Focus masking

[purplehaze]

Capture One provides a focus masking option, but I have never found it to be very accurate and I am wondering if someone can shed some light on why there is such a discrepancy between what my eyes tell me is in focus (or reasonably so) and what Capture One thinks is. Today I am looking at images where Capture One has masked a clearly OOF background. Is it because that background is featureless? Is it because C1 has still not profiled this particular lens? Is it because my eyes are lying to me? How does this focus masking work? I have been taking full advantage of the D500's high frame rate and so would really like to be able to rely on this tool for rapid culling.

[billtils]

I've never used it Janis, but I may shoot in a different way to you. I normally have less than 10 shots in a burst, on the basis that the D500's excellent AF will do the trick and the reason for the burst is to provide a range of "poses" - wings up or down for example. I either have a burst where I got the lock on right and all are in focus or didn't get it right and they're all immediately for the bin.

EDIT to clarify: As an example I don't focus on a bird in flight and fire a continuous burst while I follow the bird for, say, 6 to 10 seconds. I follow it (using BBF) but only fire when it looks "right" and then only for half a dozen shots; keep following it and repeat as required. That makes things easier at the editing stage - either it's a burst with interesting shapes and is kept, or they're not and isn't. The ony time I've had to dump interesting shapes was in the Gannet shoot I posted recently where some of the action was too close to the boat and it was more a depth of field issue than an absolute focus problem - see "It's mine!".

One thing to check - did you set it up as in the User Guide, especially in regard to the threshold slider?

[Manfred M]

Janis - to the best of my recollection, the distance to the subject, i.e. where the focus plane is is recorded in by the camera and written to the EXIF (metadata). Knowing the aperture and focal length you are shooting at, some assumptions can be made as to what is in / not in focus. Algorithms that are similar to contrast detect that are used in-camera autofocus programs could also be applied.

That being said, Nikon's View NX2, which shipped with my D90 had that functionality and when I played with it back then, I found it to be rather imprecise and was not at all surprised when other raw convertors did not have this functionality. It seems that Phase One has decided to introduce it.

[purplehaze]

Thanks, Bill and Manfred. Based on comments I have since found in the C1 community support forum, I gather that the tool just isn't very good. Good as the D500's AF is, it sometimes fails in low contrast, narrow DOF situations, like the one I was shooting in yesterday morning. I was trying to capture the sparrows splashing about in the bird bath and was sacrificing DOF for shutter speed and ISO in what were shady conditions. The solution is likely just to wait until the sun hits the bird bath and I can use a narrower aperture.

[billtils]

Janis - there's somethng missing in your analysis: what was your ISO?

It was dull when I took my gannet shots. I'm still playing/learning with the D500 and all were shot on manual shutter and aperture and auto ISO. Most came in around F/6.3 and 1/2000s with ISO in the range 1250 to 5000. All worked well, other than ISO noise which NIK Define + Control Points dealt with reasonabley well.


Bill

[Manfred M]

Janis - I know several excellent bird photographers and they shoot with cameras that handled high ISO extremely well and they often set the ISO to at or near the highest that their cameras can handled. Much like what Bill has suggested, it is something that they can always handle in post.

The other thing that they usually do is to pre-focus their camera and wait for the bird to get into the zone where they have pre-focused to take the shot. Depth of field will be selected based on getting the maximum they can afford to at the shutter speed / ISO they have chosen.

[purplehaze]

I was shooting manual with auto ISO. Yes, I am quite impressed with some of the high ISO results I get from the D500, but my choices were quite deliberate based on what I was trying to achieve with these shots, which were going to have to be cropped somewhat and in which I wanted maximum quality of detail (wet feathers, water drops flying, etc.). My question was about the focus masking tool, which I personally would find useful for rapidly sorting through multiple shots that otherwise vary little, and is advertised as being for that exact purpose, but does not in fact actually work.

[billtils]

The other thing that they usually do is to pre-focus their camera and wait for the bird to get into the zone where they have pre-focused to take the shot. Depth of field will be selected based on getting the maximum they can afford to at the shutter speed / ISO they have chosen.

Yes I've seen that but don't see how it can work reliably with fast moving objects like birds ..

[Manfred M]

Yes I've seen that but don't see how it can work reliably with fast moving objects like birds ..

It was pointed out to me when I asked that question that if one positions themselves where the birds are flying in a direction that is roughly parallel to the sensor plane, the technique works well. The other comment was that they don't fly all that fast just after takeoff or when coming in for a landing.