Can I improve my keeper rate here?

[Julian Mole]

Hi Everyone,

I have been practicing my flight photography recently but really find it frustrating at times! I love my Canon 40D but BIF really pushes it to it's limits, worst of all is when i know I have done my bit right (orienting myself to the light, pre-focusing, picking up and tracking the bird, bumping the focus regularly and waiting for the decisive moment to press the shutter) and still the image is out of focus!

This image features a bird flying straight at the camera, full frame, which is when I experience most misses. Is there anything i could do to improve results with my current set up?


The techs here were:

Shutter speed - 1/1250 sec
Aperture - 6.3
ISO - 400

Canon 40D and 400mm 5.6 lens (focus limit set to 8.5m to infinity)


Julian.

136242

[Julian Mole]

Here is a screen capture, Artie style, to show the camera was focusing on the right place! (my original RAW capture was under-exposed by 2/3 stop)

136243

[John Chardine]

Hi Julian- I assume when you say "out of focus" that the image is soft at the focus point. There are many reasons for a soft image, focus only being one of them. It would be helpful to see a 100% crop of this image centred on the focus point. Don't run any sharpening during raw processing or after. As the image is dark, lightening it some would help in the diagnosis. Perhaps you are getting soft images because the camera/lens combo is back or front focusing? If so I don't think the 40D allows AF fine-tuning.

[Julian Mole]

Hi John,

Happy New Year, thank you for your input.

Here is a 100% crop as requested, this is from the DPP conversion with a new white point set in Photoshop.

It is closer to focus towards the tail, but nothing is quite 'in focus'! Given the crop body and close distance to the subject perhaps a slightly higher shutter speed would help as well?


Regards,

Julian.

136246

[WIlliam Maroldo]

I have two guesses of what caused this. In the first place the shutter-speed is a bit slow; with BIF I find a minimum of 1/2000 sec or faster is best. Also there is a very short lag between when focus is acquired and the shutter released that becomes a problem when the subject is moving toward the camera. This would explain why the tail seems more in focus. regards~Bill

[John Chardine]

Quick question- is it possible the bird was closer than 8.5m?

[Diane Miller]

It does look like you might have some back focus, but hard to tell from here.

I see you are using AI Servo. You might try using back button focus, holding the focus button down continuously and then hitting the shutter when everything seems stable. Also, the image shows one focus point, but did you only have one selected? That would save the camera trying to make a decision on which one to use.

You could just be up against the camera's lag time, as William said. Shooting BIF makes big demands on many cameras.

[arash_hazeghi]

Hi Julian, it seems to me that you did everything right on your part. Was this the first frame in the sequence? The first frame is always more difficult, because the camera has not been tracking the bird so it hasn't calculated its speed and direction yet. Also incoming bird is going to be a challenge for the EOS 40D as its AF system is not really capable of nailing this kind of shot consistently, it is possible to get it sharp but you have to try many times to get one keeper file and there will be disappointment. I suggest you focus your efforts more on banking shots where the bird is flying parallel to the back of the camera, that type of shot (dorsal view) works better for a gull also from an aesthetics point of view.

Good luck

[Julian Mole]

Hi folks,

Thank you for your help and suggestions.

William your point about shutter speed and shutter lag are good ones. My shutter speed was probably border line with the bird coming straight at me at close range (i have made sharp ones at this ss but everything has to line up), shutter lag is probably a significant factor.

John, good question! However, I am pretty sure that the bird was further than 8.5m as I made images with these birds larger in the frame and still in focus that afternoon, although those were panning shots.

Diane, yes was using single point AF at centre. Back button focus is something i haven't tried yet, am currently just trying to master making consistent exposures with reasonable compositions on a regular basis! (although may give it a go in time) Yes, definitely BIF asks a lot of a camera!

Arash, thank you, yes it was the first frame in a burst of 2 shots. I try to take 2 to 3 shots at the right point, bumping the focus regularly until then. Yes, i definitely get more in focus shots with panning type BIF shots. Straight on shots are more consistent with birds which are slightly further away, annoyingly! (Although full framers are possible, see below)
Your thoughts are helpful though, I will just have to work that bit harder for my shots. :-)

Happy New Year to all! :-)

Julian.

136251

[Roger Clark]

Also there is a very short lag between when focus is acquired and the shutter released that becomes a problem when the subject is moving toward the camera. This would explain why the tail seems more in focus. regards~Bill

With AI servo, also called predictive autofocus, the camera measures the velocity of the subject, and predicts where the focus will be at the time the shutter actually opens, taking into account the lag from raising the mirror. But to do this, the AF system needs to be tracking the subject for at least a short time. If the first frame is made quickly after the AF system is started, the system may not have had enough time to track the subject to establish the velocity, in which case the first frame may be out of focus. Thus, it is better to be tracking the subject for at least a fraction of a second, rather than wait and fire rapidly from no tracking.

Also, if the subject is accelerating, the AF system on most cameras will have tracking errors. From my understanding, only the 1DX has parameters in the AF system for accelerating subjects. Cameras except the 1DX only have linear velocity tracking prediction, meaning after a couple of measurements, the predict projection for the focal point assumes a linear velocity projection. If the subject is accelerating, the change in velocity translates to a predict error and the subject will be not at the predicted focus. Cameras with shorter shutter lag (which generally means faster frame rates) will have less error in this regard simply because the predict projection is over a smaller time interval.

Roger

[David Stephens]

Equipment is an issue here. Given the technique that you're showing, I suspect that if someone handed you a 1D X, your keeper rate would more than double, with no change in your technique. Between the 40D and the 1D X, there are lots of increments. A used 1D MkIV would be give you a large percentage improvement. A 5D MkIII would also improve your keeper rate. Your 400/5.6 is a sharp, fast lens and will serve you well on the very best bodies.

As suggested, raise your SS a bit more and track and shoot burst more (within the limits of the 40D). Those will give you small, incremental improvement; however, to experience substantial improvement in your BIF keeper rate, I'm afraid that you'll need to upgrade your camera body.

[Julian Mole]

Hi Roger,

Thank you for that extra information, helps me better understand what's happening here.

Regards,

Julian.

[Julian Mole]

Hi David,

Thanks for the encouragement, I will keep at it and experiment a bit and hopefully the Gary Player mantra* will come into play!

Cheers,

Julian.

*The more I practice the luckier I get!

[Julian Mole]

Hi Diane,

With a lovely sunny day here I decided to try out back button focus like you suggested, ...and I love it! I didn't think it would make that big a difference, but I found it really worked for me. :-)

I found it much easier to track a subject and hold focus, even with a busy background and I managed to get back on it easier if I lost it. Am very pleased, even though it wasn't exactly the same situation as in the OP I felt I got better results than I'm used to.

Here are my best two shots. :-)

136598

[Julian Mole]

And this one is a shot I've been trying to nail for a while, although I needed a slightly higher shutter speed than the 1/1600 sec I used to totally freeze the bird. (PS. Both are full frame images)

136599

[Diane Miller]

Lots of great information here! And great results in these last two shots! Practice is key...

Looking forward to your future shots!

[Julian Mole]

Thanks Diane. :-)