Sample of 10D focus problem

[The Raven]

From where I'm at, that has to be an excellent idea. Not limited to
a 10D, not limited to photographic gear.
I test every single piece of equipment thoroughly before I use it
in the field. But then again, my livelyhood depends on it, for the
most part.
Blind trust in modern industrial manufacturing standards and QC can
be translated into naïvete imho.

Greetings
Stefan

But testing equipment you're not entirely familiar with first CAN get erroneous results. I believe someone ought to use their camera enough to get familiar with it before wholesale torture testing. Let me ask, over the years, how have you tested your camera equipment? What's YOUR process?

 

[pcockerell]

Well Pete, since I seem to be your favorite subject,

I've noticed that too! It's funny, sometimes you come across someone with an attittude a mile wide whose almost every post manages to raise your hackles, and it's really hard to resist the temptation to react to it. But I'll try to restrain myself in future. Sorry.

Anyway, to summaize you reply, it's all a matter of what you personally believe (as opposed to know) based on your (low) opinions of your fellow Canon users. That's all I was trying to confirm, thanks. I still stand by MY belief that people are unlikely to go to the hassle of sending the camera back unless they have real evidence in real-world photos that there's a problem, but maybe I'm doing the opposite to you and given them too MUCH credit. Who knows? It's a shame we'll never know what percentage of returned camera need an adjustment. Or indeed what percentage of all cameras produced are out of wack.

Cheers,
Pete

--
http://www.pbase.com/pcockerell
http://www.peter-cockerell.net:8080/

 

[The Raven]

I've noticed that too! It's funny, sometimes you come across
someone with an attittude a mile wide whose almost every post
manages to raise your hackles, and it's really hard to resist the
temptation to react to it. But I'll try to restrain myself in
future. Sorry.

Anyway, to summaize you reply, it's all a matter of what you
personally believe (as opposed to know) based on your (low)
opinions of your fellow Canon users. That's all I was trying to
confirm, thanks. I still stand by MY belief that people are
unlikely to go to the hassle of sending the camera back unless they
have real evidence in real-world photos that there's a problem, but
maybe I'm doing the opposite to you and given them too MUCH credit.
Who knows? It's a shame we'll never know what percentage of
returned camera need an adjustment. Or indeed what percentage of
all cameras produced are out of wack.

Cheers,
Pete

--
http://www.pbase.com/pcockerell
http://www.peter-cockerell.net:8080/

Sorry I raise your "hackles" but it might make you feel better that you get my "dander" up so we both need to make a trip to the drugstore. ;-)

Anyway, I don't have a low opinion of my fellow Canon users. I'm only reporting what I'm seeing here. Just this morning, I've seen someone ask if they should send their brand new camera back to Canon for recalibration because they feel that it'd work better even though they have no problems with it. Also, some of these new 10D owners have come from point & shoot cameras and while some have had film SLR experience, some haven't. So, we've got a lot of people testing their cameras to MAKE them fail and if they glitch, they've been so worked up about getting the problem fixed NOW, that there's BOUND to be some who've sent theirs in when it COULD be something they overlooked. I just wonder how many cases Canon's seen of this. Also, have you seen the call for Canon to "own up"? This person wants Canon to recall every camera and have every store send back their stock just because HE got a bad one and has read that others here have gotten bad ones as well. Never mind that he took the camera back to the store, got a new one, and it works fine. His isn't the first "recall" thread here either. So it's my opinion that people are being whipped into a frenzy to get their cameras checked. One guy checked his camera and found that it had a slight problem at a certain focal length and stated that he might send it in one day to have it calibrated. For now, he'll just work around it. He was told to go ahead and send it in NOW. It might break down AFTER the warranty expires and then he'd be out some bucks. I believe the 10D owner knew that and I expect he'd take care to not let that happen. It's just this sense of "urgency" imparted by these people that I feel makes new 10D owners jump the gun and send the camera in anyway. Oh well, I'm off to Walgreens. ;-)

 

[j-a-s-o-n]

I remember my first leap, and bad shots, transitioning from P&S to SLR. I even bought a top of the line camera. I feel bad for guys comming from P&S cameras, dealing with SLR issues, and now digital ones too. I agree some of these cameras may be bad, but too many people are aimlessly posting unscientific and non-systematic tests.

My first experiences with SLR ... I was shooting a group of kids on a field-trip to see baby goats in a dark barn, using a 1N and the 28-135 and 100-300 f/3.5-5.6 with ISO 100 film. That didn't work. Kinda like getting 2" from a ruler, with a lens that needs 1.5' to focus and shooting the widest angle at some extreme apperature, ISO, while using autofocusing, that doesn't work. Did I go around showing everybody those photos? Did I take the camera back? NO. I told my wife my lenses are no good and I had to buy the better ones, if she wanted any good pictures of the kids. LOL

Now that I know the best photos are taken at a different focus ranges and aperatures, depending on the lens ... and I know the limitations vs. advantages of those lenses; I want those small lenses back. Just like SOME of the photogs here will learn that the 10D has its limitations and strengths.

To those not experienced enough to find those strengths and exploit them, I say buy a camera with fewer weaknesses. Those who know how to systematically test, test and return if needed. But we need to curb the tide of people driving 95mph in first gear and are upset when they crash or break down, somehow. If you test a camera and a lense at extremes; you're gonna get extremely bad test results. If you camera is bad no matter what lens you use, at all settings; your camera may be broke. Call Canon. Post that. But if you post 1 pic, with 1 lens, on 1 setting and make a judgement; you're gonna get blasted by somebody.

I think my last post spelled out things pretty well. Fully/properly test the camera then post.

 

[Stefan Schott]

But testing equipment you're not entirely familiar with first CAN
get erroneous results.

Reading the manual beforehand will take care of that.

Let me ask, over the years, how have you tested your camera
equipment? What's YOUR process?

Are you seriously suggesting I write an essay about what kind of testing of my gear I did, and how, over the last 20-odd years? You absolutely have to be kidding. Maybe I'll include this in my memoires when I retire - perhaps then I'll have the time to do it.

Apart from that, I have let people participate in some of it by posting threads here, at least since I've gone digital (and made the switch to Canon). In case you've missed them, here are a few links:

http://forums.dpreview.com/forums/read.asp?forum=1019&message=4229010

http://forums.dpreview.com/forums/read.asp?forum=1019&message=4436302

http://forums.dpreview.com/forums/read.asp?forum=1019&message=4877068

http://forums.dpreview.com/forums/read.asp?forum=1019&message=4879046

http://forums.dpreview.com/forums/read.asp?forum=1019&message=4542842

Greetings
Stefan

 

[The Raven]

But testing equipment you're not entirely familiar with first CAN
get erroneous results.

Reading the manual beforehand will take care of that.

But many people DON'T as evidenced here.

Let me ask, over the years, how have you tested your camera
equipment? What's YOUR process?

Are you seriously suggesting I write an essay about what kind of
testing of my gear I did, and how, over the last 20-odd years? You
absolutely have to be kidding. Maybe I'll include this in my
memoires when I retire - perhaps then I'll have the time to do it.
Apart from that, I have let people participate in some of it by
posting threads here, at least since I've gone digital (and made
the switch to Canon). In case you've missed them, here are a few
links:

No, you don't have to be explicit. Just a general rundown would've been nice but...

 

[The Raven]

I read all the links you provided and found it interesting that you didn't use charts or rulers until recently due to the postings of this forum. You took test PICTURES of subjects you NORMALLY would photograph. That's what I've been saying SHOULD be used as testing your equipment.

 

[Yakuza]

First off, as has been well discussed, all consumer camera's have a slight focus proablem, a tad forward or backward. Only pro camera's "nail" it. The 10D is based on the Elan7's consumer focusing mechanisim. 1D owners aren't compaining about front/back focus.

I have an example of a shot with the ElanIIe, and a 75-300IS. My camera has always had a slight front/back focus problem (not If I manually focus, of course). Canon Germany released there official response to all this and said basically, it's usually going to be a touch off, within spec. If it's way off, return your camera. They also said that do the the size of the sensor, vs. 35mm, the slight innacuarcy is made more apparent. That doesn't make sense to me, but, that's what they said.

Anyway, in this photo, the monkeys clothes are tack sharp, but his head is just a touch off. When I shot this, I focused on his head, and then recomposed. I shot several pictures, and most are fine, but I like the monkeys expression in this one.

This pic was shot at 1/15th, at 5.6, at 300mm. Thank you IS!

http://www.pbase.com/image/16091298/original

 

[Stefan Schott]

You took test PICTURES of subjects you NORMALLY would
photograph.

Sort-of, yes. Shooting focus-charts was just killing me because it's too boring. I'm fortunate in that I live close to a location where I can shoot a village from a hill, about a mile or two in the distance. The subject is at quasi-infinity but with details in front and in the very distance that let me see even the slightest front/back focusing.

That's what I've been saying SHOULD be used as testing
your equipment.

Whatever. You know what I think the best thing about disagreeing with you is? That we both are civilized enough not to resort to name-calling ;-)

Later
Stefan

 

[The Raven]

You took test PICTURES of subjects you NORMALLY would
photograph.

Sort-of, yes. Shooting focus-charts was just killing me because
it's too boring. I'm fortunate in that I live close to a location
where I can shoot a village from a hill, about a mile or two in the
distance. The subject is at quasi-infinity but with details in
front and in the very distance that let me see even the slightest
front/back focusing.

That's what I've been saying SHOULD be used as testing
your equipment.

Whatever. You know what I think the best thing about disagreeing
with you is? That we both are civilized enough not to resort to
name-calling ;-)

Later
Stefan

Thank you!!! I sincerely appreciate that!

 

[pcockerell]

They
also said that do the the size of the sensor, vs. 35mm, the slight
innacuarcy is made more apparent. That doesn't make sense to me,
but, that's what they said.

I think this explains it...

http://forums.dpreview.com/forums/read.asp?forum=1019&message=5003481

Cheers,
Pete

--
http://www.pbase.com/pcockerell
http://www.peter-cockerell.net:8080/

 

[Stefan Schott]

1D owners aren't compaining about front/back focus.

I'm sorry, but I have to strongly disagree. The fact that 1D owners don't come whining about it all that much has different reasons and should not be mistaken for this error not occuring in 1D's. That just isn't true.
I invite you to browse through my own 1D focusing nightmare:
http://forums.dpreview.com/forums/read.asp?forum=1019&message=4229010

Regards
Stefan

 

[MIB]

Yakuza wrote:
. 1D owners aren't compaining about front/back focus.

Ask DividP about this. He has 2 1D-s
--
'MIB'

 

[Ben Strasser]

Lou,

This may sound dumb and off the wall but I have been reading the postings on the focus issues with digital AF cameras and I remember when I first got my EOS A2E and tried to shoot (photograph) my dog. He is a shaggy little liver and white Shid Tsu. I could not get the camera to focus on him for love or money. Everything else auto-focused great but him. When you think about it, his coloring was about as bland as you could get and not a sharp defination anywhere on the dog. When you think of how auto-focus works and at the sharpest focus point you have the maximum brightness and contrast, I could see how the AF was fooled. The A2E has 5 focus points and I have seen it randomly grab two or three of these points even when using eye controlled focus. It is interesting to note that I also have an EOS 1 with one focus point and have not had the problem focusing on my dog. BTW - I use the same lens on both cameras. My thought is that the problem with focusing is related to the contrast of the subject and if the CPU in the camera cannot determine the sharpest point. I never did try shooting the dog in dim light where the assist beam was active - this may have helped. I should be getting the 10D I ordered in about a week or so, I will have to see if the multi-AF points make a difference. I also opted of a higher contrast dog so I cannot duplicate the exact situation. Just a thought.

Here are 2 10D photos that show exactly what the problem my 10D
has. I would like to verify if any others experienced the same.

Both were shot from the same distance at the same settings at about
10 feet away. The only difference was in HOW I focused. (Lens :
Canon 24-70mm L)

Knowing the problem my 10D had from testing it many times, I knew
that when I shot at the lenses widest (24mm), I knew the AF (center
spot) would give me a bad reading and not focus properly. Though I
also knew that If I shot the image at 70mm (fully zoomed) it would
be pretty sharp. So I determined the problem was shooting in wide
angles.

In this example when fully zoomed at 70mm the lens registered a
distance to subject of just about 10 feet. When AF'ing while at
24mm, the distance registered was much higher almost hitting
infinity. Result: a soft image.

PHOTO 1
Here is a shot of the soft image. Taken stright with the center AF
spot with the lens at 24mm. Very soft. The center spot was fully on
the magazine and there was nothing to fool it. This was done 10
times with the same result.

http://www.photo.net/photodb/photo?photo_id=1460766&size=lg

PHOTO 2
Here is the same shot EXCEPT that I first focus at 70mm then held
the shutter half way and backed the lens up to 24mm. Much sharper
BUT not acceptable, in the real world I will not have time to do
this for all my shots at an event. The camera should focus properly
at both ranges.

http://www.photo.net/photodb/photo?photo_id=1460734&size=lg

Both images are a 100% crop of a much wider photo.

 

[Yakuza]

I wouldn't ask davidp how to open a door

Ask DividP about this. He has 2 1D-s
--
'MIB'

 

[Yakuza]

Thanks for a great laugh!

I should be getting the 10D I ordered in about a week or
so, I will have to see if the multi-AF points make a difference. I
also opted of a higher contrast dog so I cannot duplicate the exact
situation. Just a thought.