it much, but when I do, it would be nice if it worked. I am using a GH1, and keep wondering if another MFT body might do it....seems not....guess I will wait for a bit, since I have four lenses already...and the GH1 does fine, otherwise.

I have found that I can focus on BIF and planes and get some pretty good shots with my Pana G3 - but I don't use the LCD, I use instead a red dot sight and keep the focus box at the next-to-the-smallest size, using AFS and not AFC. Thom Hogan does not say how/what he tried on the G3 to get the shots, just that he tried and had problems.

Anyway for me Mr. Hogan's pronouncements don't carry a whole lot of weight especially when my experience is to the contrary.

Yes, for objects causing the focal plane to change as fast as or faster than the rate of change a given lens can manage combined with CDAF causes focus problems on the G1 and my wife's GH2. Neither perform well on fast moving subjects such as birds in flight.

There are ways to improve performance:

Stopping down the lens while increasing ISO. Allows system to lock focus on a moving object much faster due to greater DOF. This is likely why the V1 operates so much better, the DOF due to the smaller sensor is MUCH deeper, so it doesn't have to try so hard to get perfect focus. Quick math says if DOF is twice as deep, focus on same system needs to work for 1/2 as long.

Use a different lens: example the 20mm is fairly slow and does not do well on tracking fast moving objects. The old 14-45 kit is pretty fast and does well, but not enough reach for BIF. The 45-200 does a lot of hunting on moving objects, and I have had no luck with BIF on this lens. Probably due to busy bokeh confusing the CDAF at times. I have no experience with the 100-300 which I wonder how well that may serve, or the 14-140.

--
My pictures...
http://picasaweb.google.com/wymanfamily3
http://www.markwyman.com/photos/default.asp

You need to mention this is a test of continuous focusing, and NOT focusing speed.

Big difference. Never use continuous AF on a CDAF camera. Always use S-AF.

FWIW, I wrote to Thom Hogan about the difference between AFC and AFS, and he said he sets both his Oly and Panny to AFS almost all of the time. He confirmed what I experienced, that AFS gets you better stills than AFC.

Everdog said:

You need to mention this is a test of continuous focusing, and NOT focusing speed.

Big difference. Never use continuous AF on a CDAF camera. Always use S-AF.

Click to expand...

I certainly came up against this issue today, trying to shoot some birds in flight with a G3 and 45-200. Shooting at 200mm, AFS mode, centre focus point, and not one sharp image of a bird in flight. It got me wondering how much of this is down to the lens, and how much to the camera's focusing system. Put another way, I wonder if the 100-300 would be an improvement or if it's just a limitation of the camera at such high magnifications?

Andrew

That images of birds from more than 3 years ago even exist isn't it?

I mean, according to most people who write on these forums, and Hogan, their Nikon D3X is the only camera capable of taking such pictures, and that camera is only 3 years old!

One thing is that, while testing AF in these folly extreme cases, you get some kind of result you can post on the net to be interesting, another is how many people actually put stock in these numbers... And this is not just in the camera business mind you... I own a car also, and on car forums, you get various test and measurements on 0-100km/t acceleration, speed tests, cornering tests etc. All as much use as these tests...

Now if Hogan actually put these cameras to real world tests that had meaning, then I would give him some credit... But this? c'mon...

liquidsquid said:

Stopping down the lens while increasing ISO. Allows system to lock focus on a moving object much faster due to greater DOF. This is likely why the V1 operates so much better, the DOF due to the smaller sensor is MUCH deeper, so it doesn't have to try so hard to get perfect focus. Quick math says if DOF is twice as deep, focus on same system needs to work for 1/2 as long.

Click to expand...

This is interesting - I haven't read much commentary on comparing the Nikon 1 system AF at equivalent DoFs to larger sensor systems. I wonder whether the same holds for NEX, which seem to have slower AF speeds than the M4/3 cameras.

I had thought that the Nikon 1 focus speed was more directly related to Nikon's CDAF/PDAF hybrid focusing method, but that it used only CDAF in lower light?

Every once in a while, I take a fast manual focus lens (like a 100 f2 Rokkor) to an indoor soccer match, and I am surprised by how reasonable it is to manually focus....

I don't think most people will argue that CDAF is less than ideal for C-AF on fast moving subjects. He hasn't really demonstrated anything new regarding the m43 camera. What was more interesting to me was that the V1 out focused the supposedly better DSLR with PDAF in this test.

I'm not that impressed with the Nikon 1 system as a whole, but everything I've read convinces me that Nikon has done a great job with the AF (even granted that the deeper DOF of the smaller sensored Nikon probably helps in this test). If Nikon moves this technology to their DX and FX line of cameras, and it works as well there as it seems to work in the CX camera, they're going to steal a march on other DSLR vendors.

I don't shoot BIF as a general rule, but with existing m43 cameras I'd be more likely to try using AF-S and repeated presses of the shutter release rather than AF-C and burst mode. that technique works quite well with moving automobiles. Shooting birds against a solid sky you could make the AF area pretty large, knowing that the only object in the focus area is going to be the bird.

--

Bokeh is the aesthetic quality of the blur in out-of-focus areas of an image, or the way the lens renders out-of-focus points of light. Bokeh is not the same as depth of field (DOF).

MadsR said:

Now if Hogan actually put these cameras to real world tests that had meaning, then I would give him some credit... But this? c'mon...

Click to expand...

Shooting birds in the Galapagos isn't real-world?

--
MFBernstein

'Wilderness is not a luxury but a necessity of the human spirit.' - Ed Abbey

Using only continuous AF on a camera while admiting the Single AF is much better, is not a good "focus ability" test.

I really love my GH2 with the 100-300mm,

BUT BIF is indeed bad, when i'm with my friends, with their DSLR's, it's really frustrating how many good pictures they can take, even of small birds!

I have to focus on a subject, about the same distance as the BIF and then try to get it in focus.

During that time, the bird almost past away.

I hope there will come a solution for this, it would be the one and only reason for me to move to DSLR.

Bob Meyer said:

I don't think most people will argue that CDAF is less than ideal for C-AF on fast moving subjects. He hasn't really demonstrated anything new regarding the m43 camera. What was more interesting to me was that the V1 out focused the supposedly better DSLR with PDAF in this test.

Click to expand...

In theory, the 1's focusing system is the best of both worlds, and no DSLR on the market can compete with it. Even the SLTs with their magic mirror and PDAF shouldn't be as good as the combination of both PDAF and CDAF.

The 1 system is a one-trick pony, but it's a pretty damn cool trick.

Jeremy_T said:

In theory, the 1's focusing system is the best of both worlds, and no DSLR on the market can compete with it.

Click to expand...

well - actually metering sensor in dSLRs can be used for CDAF along w/ the dedicated PDAF system... it is moving towards this... now D4 can use metering sensor for face detection... plus nothing prevents dSLRs (or fixed mirror cameras) to use both of the worlds - they can have sensors w/ CDAF/PDAF combined plus they at the same time can have dedicated PDAF system and CDAF done on metering sensor... expensive, but doable for higher end cameras... certainly lenses shall be designed to support CDAF operation - but while mirrorless cameras can only combine CDAF/PDAF on sensor and on sensor PDAF is more demanding for light, mirrored cameras can have both worlds available.

The solution may be as simple as using AF-S instead of AF-C. What setting have you been using to shoot BIF?
--

Bokeh is the aesthetic quality of the blur in out-of-focus areas of an image, or the way the lens renders out-of-focus points of light. Bokeh is not the same as depth of field (DOF).

robertro said:

robertro said:

Stopping down the lens while increasing ISO. Allows system to lock focus on a moving object much faster due to greater DOF. This is likely why the V1 operates so much better, the DOF due to the smaller sensor is MUCH deeper, so it doesn't have to try so hard to get perfect focus. Quick math says if DOF is twice as deep, focus on same system needs to work for 1/2 as long .

Click to expand...

Click to expand...

AND only half as accurate .

--

Bokeh is the aesthetic quality of the blur in out-of-focus areas of an image, or the way the lens renders out-of-focus points of light. Bokeh is not the same as depth of field (DOF).

let's wait and see what the OMG and G5/GH3 come up with.

Cheers, geoff
--
Geoffrey Heard
http://pngtimetraveller.blogspot.com/2011/10/return-to-karai-komana_31.html

Bob, i always use AF-S.

Also i use a large focus area, hoping the bird comes in this area.

Do you have other settings you can advise?

Guy.