[slip]

this is a broad question, but on average what is the percentage of keeper rates as far as sharp as the lens can perform? (stationary subjects)

with my Sigma 70-200 2.8 the "keeper" rate is below 50% with the exceptionally sharp rate in the 20-30% range.
what are others autofocus performance rate on average?
yes, I do know it will be lens specific but I am looking for an overall average for the K3

thanks

randy

[jatrax]

With the k-5 my hit rate was around 50% give or take. On the k-3 it approaches 100%. Simply bordering on ridiculous.

That is with a variety of glass but mostly the 16-85 or dfa 100mm.

And that is stationary subjects not tracking.

[Quartermaster James]

I may not understand this question properly, but I rarely have non-keepers due to autofocus issues.

How are you shooting?

[victormeldrew]

With my K3II and the lenses I have, I'd say nearly 100% unless I myself mess up. With the tammy 70-200 at long focal lengths and wide open it's sometimes hard to hit the right spot and the leeway is very small (focus on a dog's nose and it's eyes will be soft) but again that's not really an autofocus issue, more an aiming issue.

[Venom3300]

On my k5 with stationary stuff it was maybe 60-70%

My k-3ii approaches 100. I miss more than it does.

[stevebrot]

I don't use AF for fine focus, though I am generally happy with the keeper rate when I do use AF. That is not a direct answer to your question, but does indicate how I use the feature and my expectations when I do.


Steve

[slip]

I may not understand this question properly, but I rarely have non-keepers due to autofocus issues.

How are you shooting?

to test: lock focus on a distant subject, then lock focus on intended target, then fire. repeat many times and find out the keeper rate.

thanks

Randy

[Quartermaster James]

to test: lock focus on a distant subject, then lock focus on intended target, then fire. repeat many times and find out the keeper rate.

thanks

Randy

Okay, now I am really confused: are you talking about real world, actual use results or a rate you get when putting the camera through some sort of drill?

[hcc]

As stated by others, the AF of the K-3 is really good and your should not loose may shots because of AF issues.

It may be an issue with the lens itself. Consider to update the firmware. (Could you use the Sigma USB dock ?)

Do you use the lens Sigma OS or the in-camera Pentax SR? One should be switched off (they do not work together.)

Also what AF focal point(s) do you use? Using centre point is strongly recommended.

I hope that the comment may help.

[slip]

Okay, now I am really confused: are you talking about real world, actual use results or a rate you get when putting the camera through some sort of drill?

kind of both. today I was shooting a chipmunk for practice and noticed when I locked focus and fired (sometimes several times) that some where front focusing, some back focusing and others (be it lower rate) focus perfectly. so instead of loving focus and firing away, thus getting a group of them unfocused, I decided the best way to figure out the success rate would be to do what I described above

randy

---------- Post added 05-06-16 at 07:40 PM ----------

As stated by others, the AF of the K-3 is really good and your should not loose may shots because of AF issues.

It may be an issue with the lens itself. Consider to update the firmware. (Could you use the Sigma USB dock ?)

Do you use the lens Sigma OS or the in-camera Pentax SR? One should be switched off (they do not work together.)

Also what AF focal point(s) do you use? Using centre point is strongly recommended.

I hope that the comment may help.

the lens is the DG model thus to old to upgrade the firmware. I have been using center point only

[clackers]

kind of both. today I was shooting a chipmunk for practice and noticed when I locked focus and fired (sometimes several times) that some where front focusing, some back focusing and others (be it lower rate) focus perfectly. so instead of loving focus and firing away, thus getting a group of them unfocused, I decided the best way to figure out the success rate would be to do what I described above

randy

---------- Post added 05-06-16 at 07:40 PM ----------


the lens is the DG model thus to old to upgrade the firmware. I have been using center point only

Have you chosen a FL and done the microadjustment?

[slip]

Have you chosen a FL and done the microadjustment?

yes but they don't seem to be consistent results

Randy

[C_Jones]

If you are using center point, continuous AF mode, a reasonable depth of field for your subject, the subject is stationary and the subject has reasonable contrast, I do not see a reason why you would have an AF problem.

If your other lenses do not have difficulty with the camera, you may want to test the lens on another camera and see if it behaves similarly.

You may want to check to see if you have your camera SR on, and not both the lens OS (if it has it) and the camera SR on at the same time.

[stevebrot]

some where front focusing, some back focusing and others (be it lower rate) focus perfectly.

yes but they don't seem to be consistent results

Some front, some back, some just right? It sounds like you are getting a real-world lesson in precision* (or lack thereof). What you are seeing is what you should expect when the camera is having a tough time of it or when your standards are higher than what the system can produce.

The base precision at moderate distance with a sharp f/2.8 lens using the center focus point on a high contrast (dark black bordering bright white) subject is pretty good. Increase the distance, soften the lens, and/or substitute a low contrast subject and attaining focus becomes more and more a guessing game for the camera. At worst, it simply gives up. At about 150m with a 200mm lens, my K-3 can be off +/- 7-10m easily; assuming, of course, that the focus mechanism for the lens is able to move in that small an increment.

That is why AF fine adjustment should be done at moderate distance with a high contrast target and should involve multiple focus attempts. (I usually suggest a minimum of 10 with 20 being more acceptable...half focusing from near and half from far.)


Steve

* The tendency in everyday talk is to treat precision and accuracy as meaning the same thing, but they aren't. A measurement system can be accurately calibrated, but provide unacceptable measurements most of the time due to poor precision (the probability of getting the same answer twice in a row).

[Quartermaster James]

kind of both. today I was shooting a chipmunk for practice and noticed when I locked focus and fired (sometimes several times) that some where front focusing, some back focusing and others (be it lower rate) focus perfectly.

Okay. Thanks for explaining that.