kazman442 wrote:
We noticed the focus system seems real sensitive and seemed to jump from one thing to another real quick like when we tracked the dog running down the ramp to jump the person would throw a toy in the water and it would splash the tracking would jump righ to the water splash instantly. Ron & Chad
This is also what I'm seeing, that the AI Servo seems "hyper" active. Adjusting CF III-2 to slow might help keep the focus on the dogs when the water splashes in front of them, but I've found so far that the Mark III seems to be quick to want to refocus, perhaps too quick. AI Servo AF appears to be much more sensitive than the N, or at least tuned in this way. For the time being I'm undecided whether it's a good thing. Many sequences I'm seeing from my camera have soft frames from a few different sports I've covered so far. Part of the issue may be AF point placement, but there are certainly soft frames that are not due to user error.
You post some appealingly sharp images here, but on the whole, how many soft images are you getting? Better than you feel you get with the N, or worse? Did you shoot any 5-8 frame bursts and if so, how do you feel the Mark III is in maintaining focus lock?
HI, I fell that I have gotten more OOF frames on the mark3 then with my m2n, This may be in part because it's a new camera and I need to learn it's AF more. The other part may be that it seems to get confused on say a brown dog coming at me. I set the response time down a notch and next time I will see how that reacts. Of the dog's I shot in the water jump I had about 60-70% keepers, Some had the splash focused on in front of the dog (I posted a few of those in previous post) this may have been helped like you said adjusting the time it sees a new focus point to a slower setting, I felt a few of the rest had to do with lower contrast dogs and it getting the background instead. Ron
Where I'm seeing the most AF issues with the Mark III so far is in lower contrast situations. And I don't mean dark situations, but sunny conditions with backlit subjects. It's how I prefer to photograph most sports whenever possible. So far I'm not blown away with the results, though granted, I've only shot a few sports events with the camera. Focus tracking on front lit subjects seems to be less of an issue but soft frames do happen here as well.
Otherwise for non action everyday type photography, I've for the most part been very happy with the camera.
Where I'm seeing the most AF issues with the Mark III so far is in lower contrast situations. And I don't mean dark situations, but sunny conditions with backlit subjects. It's how I prefer to photograph most sports whenever possible. So far I'm not blown away with the results, though granted, I've only shot a few sports events with the camera. Focus tracking on front lit subjects seems to be less of an issue but soft frames do happen here as well.
Otherwise for non action everyday type photography, I've for the most part been very happy with the camera.
I've been taking some tracking photos of birds against an empty blue sky using the basic set of AF points and AI servo. The results are very good. The camera tracks well and virtually all the images are in focus. So, for me, that suggests the camera is competent at tracking under "ideal;" circumstances.
Well I just gave mine a test run and all I can say is WOW! Now I went from a pair of M2's to one M3 and one M2 still.
First the screen on the 3 is awesome, it is like going from a old Sony watchman to a big screen plasma or LCD. My second reaction was how many keepers I got shooting my soccer tournament this weekend. I dont think I had more than 10% out of focus and probably those were me missing my center focus point on the subject. My keep rate went way sky high using my 400 2.8 on it. Most of the ones I tossed were dups from the fast shutter. I tracked sooo fast even playing with it really would not fool it. I wish I could replace my other one too but I have to watch out flow right now as I will be shooting a couple tournaments in Europe next month.
Again jaw dropping would be another word for the results.
Have any of you used the following on the 1D3 yet? A custom function, C.Fn III-7, allows the user to set into the camera registry a correction factor for consistent auto-focus error for up 20 individual lenses.
I just rescued an EF 50mm f1.4 I’ve had for years from dustbin or trade-in fate by the above process. Previously, the only way to handle this matter would be to send body and lens in to Canon Service Center. This is what I call ‘progress’!!!
I borrowed a 500 mm f4 L IS for a shoot. Just for grins, I gave it a try. It seems that the lens required a -10 factor. It worked great !
I tested all my other lenses and they were dead on.
It seems that the longer the focal length, the more critical the calibration.
YMMV
Bill
BTW, I also have a 5D which I tested the same way, and the lenses were spot on
I had to +9 into my 300mm F4 lens, my brother and i both had to do the same thing to both of our cameras with two different 300mm F4 lenses. I would be curious to know if anyone else has a 300mm F4 and a M3 that had to do the same, if so how much?
Side note, we even had to add this to our lenses adjustment CF with the 1.4x on as well.
Thanks for the quick reply, I forgot to mention ours are IS as well. we tested them about 12 feet away, using a chart at a 45 degree angle, I should also say that they were not off much about 1/2 inch front focusing in this distance, I was surprised how much we had to move the adjustment to get it in for how much it was out. we did see a big improvement of the overall sharpness of the shots from greater distances.
phrog wrote:
I have a 300 f4 L IS, and it did not require any correction.
B
Just an FYI, there is a guy over at DPReview that posted the plans for a test device. It is simple to make and it works quite well.
Take multiple shots, defocus in between and compare results. they should always give the same results. (IS on and IS off - depending on your version of IS.)
Sound like what we did.. Do you have a link to that post? phrog wrote:
Just an FYI, there is a guy over at DPReview that posted the plans for a test device. It is simple to make and it works quite well.
Take multiple shots, defocus in between and compare results. they should always give the same results. (IS on and IS off - depending on your version of IS.)
With the III's magnified live view its easy to check focus just shooting a dollar bill taped on a wall. So far my lenses are fine, the III's AF is very impressive.
DonM2 wrote:
Have any of you used the following on the 1D3 yet? A custom function, C.Fn III-7, allows the user to set into the camera registry a correction factor for consistent auto-focus error for up 20 individual lenses.
I just rescued an EF 50mm f1.4 I’ve had for years from dustbin or trade-in fate by the above process. Previously, the only way to handle this matter would be to send body and lens in to Canon Service Center. This is what I call ‘progress’!!!
DonM
I haven't used it on all my lenses yet, but so far the 135L and 300/2.8 IS are quite correct at 0 as I have tested both of these.
A 'straight' metal tape measure at 45 degrees to the lens axis, camera/lenses mounted on a tripod with IS off, works well enough for me.
