AF speed question

mercphoto

I have a question about auto-focus speed, was wondering if anyone could comment. I know that my Canon 20D will focus faster when using 2.8 or faster glass. My 70-200/2.8 focuses very quickly. But it slows down quite a bit with my 1.4 TC attached. How much of that is because the lens just became an f/4, and how much of that is inherint in the TC for some other reason?

In other words, would a 300mm f/4 prime focus faster than my 70-200/2.8 with 1.4TC? Would the 400mm f/5.6 prime be a dog at focusing? Would it focus faster than the 100-400/5.6?

luckyrwe

You are in a paradox...the f/2.8 has more light and can resolve the image faster, however there is more glass to move. The f/4 sees less light but has a lighter amount of parts to move.

ian408

Best to understand how AF works in the first place. The thumbnail is that it
want's to compare contrasting objects (ideally). It's easier to determine that
contrast the more light you have. This is more true when the camera is trying
to pick a focus point.

You can try and shoot with single point AF. Sometimes or you can try and
use the flash's beam to assist in focusing in low light situations.

But there is no real substitute for more light.

Ian

mercphoto

luckyrwe wrote:

You are in a paradox...the f/2.8 has more light and can resolve the image faster, however there is more glass to move. The f/4 sees less light but has a lighter amount of parts to move.

Putting lens mass into the equation answers a lot of questions for me. I understand why fast glass resolves quicker. But based on this, a genuine f/4 lens should AF faster than a 2.8 lens with a 1.4 teleconverter, since it has the reduced light gathering coupled with the higher mass of a 2.8 lens system. The true f/4 lens has the reduced light gathering but reduced mass as well.

OlgaJ

IMHO, having a 2.8 or faster lens on a 20D only means that the focus would be more accurate (closer to 1/3 DoF). Whether the focus is fast/slow depends on the lens and the lighting. My 24-70L f/2.8 focuses faster than my Tamron 28-75 f/2.8 in the same lighting. And in low light my 24-70 could hunt a tad.

Or have I misunderstood your question?

Olga

mercphoto

OlgaJ wrote:

IMHO, having a 2.8 or faster lens on a 20D only means that the focus would be more accurate (closer to 1/3 DoF)

I believe the center point on a 20D reacts differently to fast glass, but I don't know if that results in more accurate, or more speedy AF.

My question asked a different way, I can find plenty of sources about the quality of a lens. Charts, comparison pictures, etc. Is there a source that discussing the AF speed of particular lenses as well? Is it even measured and compared in some objective manner?

OlgaJ

Been looking for the Rob Galbraith review that talked about the 20D and fast lenses, but haven't been able to find it.

I did find Bob Atkins' page though:

http://www.bobatkins.com/photography/digital/eos20d.html

where he says:
"The 20D has a new high-precision cross-type sensor in the center position. It provides full cross-type performance with maximum apertures as small as f/5.6, yet it achieves up to 3 times the standard focusing precision when used with EF lenses featuring maximum apertures larger than or equal to f/2.8. Previous designs either worked as a cross at f5.6 and faster, with normal precision (consumer cameras), or as a high precision cross at f2.8 but a normal precision linear sensor with slower lenses (pro cameras) "

Olga

pathfinder

luckyrwe wrote:

You are in a paradox...the f/2.8 has more light and can resolve the image faster, however there is more glass to move. The f/4 sees less light but has a lighter amount of parts to move.

I don't think the weight of the glass is as important as how far the optical element has to be moved, and how powerful the drive motor is. Posters ramble on at length about the optical quality of various lenses on the web. Try to find a good discussion of barrel tube construction, iris mechanism mechanical consrtuction, or durability of the AF motors with heavy use.

The pros shooting the NFL games may shoot 5,000 frames or more in an afternoon - that means 5,000 shutter openings and closings, and 5,000 iris diapragm openings and closings. This calls for mechanical robustness. They may use more AF time in an afternoon, than many of use will use in a year.

Canon's f2.8 100 macro can be slow to focus if the lens is focused on infinity and you decide to focus on a bug at 1:1 , even at f2.8
Canon's f4 500mm telephoto will AF quickly, IF the lens os already reasonably close to focus. But if you change from infinity to 20 feet, focus takes longer due to the distance the optical element has to move.

SO I think there is more to discuss than just aperature. It is true, that Canon's AF system is more accurate with f2.8 lenses and they do perform better in dim light - both vertical and horizontal contrast change.

But even my 85mm f1.2 can fail to AF on my 1Dmkll if it gets dark enough. Just a fact of life.

mercphoto

pathfinder wrote:

Posters ramble on at length about the optical quality of various lenses on the web. Try to find a good discussion of barrel tube construction, iris mechanism mechanical consrtuction, or durability of the AF motors with heavy use.

Agreed. I can't find any objective reviews of AF speed. You can find sample pictures, MTF charts, lots of stuff to judge image quality. But if you want to judge a lenses ability to track a moving object, that information doesn't appear available.

Oh well. Thanks for the inputs guys.