Victor Engel
Forum Pro
Have you tried comparing with IS turned off?
-
Welcome to the new forums! Please read this first. For known issues we are working to resolve, click here.
Anyone having alot of weird slightly out of focus shots with R7
- Thread starter K A Wray
- Start date
axlotl
Senior Member
This morning I photographed the same birds on the same perch. Put the RF 24-240 on the RP and focussed the old fashioned way placing the focus box on the birds eye. The RP has by current standards an unsophisticated AF system but apart from the frames where the bird moved all were sharp.
Andrew
Andrew
BirdShooter7
Forum Pro
I can report that I have seen similar behavior with my R7 with both adapted EF lenses as well as native RF lenses (100-400, 100-500L and 800mm f/11). However, it isn’t exclusive to the R7, I’ve also seen plenty of this behavior from the R6 and R5. My feeling is that this is just a limitation of Canon’s AF at the moment.
[/QUOTE]
BirdShooter7
Forum Pro
Sounds good, if you use the single point AF with no eye detection on the R7 what are your results?
--
Some of my bird photos can be viewed here: https://www.flickr.com/photos/gregsbirds/
BirdShooter7
Forum Pro
Interesting, so far I’m not really seeing what I can say is systematically different behavior from the R5, R6 and R7. I’m not really sure how you can do a real controlled test. It’s still my feeling that what we are seeing here are limitations of Canon’s eye-detection AF but I’m definitely open to the idea that there could be something specific to the R7 that makes it even worse. Maybe it could be related to the slower readout speed of the R7 sensor?
My suggestion is to maybe get one of those Atamos recorders and record the behavior you are seeing in real time and send those to Canon to see what they say.
For sure I’ve been very frustrated with every R camera so far when it comes to AF performance in backlit situations. I’d love to see some improvement there.
I can safely say it is not heat waves, all my shoots are early morning starting around sunrise. It also occurs taking pictures of birds in water that is still very cool from night before. It might have something to do with IBIS. I will try shooting with no IS at all when able.
Looks pretty sunny in your posted shots. And it doesn’t have to feel hot or even warm for heat distortion to occur. All it takes is a difference between the surface temperature and the air temperature. I’m not saying that the problem is caused by heat distortion. Just saying one might be too quick to rule out other possibilities because of antecedent beliefs.
Yes, many of the bad pictures are showing non-ciircular images of the sun in their eyes. Too bad I dont think IBIS can be turned off by itself and leave lens is on, to rule it out. Will have to break out tripod and turn them both off for tests. Attached pics of zoomed in sun in eye pic. This was from a series of shots of a bird super close not moving. Easy shot that is near imposable to miss.
Victor Engel
Forum Pro
Since I can see the reflection of the sun in the eyes, I'm not so sure you can safely say that. If the sun is hitting the bird's eye, it is also hitting the sand and water. Also, there could be a residual temperature differential from overnight cooling. Probably the sand is warmer than the water during the day, but the water is probably warmer than the sand at night. Additionally, there is likely to be a humidity differential.
R2D2
Forum Pro
Yeah those mornings when I see mist coming off the water I know I’m in trouble. Ironically the mist creates some of the most pleasing compositions, but for detail it can really s#ck.
R2
Adam2, your experience seems very similar to mine.
Again, the R7 performs very well most of the time, though I am noticing this "pulsing" of the af where in some sequences the first few images will be in focus followed by OOF followed by more in focus. If I could characterize it better, I would. These are very complex systems with IBIS, lens communication, closed loop autofocusing, AI, etc. and I'm not surprised that some kinks would manifest.
In another example, I was shooting a sandpiper under similar light, temperature, etc. side-by-side with the R5. The bird was front lit and R5 nailed the eye af in every sequence. The R7 appeared to nail the af in the EVF as evidenced by the blue eye AF indicator, however looking at the sequences back on the computer revealed a "pulsating" af where the first few frames of the sequence were in focus, the next several were severely BF'ed (in spite of the blue box squarely on the eye), and then the final few were in focus again. I didn't move and the bird didn't move appreciably towards or away from the camera. If the lens is pulsating on the R7, it is imperceptible in the EVF during the short sequence.
Again, the R7 performs very well most of the time, though I am noticing this "pulsing" of the af where in some sequences the first few images will be in focus followed by OOF followed by more in focus. If I could characterize it better, I would. These are very complex systems with IBIS, lens communication, closed loop autofocusing, AI, etc. and I'm not surprised that some kinks would manifest.
axlotl
Senior Member
Good question. So today I put about 600 frames through the R7 and about 400 through the R10, with Subject tracking OFF, various shutter speeds and types (need about 1/500s or faster to avoid shutter shock with continuous drive, even low speed continuous if EFC is used, there is no shutter shock with E Shutter) .
I photographed people, insects, flowers, birds, books on a shelf, foliage.
Anyway the main finding is that AF with essentially static subjects is more reliable with Subject tracking OFF. Still not perfect but that disconcerting focus drifting back and forth which happens with subject tracking ON is no longer present.
Which presents the R7/R10 user with a dilemma given that subject tracking is one of the selling points of these cameras.
Andrew
BirdShooter7
Forum Pro
Are you getting similar results on overcast days?Yes, many of the bad pictures are showing non-ciircular images of the sun in their eyes. Too bad I dont think IBIS can be turned off by itself and leave lens is on, to rule it out. Will have to break out tripod and turn them both off for tests. Attached pics of zoomed in sun in eye pic. This was from a series of shots of a bird super close not moving. Easy shot that is near imposable to miss.
--
Some of my bird photos can be viewed here: https://www.flickr.com/photos/gregsbirds/
BirdShooter7
Forum Pro
I have also had similar findings. Less focus excursions when subject detection is turned off. However, this has also been the case with the R6 and R7 for me. It can be quite frustrating and I’d love to see improvement in this area. I can’t say for sure if it happens more frequently with the R7 or not and I really don’t know how to test it in a controlled way.
Maybe reserve subject tracking for situations where one intends to track a moving subject?
Mike Engles
Senior Member
I have had ficus problems with my R5 and adapted EF lenses, especially when I use my EF100-400 2 with my x1.4 TC111.Also with my 100mm IS macro.
My TC had its internal glass replaced by Canon, but the problems still persists
My TC had its internal glass replaced by Canon, but the problems still persists
Mike Engles
Senior Member
"You will get plenty of naysayers who don't want this to be true".
It is very possible that adapted EF lenses are not as compatible with the R5 and newer cameras as the should be.
I have had eleven months of the naysayers, whose total belief is that mirror less cameras cannot have focus problems. The camera might not need calibrating, but the lenses might. My R5 has been to Canon 4 times and it still has problems with my 100-400 2,my x1.4 111 TC and my 100mm IS macro.
Pretty well all pictures are just slightly off, even if the AF focus spot is on the eye.
No heat haze, I was about a meter away
It is very possible that adapted EF lenses are not as compatible with the R5 and newer cameras as the should be.
I have had eleven months of the naysayers, whose total belief is that mirror less cameras cannot have focus problems. The camera might not need calibrating, but the lenses might. My R5 has been to Canon 4 times and it still has problems with my 100-400 2,my x1.4 111 TC and my 100mm IS macro.
Pretty well all pictures are just slightly off, even if the AF focus spot is on the eye.
No heat haze, I was about a meter away
Last edited:
Victor Engel
Forum Pro
It's beginning to sound like an issue with the algorithm(s) used. Maybe a faulty assumption when doing interpolation or an unhandled positive feedback loop. Just wild speculation here.
Victor Engel
Forum Pro
Your photo example appears to be in perfect focus. Look at the square indicating where the focus was taken. Now, within that square find the highest contrast fine detail. It is the whiskers in front of the eye. That is in focus. Could that be the issue here?
Of course, then the question is whether that is the right thing for the AF system to do. Most people would say no, but how is the camera to know not to look at what otherwise would be the best thing to focus on?
Of course, then the question is whether that is the right thing for the AF system to do. Most people would say no, but how is the camera to know not to look at what otherwise would be the best thing to focus on?
Keyboard shortcuts
- f
- Forum