Well crap I did it.... bye bye 6D... hello 5D3!

Scrumhalf wrote in post #16766466
Every camera has its worst feature. Not sure I would classify the AF of the 6D as its "downfall." Not even sure what that means. Sure, it is not as sophisticated as the 5D3 or the 7D. But it is better than its worst critics give it credit for, and it is perhaps not as good as its most rabid supporters tout. The truth, as in most things, is somewhere in the middle, but it can be perfectly adequate for many people who get upsold to the 5D3 even though they may not need it. There is an opportunity cost for the extra $1K premium the 5D3 demands.

Don't compare with other 1-digit bodies indended for professional use.

But compare with the xxD bodies with 9 full-cross points. Shouldn't a 1-digit body at least try to match 2-digit bodies?

pwm2 wrote in post #16766705
Don't compare with other 1-digit bodies indended for professional use.

But compare with the xxD bodies with 9 full-cross points. Shouldn't a 1-digit body at least try to match 2-digit bodies?

-3EV Center Point, best in the Canon catalog

mickeyb105 wrote in post #16766991
-3EV Center Point, best in the Canon catalog

Yes - but it should still have been complemented with "not worse than" of-center points - why worse off-center points than a 700D?

mickeyb105 wrote in post #16766991
-3EV Center Point, best in the Canon catalog

My center point can't focus on a large ivory candle. In bright light.. It has trouble with anything not contrasty.. Mark3 had no problem, tested them out side by side.

Scrumhalf wrote in post #16766483
Not disagreeing with you, but what you consider its downfall, I consider a reasonable tradeoff for the $1K price difference with the 5D3. IMO of course.

Actually I wish the difference was only $1K... It's only $1K when 6ave, BigValue, or getitidigital have their deal of the day sales every now and then (selling the 5DIII for around $2600), and they aren't authorized dealers (although most buyers are happy with them.)

HazemG wrote in post #16767007
My center point can't focus on a large ivory candle. In bright light.. It has trouble with anything not contrasty.. Mark3 had no problem, tested them out side by side.

It's quite possible due to dual-cross crosstype center point(s) on 5D3 vs standard crosstype center point on 6D.

mgk2 wrote in post #16767044
It's quite possible due to dual-cross crosstype center point(s) on 5D3 vs standard crosstype center point on 6D.

The 6D has a standard cross type AND an extra sensitive - in the center . So it's not dual cross, but not a simple single cross either. Sort of one and a half cross (Basically an x and a -, instead of an x and a +)
But I don't think that's the issue; I think even a single cross type should pick up what a dual cross type does. The second cross in those AF points is for extra sensitivity/accuracy. It may come down to the size of the AF point instead, and exactly where it's pointed. There may be tenths of an angle differences in where we point the camera, which may bring more detail under the AF point.
Am I correct, experts?

So basically If someone starts posting BIF shots done with the 6D, we could basically throw out this whole thread ?
Anyone have any of those ?

Bsmooth wrote in post #16767132
So basically If someone starts posting BIF shots done with the 6D, we could basically throw out this whole thread ?
Anyone have any of those ?

You could only do that if the set up was this:

Contrasty bird, like a heron, flying in a blue sky. You can use a white bird (egret) or a black bird (raven) also just to really push the limits.

Compare 5D3 & 6D side by side using the outer most peripheral AF point (single point) using the same lens, tripod, settings, photographer, lighting, subject distance to photographer (as much as possible), etc

Go to the computer and see what the keeper rate is for both.

I think keeper rate is the only way to definitely say if the 6D stacks up. I have no doubt that there are great BIF shots using the peripheral AF points on a 6D. The question is how consistently focused are they in comparison to the 5D3?

gabebalazs wrote in post #16767090
The 6D has a standard cross type AND an extra sensitive - in the center . So it's not dual cross, but not a simple single cross either. Sort of one and a half cross (Basically an x and a -, instead of an x and a +)
But I don't think that's the issue; I think even a single cross type should pick up what a dual cross type does. The second cross in those AF points is for extra sensitivity/accuracy. It may come down to the size of the AF point instead, and exactly where it's pointed. There may be tenths of an angle differences in where we point the camera, which may bring more detail under the AF point.
Am I correct, experts?

The precision of the AF point depends on how distinct pattern it sees. A non-cross point can have the feature 90 degrees wrong making it not seen.

A full cross can have the feature up to 45 degrees off. At 45 degrees off, the absolute precision of the AF system is lower.

A dual-cross can not have any feature more than 22.5 degrees off.

So while a non-cross point can do well, it suffers from more situations where it will not lock, or where it will lock but with more focus error, i.e. the camera will jitter a bit too near or too far. A cross and a dual-cross will always have features aligned so the camera can use it (if the feature has enough contrast and the light level is high enough) but a dual-cross will have less worst-case focus error.

A phase-detection system splits the light path into two and try to locate the same feature (contrast change) in both images. It then measures the distance between these features (contrast changes) and if there is a shift to the left or to the right to compute in which direction the lens need to change, and how much to change the focus.

If you are going to measure the distance between two lines, it's easier to measure the distance if holding the ruler at 90 degrees to the two lines. It's way harder to try to measure with the ruler at 45 degrees. In the case of a phase-detection sensor, the transition gets less distinct when the angle is off.

HazemG wrote in post #16766438
I wouldn't hold my breath buddy not interested in winning an internet debate.

You don't have to worry about that.

HazemG wrote in post #16766438
However the fact remains the one thing you're defending on the 6D is easily it's worst feature/downfall. Riddle me that Sherlock.

Nice! Haven't heard that since 6th grade!

HazemG wrote in post #16766438
"My center point can't focus on a large ivory candle. In bright light.. It has trouble with anything not contrasty.. Mark3 had no problem, tested them out side by side"

That terrible center point! Really? Congratulations, you are now officially part of "the group"!

MikeG2012 wrote in post #16767154
You could only do that if the set up was this:

Contrasty bird, like a heron, flying in a blue sky. You can use a white bird (egret) or a black bird (raven) also just to really push the limits.

Compare 5D3 & 6D side by side using the outer most peripheral AF point (single point) using the same lens, tripod, settings, photographer, lighting, subject distance to photographer (as much as possible), etc

Go to the computer and see what the keeper rate is for both.

I think keeper rate is the only way to definitely say if the 6D stacks up. I have no doubt that there are great BIF shots using the peripheral AF points on a 6D. The question is how consistently focused are they in comparison to the 5D3?

I'm a birder and the vast majority of people I know use either the center point for tracking BIF or, if the camera has it, zones.
I have a 70D, and had a couple 7Ds in the past, but I never really used just one peripheral point for BIF.
I've taken quite a few excellent BIF images with my 6D, mostly using either the center point, or all 11 points (since they take up a relatively smaller section of the VF, at least compared to a 7D).

I've posted many of them in other threads, not sure if I need to post them here. But if people want me to, I'd gladly do it.

Bsmooth wrote in post #16767132
So basically If someone starts posting BIF shots done with the 6D, we could basically throw out this whole thread ?
Anyone have any of those ?

Ouch the logic (unless it's just a joke post). One could conceivably travel around the world in a bicycle as in your avatar. But the existence of someone doing that does little to prove that a bike is on the same level of a car, train, or airplane as an ideal form of transport.

Can a 6D capture a BIF? Of course it can. But, the efficiency/speed of achieving focus, not the ability of ever achieving focus, is perhaps a differentiator.

I am not, however, contending that the 6D is not as good as the 5D III for BIF - I simply don't know that and defer to experts like Gabe and other skilled BIF shooters for that. I'm only protesting the suggestion that BIF shots from a 6D might be some sort of proof that would seal up and end the thread discussion.

EDIT: And as Gabe suggests above, BIF shooters rely on the center point anyway, so perhaps BIF shooting is not a subject type that really differentiates the AF systems of various cameras.

pwm2 wrote in post #16767157
The precision of the AF point depends on how distinct pattern it sees. A non-cross point can have the feature 90 degrees wrong making it not seen.

A full cross can have the feature up to 45 degrees off. At 45 degrees off, the absolute precision of the AF system is lower.

A dual-cross can not have any feature more than 22.5 degrees off.

So while a non-cross point can do well, it suffers from more situations where it will not lock, or where it will lock but with more focus error, i.e. the camera will jitter a bit too near or too far. A cross and a dual-cross will always have features aligned so the camera can use it (if the feature has enough contrast and the light level is high enough) but a dual-cross will have less worst-case focus error.

A phase-detection system splits the light path into two and try to locate the same feature (contrast change) in both images. It then measures the distance between these features (contrast changes) and if there is a shift to the left or to the right to compute in which direction the lens need to change, and how much to change the focus.

If you are going to measure the distance between two lines, it's easier to measure the distance if holding the ruler at 90 degrees to the two lines. It's way harder to try to measure with the ruler at 45 degrees. In the case of a phase-detection sensor, the transition gets less distinct when the angle is off.

I see, most of that I knew but some new info there too, thanks.

So basically, not a huge difference between a double cross and a cross plus an extra -.
So worst case scenario is that in extreme cases there may be a 45 degree angle in one possible position since the second | is missing; we have an X and a - but no |. We are only talking about the center point, at least that's what the other poster complained about.

We also don't know what lens he was using (or I don't remember reading it). If it was slower than f/2.8 then it's only single cross type, and then it doesn't matter if the second element is a cross or simple -

Well I think If it could do it and get a good keeper rate then it would at least prove it could do it. I haven't seen a lot of BIF shots with it.
I also do BIF shots using only a center point on my 1DMKII.
Birds are tough to shoot in flight, but If you could keep the center point on them long enough , you should be able to get some good keepers.
I'm sure having multiple focus points makes it easier, but then again do you get what you want on the bird in focus ?
Whatever makes it easier, but If you want to shoot BIF and you can save some cash, looks like the 6d would be a winner.
Keep it simple whether it be on a bicycle or a train, its what works in the end. If you have enough money to try them all and keep just what you want, more power to you. Most of us don't have that luxury.
I would like to see a few of the BIF shots though, that would be nice to see.