Autofocus Stress Test: Can Canon’s Dual Pixel AF II

[Canon Rumors]

While Canon’s Dual Pixel AF II system is known for accuracy and speed, its real-world performance in challenging environments rarely gets a true stress test. I put this innovative system through the most brutal focusing scenarios—low contrast, fast motion, foreground interference, and edge-of-frame compositions—to see if it lived up to the hype.

Read the Full Article

 

[P-visie]

The RF 70-200mm is an “unusual” choice for birds in flight photography. I doubt the birds in flight test with a RF 70-200mm is representative of the actual performance of the R5 Mk II for birds in flight. How many pictures are taken with a 70-200mm in the “Birds in flight” section of the forum?

 

[R1-7D]

Another ridiculous “click-bait” article. What’s with all these “battle of the titans” / “show down” pieces lately? They’re horrible and all read as if put together by ChatGPT.

 

[wbrock001]

The RF 70-200mm is an “unusual” choice for birds in flight photography. I doubt the birds in flight test with a RF 70-200mm is representative of the actual performance of the R5 Mk II for birds in flight. How many pictures are taken with a 70-200mm in the “Birds in flight” section of the forum?

Agreed, have a sports photographer shoot wildlife and that’s what you get!

 

[Lando Lowndes]

Unfortunately this autofocus stress test has a major omission: camera firmware version. Especially when autofocus with the current version 1.0.3 is considered to be inferior to 1.0.2 under certain circumstances.

 

[AlanF]

The RF 70-200mm is an “unusual” choice for birds in flight photography. I doubt the birds in flight test with a RF 70-200mm is representative of the actual performance of the R5 Mk II for birds in flight. How many pictures are taken with a 70-200mm in the “Birds in flight” section of the forum?

The major difference between her tests on still subjects, sports and the birds in flight is the size of the subject in the image. The birds are so small that she needs to use a longer lens to have more pixels on the subject. I would be using the RF 200-800mm at that distance or at least the RF 100-500mm.

 

[AlanF]

Unfortunately this autofocus stress test has a major omission: camera firmware version. Especially when autofocus with the current version 1.0.3 is considered to be inferior to 1.0.2 under certain circumstances.

That is not universally true. The AF of my R5ii, like that of several others reporting here, works just as well with 1.0.3 as with 1.0.2.

 

[justaCanonuser]

While Canon’s Dual Pixel AF II system is known for accuracy and speed, its real-world performance in challenging environments rarely gets a true stress test. I put this innovative system through the most brutal focusing scenarios—low contrast, fast motion, foreground interference, and edge-of-frame compositions—to see if it lived up to the hype.

Read the Full Article

Regarding the BIF test: my experience with the R7 (not sure what DPAF system) is that the bird needs to cover more than a few pixels to get safely recognized by the AF system in front of a busy background. Could well be happened here, 200mm is very short for BIF.

 

[wyotex43n]

While I regard the autofocus on the new camers to amazing I still see a few problems. The biggest issue I personally see with the autofocus on the r5mk2 is it will pick up a bird in the sky and track very well until the bird crosses in front of a background. This is especially true when the bird is less than a quater of the screen. I have tried a number of different settings. If anyone has a setting they think works well I would be glad to hear them. I know this is an extreme case but things can always be better.

Also I have not personally experienced the autofocus issues others have but I do have on anomaly that happens and wondering if others have seen this. 1 out of maybe 15 times when I turn on the camera it shows an almost white screen. LIke an extreme overexposed view of what you are pointing it at. As if you the exposure is frozen in an overexposed state. I have to turn it off and on to get it to act normally.

 

[wbrock001]

I definitely noticed a lot of hunting and not tracking ideally with the 1.0.3, but I did what Ordinary Filmmaker suggested by turning off sleep and seemed to make a difference as I shoot a lot of events, long head shot sessions, etc so camera would normally go to sleep and wake up drunk! ha ha. So yes, I'm chomping at the bit for a REALLY GOOD firmware update and would like to see sensor shift high resolution return. Not that I used it much but its much better than AI uprez.

 

[Birdshooter]

The article is a joke..

I am shooting falcons going in an out of a metal bridge and the auto focus sticks like glue, and I have the new firmware 1.10 installed and no issues tracking fast birds. Zero, and with obstructions I have no issues. I can tell you I only use 1 single auto focusing point. Spot the tiny one that most people say is the hardest to use. I laugh inside when I hear that as I have been using single point since the 40D came out to shoot birds in flight.

Here is a special little peregrine falcon, the runt of the litter and one like no other I have seen. A very dark bird with a dark hood so I try and single this bird out for shots as it's different, and boy oh boy is this little falcon a speedster.

This should be a good example of shooting with obstructions and staying on the bird.

These images were take with a 600mm f/4 Hand held with the R5II and firmware 1.10
And I slowed the gif down a whole lot so you can see the obstructions had no effect on tracking the bird.

Cheers

 

[Appelflapje]

The article is a joke..

I am shooting falcons going in an out of a metal bridge and the auto focus sticks like glue, and I have the new firmware 1.10 installed and no issues tracking fast birds. Zero, and with obstructions I have no issues. I can tell you I only use 1 single auto focusing point. Spot the tiny one that most people say is the hardest to use. I laugh inside when I hear that as I have been using single point since the 40D came out to shoot birds in flight.

Here is a special little peregrine falcon, the runt of the litter and one like no other I have seen. A very dark bird with a dark hood so I try and single this bird out for shots as it's different, and boy oh boy is this little falcon a speedster.

This should be a good example of shooting with obstructions and staying on the bird.

These images were take with a 600mm f/4 Hand held with the R5II and firmware 1.10
And I slowed the gif down a whole lot so you can see the obstructions had no effect on tracking the bird.

Cheers

Hi, amazing shots! Could you elaborate on your settings? 1. Area spot (The smallest one with the 2 inner squares) or 1-point (shows up as only 1 square)? 2. Locked or not? 3. Subject detection animals? 4. Eye detection auto or off? 5. Whole are tracking servo Af=on or off? (I assume you use Back button focussing?) I would be gratefull if you could help me out.

 

[P-visie]

Hi, amazing shots! Could you elaborate on your settings? 1. Area spot (The smallest one with the 2 inner squares) or 1-point (shows up as only 1 square)? 2. Locked or not? 3. Subject detection animals? 4. Eye detection auto or off? 5. Whole are tracking servo Af=on or off? (I assume you use Back button focussing?) I would be gratefull if you could help me out.

This question is answered by @Birdshooter in another thread: 'Is Your Canon EOS R5 Mark II Autofocus Affected by the Latest Firmware?'
https://www.canonrumors.com/forum/t...ted-by-the-latest-firmware.44491/post-1028861

 

[snapshot]

i have found R52 auto focus in tennis from the back fence to be very good, faces are identified at 90 ft / 200mm focal length. tracking / servo is great as the precapture. subject tracking through the view finder during HS+ shooting is glitch free. i would say the electronic shutter read out speed is a little on the slow side as i can perceive distorted balls/rackets.