Continuous AF in 1Dsiii vs 1Div and 1Dx

Messages
6,512
Solutions
2
Reaction score
973
Location
cork
I've a 1Dsiii, a 1Div and a 1Dx...

Now in assessing AF you have to allow for the interaction of the body with the lens. I don't use the 1Dsiii body much (got it secondhand), but use the others a lot, much of it for running photography.

On a shoot of my daughter's running club training, yesterday, I decided to use the 1Dsiii body with my best AFing lens, Canon 300 f2.8 IS mark one.

How did it fare?

The answer is straightforward. Not as well as the other two. In this comparison the 1Dx is the best, producing the highest number of spot-on shots. The 1Div lags a little, but not a lot and the 1Dsiii is just generally less accurate. Lots of nearly spot-on shots and some spot on. I was shooting very carefully, monoopod etc.

Would it do in a crisis? Yes but you'd have fewer really crisp ones.

Recently (I posted some) I'd something like 30 spot on frames in a row from the 1Dx with this lens.

I've found a similar gradation with the 1Dx and 1Div with another really good lens, the 70-200 F2.8 L ii.

These are some (of the good ones) from yesterdays shoot. All 1Ds3 + 300

 
Thing with the Canon all points tracking AF is that it doesn't show you the focus point whilst shooting. Without that information I have blasted away for hours taking half the shots slightly out of focus on low tracking sensitivity. On Nikon one has the choice of 3D tracking displaying focus points or Auto-area not. It is a serious grievance there is not more Canon viewfinder information about how well it is focusing at the time. I would't mind a focus dot on the edge of my vision giving me an idea how often I am in critical focus. I wouldn't be looking straight at it but if it was absent then after a while I would notice.

There was talk at the time of the 1D Mark III that Canon had tried to give it faster first shot acquisition and lower light capability but that the tracking was more jittery than before and this was exacerbated by some notorious production inaccuracies.
 
Last edited:
1Dsiii was what I was using not the 1D3.

If you are using all point servo af if you watch it you can see it lock onto something. Then, unless there is something else in front of the target, it's pretty good.
 
About what one would expect, I think. AF in EOS, especially AI Servo has been a work in progress since the beginning.

1Ds III: 45 pt (19 cross), circa 2007 AF algorithms processed by dual DIGIC III. Pretty good, but 2 years later ...

1D IV: 45 pt (39 cross), circa 2009 AF algorithms ("AI Servo II") processed by dual DIGIC IV. Even better, but 3 years later ...

1DX: 61 pt (46 cross), circa 2012 AF algorithms building off the 1D IV adding iTR, processed by dual DIGIC 5+ ("17x processing power of the DIGIC IV)

I've a 1Dsiii, a 1Div and a 1Dx...

Now in assessing AF you have to allow for the interaction of the body with the lens. I don't use the 1Dsiii body much (got it secondhand), but use the others a lot, much of it for running photography.

On a shoot of my daughter's running club training, yesterday, I decided to use the 1Dsiii body with my best AFing lens, Canon 300 f2.8 IS mark one.

How did it fare?

The answer is straightforward. Not as well as the other two. In this comparison the 1Dx is the best, producing the highest number of spot-on shots. The 1Div lags a little, but not a lot and the 1Dsiii is just generally less accurate. Lots of nearly spot-on shots and some spot on. I was shooting very carefully, monoopod etc.

Would it do in a crisis? Yes but you'd have fewer really crisp ones.

Recently (I posted some) I'd something like 30 spot on frames in a row from the 1Dx with this lens.

I've found a similar gradation with the 1Dx and 1Div with another really good lens, the 70-200 F2.8 L ii.

These are some (of the good ones) from yesterdays shoot. All 1Ds3 + 300

https://www.flickr.com/photos/16897255@N05/sets/72157648246665063/
 
The 1Dx with a really good Canon lens is about as good as you can get, I think. I've a few friends with Nikon outfits and they don't quite manage the accuracy level I'm getting with the 1Dx. They have D700 and D3s
 
1Dsiii was what I was using not the 1D3.
?? You don't think they shared the same AF module? Compare the viewfinders.

The 1Dx with a really good Canon lens is about as good as you can get, I think. I've a few friends with Nikon outfits and they don't quite manage the accuracy level I'm getting with the 1Dx. They have D700 and D3s
You like a fair fight between a camera of 2007 and one from 2011? Was this thread just a vanity exercise for someone who owns a few 1D-series??
 
1Dsiii was what I was using not the 1D3.
?? You don't think they shared the same AF module? Compare the viewfinders.

http://www.usa.canon.com/cusa/support/consumer?pageKeyCode=prdAdvDetail&docId=0901e0248004cd9e
The 1Dx with a really good Canon lens is about as good as you can get, I think. I've a few friends with Nikon outfits and they don't quite manage the accuracy level I'm getting with the 1Dx. They have D700 and D3s
You like a fair fight between a camera of 2007 and one from 2011? Was this thread just a vanity exercise for someone who owns a few 1D-series??
Don't be insulting! This is information. It's very very difficult to source actual information on sports stuff on the net. Whenever I tedioiusly work something out I post it. It helps people.

Here's a question for you.

What is THE major factor in out-of-focus sports photos? (It took me quite a while to nail this answer)
 
About what one would expect, I think. AF in EOS, especially AI Servo has been a work in progress since the beginning.

1Ds III: 45 pt (19 cross), circa 2007 AF algorithms processed by dual DIGIC III. Pretty good, but 2 years later ...

1D IV: 45 pt (39 cross), circa 2009 AF algorithms ("AI Servo II") processed by dual DIGIC IV. Even better, but 3 years later ...

1DX: 61 pt (46 cross), circa 2012 AF algorithms building off the 1D IV adding iTR, processed by dual DIGIC 5+ ("17x processing power of the DIGIC IV)--

Unapologetic Canon Apologist :-)
Why would the image processor used to process the images in these cameras make a difference, when all those mentioned have a separate, dedicated CPU for AF calculations?
 
Well, question not asked of me, but it's a good question. It doesn't matter to what is the actual result, though..
 
It's not exclusive to image processing.

 
Last edited:
Here's a question for you.

What is THE major factor in out-of-focus sports photos? (It took me quite a while to nail this answer)
Is it because they have a Garry behind them?

I haven't picked up a 1D in months, it wouldn't surprise me to find an AI servo focus confirmation is there but I forgot about it. One is five feet away from me, I just can't be bothered to pick it up...

Is the answer Sloth?
 
Here's a question for you.

What is THE major factor in out-of-focus sports photos? (It took me quite a while to nail this answer)
Is it because they have a Garry behind them?

I haven't picked up a 1D in months, it wouldn't surprise me to find an AI servo focus confirmation is there but I forgot about it. One is five feet away from me, I just can't be bothered to pick it up...

Is the answer Sloth?
Well it was but no longer is. It could now be because a fishywisht is behind them.
The answer is because not enough time is not being allowed for the AF to lock on. It's by far the biggest factor.
 
It's not exclusive to image processing.
In the mentioned cameras it is. So my question remains. But I understand now that the statement I questioned was written in lack of knowledge.
 
The answer is because not enough time is not being allowed for the AF to lock on. It's by far the biggest factor.
Canon does not help by mis-labelling a setting as AF Tracking Sensitivity when it is in fact a dumb delay where no more AF will take place.
 

Keyboard shortcuts

Back
Top