Your Thoughts on AI Servo Image Priority

[Arthur Morris]

Would love to hear what folks think is best for bird in flight for both AI Servo 1st image priority and AI Servo 2nd image priority...

With the 5D III for 1st image priority you can set anywhere from Release Priority which tell the camera to take the image even when focus has not been achieved or Focus priority where the shutter will not release until the subject is in focus. Equal priority is the in between setting.

AI Servo 2nd image priority gives you a choice between continuous shooting speed and focusing. Or Equal priority.

On the Mark IV it is different at C.Fn III:3.

You can choose either AF Priority or Release Priority for AI Servo 1st image and either Tracking Priority or Drive Speed Priority for AI Servo 2nd image by choosing 0, 1, 2, 3.

Last question: with it being impossible to replicate even a simple bird in flight situation and thus being unable to perform any type of test has anyone ever seen any differences in performance with any of the settings above????

[arash_hazeghi]

Hey Artie,

For 5D3 I have 1st image set to AF priority and 2nd image equal priority for focus speed.

For 1D4 I have 1st image set to AF priority and 2nd image tracking priority.

The reason is I don't want the camera to fire off when it has not locked. These settings have given me more consistent results overall but I have no way to quantify by how much or in which condition. As you pointed out given the erratic nature of subject and also the skill of the photographer which is the biggest variable here, it is not possible to do a controlled test and quantify the effect of these settings in a measurable way. The only setting which I find affects AI servo performance noticeably in these cameras is the AI servo tracking sensitivity which I set to slow for most conditions.

[Roger Clark]

Would love to hear what folks think is best for bird in flight for both AI Servo 1st image priority and AI Servo 2nd image priority...

With the 5D III for 1st image priority you can set anywhere from Release Priority which tell the camera to take the image even when focus has not been achieved or Focus priority where the shutter will not release until the subject is in focus. Equal priority is the in between setting.

AI Servo 2nd image priority gives you a choice between continuous shooting speed and focusing. Or Equal priority.

On the Mark IV it is different at C.Fn III:3.

You can choose either AF Priority or Release Priority for AI Servo 1st image and either Tracking Priority or Drive Speed Priority for AI Servo 2nd image by choosing 0, 1, 2, 3.

Hi Artie,

I have my 1DIV C.Fn III:3 set to 0: AF priority/Tracking priority. I often find if I am acquiring a moving subject, the first frame is sometimes just slightly out of focus, but each subsequent frame is nailed (if I have the AF point properly placed). I would only set to drive speed priority for slow moving or stationary subjects that are doing something fast, but I'm not sure if there is any real difference in fps. For BIF it seems to me that AF accuracy and tracking are critical compared to speed. So setting to drive speed and/or release seems counter to what BIF needs. I remember playing with these settings with my dogs fetching a ball when I first got the camera, and mode 0 (AF priority/Tracking priority) made for better keeper ratio. With the setting of 0, the first frame is usually still sharp for web sized images, just not for a big print.

Last question: with it being impossible to replicate even a simple bird in flight situation and thus being unable to perform any type of test has anyone ever seen any differences in performance with any of the settings above????

I think it is possible and have plans to construct a system so I can test things like this. There seems to be a lot of perceptions and myths about AF. I've just been too busy the last few years with 3 active spacecraft missions, and with what free time I have I like to get outdoors in the field. Perhaps I'll get to it in the next year.

Roger

[Karl Egressy]

My setting on 1D Mark IV is exactly the same described by Arash.
It works for me most of the time.
I noticed that if I shoot BIF all day, two days in a row, like on Bonaventure Island, I hit the head of the bird 70 percent of the time.
If I shoot rarely my rate of keepers goes down to 5-25 percent.
Based on this I would add to the skill of photographer one more factor: how much BIF shooting you are doing at any given time.

[D. Robert Franz]

I currently have my mk4 set at Release/tracking priority. This setting seems to be beneficial for my wildlife photography where at times a quick grab shot is vitally important. I used to get frustrated when the camera wouldn't fire when I had a great opportunitiy such as a jumping image. I was then set at AF priority/tracking priority. Now at least I have a fighting chance and often have success capturing a sharp 1st image this way. Resuts may vary with BIF but I sure haven't noticed any major problems with BIF set up this way.

[Ofer Levy]

Like Robert I am also using Release/Tracking Priority with my Mark IV and I feel it is fantastic for BIF and action.

[Arthur Morris]

Thanks all for trying to help. Please confirm that for the Mark IV Ofer and Bob Franz are setting 1st image to Release Priority while Arash and Karl as setting it to AF priority.... In other words, they are using diametrically opposite settings and each is stating that it works great. I am just trying to make sure that I am understanding the nomenclature successfully.

[Arthur Morris]

Hi Artie,

I have my 1DIV C.Fn III:3 set to 0: AF priority/Tracking priority. I often find if I am acquiring a moving subject, the first frame is sometimes just slightly out of focus, but each subsequent frame is nailed (if I have the AF point properly placed). I would only set to drive speed priority for slow moving or stationary subjects that are doing something fast, but I'm not sure if there is any real difference in fps. For BIF it seems to me that AF accuracy and tracking are critical compared to speed. So setting to drive speed and/or release seems counter to what BIF needs. I remember playing with these settings with my dogs fetching a ball when I first got the camera, and mode 0 (AF priority/Tracking priority) made for better keeper ratio. With the setting of 0, the first frame is usually still sharp for web sized images, just not for a big print.




I think it is possible and have plans to construct a system so I can test things like this. There seems to be a lot of perceptions and myths about AF. I've just been too busy the last few years with 3 active spacecraft missions, and with what free time I have I like to get outdoors in the field. Perhaps I'll get to it in the next year.

Roger

Thanks Roger. I am gonna give AF/Priority/Tracking Priority a go on my MIV and in the same vein Focus and Focus with my 5D MIII.

As for simulating a bird in flight for testing it is just not possible. Even with a trained bird flying from A to B there are simply too, too many variables.

[Arthur Morris]

Thanks again to all. As I mentioned in my response to Roger I am going to give AF Priority/Tracking Priority a go on my MIV and in the same vein Focus and Focus with my 5D MIII. After all, it would not look good to agree with Ofer in public :). Actually, I will report back if I note any differences. I formerly went with Release/Tracking Priority with the Mark IV and with Release and Speed on the 5D III. I suspect that I will not see much difference in either case.

[Roger Clark]

Thanks Roger. I am gonna give AF/Priority/Tracking Priority a go on my MIV and in the same vein Focus and Focus with my 5D MIII.

As for simulating a bird in flight for testing it is just not possible. Even with a trained bird flying from A to B there are simply too, too many variables.

Artie, I agree that using a live bird creates too many variables. That is why a mechanical setup where, for example, a model bird on a wire where the track and velocity can be precisely controlled is the only practical way to test. That way different backgrounds can be tested and different birds, from black birds to white birds and different speeds and angles to the camera.

Roger

[Arthur Morris]

Artie, I agree that using a live bird creates too many variables. That is why a mechanical setup where, for example, a model bird on a wire where the track and velocity can be precisely controlled is the only practical way to test. That way different backgrounds can be tested and different birds, from black birds to white birds and different speeds and angles to the camera. Roger

As far as I can figure, no mechanical bird can replicate the patterns of flight, and then throw in upstroke/downstroke variables and add in the the human variables that include those of the timing of acquiring focus and pressing the shutter button and I am pretty sure that you would be grasping at straws but do feel free to prove me wrong :).