Modified 10D autofocus test

Just updating my recommended autofocus test for the 10D to address

theories that the domino test may not be valid if the autofocus

sensors are larger than the autofocus indicators shown in the view

finder.

 

To perform this test, you will need a rigid background, 5 paper towel

rolls, 5 playing cards, a ruler, a tri-pod, a 10D, the user manual, a

lens with the largest aperture you like to shoot at, good lighting,

adhesive.

 

Make the distance D 1/2 to 2 depth of fields for the aperture and

subject distance you're testing. (You may want to first do the domino

test to figure out what you want D to be). However, use the same

distance D between each card.

 

Make Y much larger than D, say Y>= 5*D.

 

Once you assemble the playing card target, set your camera on a tri-

pod. Select self timer mode or use a remote shutter with mirror lock

up.

 

Select the center autofocus sensor.

 

Manually focus your camera at infinity (or macro). Leave the lens

switch on auto-focus.

 

Take the photograph, letting the autofocus system attempt to focus on

the center card.

 

Repeat. Before each shot, set the focus to infinity or macro before

using the camera's autofocus mechanism.

 

Evaluate your shots. Note if one of the untargeted cards is sharper

than the targeted card. Note if the results were consistent.

 

Post a sample image if you wish. Return the camera if you have a

lemon. Or keep it and learn to focus manually when shooting with

large apetures.

 

Recommendations:

Use playing cards that have a lot of detail to focus on.

Keep all other variables constant. Lighting, exposure, white balance.

I'm dumbfounded by the relentless appearance of new 10D focus test scenarios. If a camera doesn't focus correctly when taking photographs of "real" subjects, how many unique focus tests are needed to confirm/corroborate that observation?

 

If my 10D didn't autofocus correctly, I'd have exchanged it during the retailer's grace period or I'd send it in for warranty service.

 

(For the umpteenth time,) I don't doubt that some 10Ds have/had focus problems. Surely the 10D isn't the first (or last) SLR, digital or film, for which every unit off the assembly line didn't autofocus within spec.

 

If you can't get your particular unit to autofocus satisfactorily, no matter how many times you've exchanged it or sent it in for service, perhaps (just perhaps) you either haven't read / don't understand the limits and tolerances of its autofocus mechanism, or you just don't possess the capability to operate the device correctly.

Jon,

 

I agree completely. I have not had focusing problems with my

10D. If I did have a problem, I would get it solved right away

and not play silly games that really don't provide any real info.

 

I have read about some folks having problems. I have advised

them to get the camera checked out by either Canon or a

qualified technician who can work under a Canon warranty.

 

Some of the problems may actually be the result of using the

camera improperly or having unrealistic expectations as to

what the camera can actually do.

 

My advice is stop talking tech and take some pictures.

I feel your pain. The domino test is perfectly valid. However, this is an attempt to design a test for those who want an objective way of determining if their 10D has an autofocus problem. The domino test has been called "bunk" by others. Yet others have claimed that dominos are too small. This test and post was address to allay those concerns. If you're not interested, you didn't need to click the link.

Mike, I appreciate the effort put forth, but as a 10D owner, I cant say I have experienced any of the <I>hype.</I><BR>

I think what you have done is nice and feel bad for anyone who is or thinks they have a focusing problem. <BR>

User error? hard to tell. Defect in the camera? Possibly. All I know is I am having a blast producing ultra sharp pictures with my 10D.

"If you're not interested, you didn't need to click the link."

 

You're absolutely correct. Call it morbid curiosity.

I actually am really happy to read about this test. I'm just starting to get into real photography and don't know what I'm doing that well. I plan to buy a 10D to learn on. I've been worried that if I get a lemon, I won't know it because I'll probably assume it's my own error if something comes out blurry. This seems like a way that I could test the camera and see if there is a problem right away. Mike, are there any common AF problems with the 10D that you know of that would not be found with this test?

Thanks so much,

Lisa

Thanks A.J. and Lisa. My current 10D focuses just fine using the domino test (which in my opinion is a perfectly objective and valid test, and much simplier to perform than this playing card test).

 

However, too many people in my view are too quick to claim 10D autofocus problems are user error problems. This leads to lost time. Hence, an objective and valid test is helpful in these circumstances.

 

It is important that others be able to quickly ascertain if they really do have a problem, so that they can return the lemon to the store for an exchange or refund before the grace period expires.

 

And in fact this is a service to Canon because if they do have a quality control problem, the feedback provided by store and service returns will motivate them economically to improve quality control.

 

I have read accounts that Canon does fix 10D autofocus problems, but more recently, I have been reading that the cameras come back from service as bad as they left. I have never had the need to send any equipment to Canon for repair or service. Besides, even if Canon service fixed the cameras, the new (excited) owner is out of the 10D for a week or two.

I think that the reality is that many AF problems perceived by 10D users are categorised as "user error" - i.e. not realising that the focussing points are significantly larger than indicated in the viewfinder (and may be misaligned with the indicators as well without being "out of spec"). Actually, I consider this a design problem (akin to the problems of getting good flash exposures with E-TTL on a 10D or 300D), partly inherent to the small viewfinder image. Maybe it will be better addressed in newer models. According to Chasseur d'Images, Canon admitted that around 0.6% of 10D's had a "real" focussing problem with being incorrectly adjusted in the early stages of the production run.

Lisa,

 

In my opinion, tests such as these are usually of little use and

are only good for people who like to talk tech. That's great for

them, it's fun, but they don't necessarily know what they are

talking about and can make the process of taking pictures

appear to be very overwhelming; especially to relative novices.

 

The Canon 10D is an expensive aquisition for a beginner. It

has many features that are great to have if you need them,

but usually are just not necessary and don't help someone

learn photography. In fact, they can actually distract one from

developing an "eye".

 

Some of the film cameras that I have are as much as 30- 40

years old; and I use them. This will not be so for digital. New

technology is being introduced at a fairly quick pace. Although

products introduced five years ago and products introduced

now ( the 10D) will be usable in the forseeable future, I believe

one is going to have to be updating much more often than has

ever been necessary. Buy a less expensive camera. Take a

class. Join a club. Don't buy an expensive piece of equipment

that you're not going to use. If you don"t think your camera is

working correctly, show it to someone who can check it out and

possibly explain how to solve the problem ( perhaps someone

such as Mike}. Dont't assume your equipment doesn't work

correctly. Take pictures.

 

 

Mike,

 

Please don't take offense to my comments. I value this forum

and the varied advice and opinions of all of the participants. I

happen to believe, as I have stated, that there is too much talk

about non-existant problems and things like "the perfect lens"

by people who are really not using their equipment for taking

pictures. I am not saying that applies to you.

 

Good luck with your 10D.

Lisa,

 

You can have an opinion, but if you *think* your 10D has an autofocus problem, be prepared to prove it to the store when you return it.

 

Mike

Opps, that shouldn't have been directed at Lisa.

Mike,

 

 

No need to get nasty. I value your info and opinions.

 

Autofocus is not infallible; far from it. As you know, many trade-

offs are made in the process of designing a camera in order to

achieve the final design and price. If at all possible, a

prospective buyer should try to use equipment in a real

situation before he or she buys it . If it doesn't feel right , don't

buy it.

 

Using the right equipment for the right job is very key; and very

subjective and personal. What is satisfactory for me, may not

be for you. I believe your test is of little use, for me. It does

not provide real situation information unless you are going to

shoot closeup still-life with a 35mm auto focus camera and

possibly an inappropriate lens. That can be done in a pinch,

but is not what the camera was really designed for.

 

If a photographer has purchased equipment that he or she

believes is not working properly, that problem can generally be

solved promptly, if it was purchased form a reputable dealer.

Sometimes it can't. That's life. Take it to a qualified technician

who can check it out under warranty. Things break. That's life.

Sometimes you have "buyer's remorse" and wish you had

bought a different product. That's life, too.

 

I have had problems with manufacturers, including a bad one

with Kodak, where they acknowledged the problem only after

over two years of complaints. Some people did not experience

the problem; some did. They replaced the equipment, finally,

at no charge, with a redesigned unit that worked great. I was

out money and my reputation did suffer with a client, though.

Perhaps they introduced the equipment and film in good faith.

Perhaps they didn't. However, problems, big problems

appeared for some when the original system was used in real

situations. They stonewalled. In my experience that is the

exception, not the rule.

 

If a manufacturer is determined not to tell the truth and to hide

a known flaw , one has a big problem, to say the least.

Perhaps this is a good reason not to be overeager to purchase

equipment when it first comes out but, rather, wait until others

have used it in real situations.

 

Manufacturers usually do want to hear about problems that

arise in real situations and make apprpriate fixes. They want

to sell product and not have photographers telling each other

that a piece stinks , thus destroying sales.

 

Again, so far, I have had no autofocus problems, in real

situations, with my 10D equipment or any other 10Ds that I

have used. I have put a lot of "film" through 10Ds during the

last six months including having to shoot closeup still lifes

when some other equipment was temporarily misplaced by an

airline.

 

I hope people will share idiosyncrasies and problems they find

with 10Ds and it turns out threre is not an auto focus flaw. If

there is a flaw, I hope Canon acts responsibly.

 

At any rate, Mike, please don't take comments that don't

agree with yours so personally. I am interested in your

opinions and everyone elses opinions. My belief is that

anyone who is not interested in opinions other than their own,

should not participate in this forum.

 

Peace.

 

 

gw

G.W., no nastiness was intended to you, G.W. I was a little confused, though.

<p>

Anyway, I agree with many of your comments. My current 10D autofocuses fine. Many that I tested did not. The errors on the failing units were consistent, though.

<p>

<i>I believe your test is of little use, for me. </i>

<p>

The domino test worked well for me. Simple to set up. Confirmed that my autofocus was not focusing as indicated on the active superimposed indicator.

<p>

This test is much more complicated to set up. But it was designed to address concerns that the dominos were to small compared to the autofocus sensors.