Reading mode: Light Dark

1Ds mk3 focusing... hmm?

Go to last post
B

Benedict Slotte

Senior Member
Messages
1,165
Reaction score
2
Location
FI
All right, I decided to test AI servo using the 50 mm f/1.4 on the 1Ds mk3, after reading in some places that a lens is supposed to remain still if the camera and the target are static (so far I have expected a slight occasional movement to be normal, since the tracking can never be 100% perfect, and the camera AI servo cannot "know" that no movement will happen at all).

No, it doesn't stay perfectly still. If I keep shooting, maybe 20% of the shots are a bit out of focus. (Test setup: white page with black text at about 50 cm distance, AI servo enabled, AF-ON button kept pushed in, shooting frames one by one with a few second intervals, lens kept at f/1.4, both camera and target kept steady on a table.)

The 50 mm f/1.4 moves randomly back and forth by a small amount. Looking at the distance scale, I would say it shifts left and right by about + - 1 mm, and rarely (maybe just 3 times during a minute) it might occasionally move up to 2 mm and then back again.

I have the center focusing point activated, not the "ring of fire".

However, with smaller-aperture lenses (such as the 24-105 mm) there is much, much less movement, and focus is OK on every shot.

I think I remember that my 1Ds mk2 showed similar behaviour, i.e. large-aperture lenses were less stable in AI servo even with static targets, but I could be wrong... I cannot test because I don't own the 1Ds mk2 anymore.

When I read threads like the following, I get worried that this is not how things should be (but I cannot repeat that specific test exactly since I don't own the same lens):
http://forums.dpreview.com/forums/read.asp?forum=1032&message=27988910

I don't have samples to post yet, but maybe later.

--
B. Slotte
Turku, Finland
http://bslotte.smugmug.com
 
I just repeated the test with better lighting, and there was much less hunting, but no, it's not 100% still if I keep it focusing for several tens of seconds.

Another thing I have noticed is that the Af sometimes seems to pick up stuff outside the center point even if I don't have the assist points activated. Could there maybe even be an alignment mismatch between the red rectangles in the viewfinder, and the AF sensor itself...

Whichever way, I have been satisfied with the 1Ds mk3 so far.

--
B. Slotte
Turku, Finland
http://bslotte.smugmug.com
 
I have tried two copies of the 1Ds3 at this point.

With the central focus point only enabled, said to be the preferred setting for AI SERVO AF tracking, with the 400/5.6L, I experienced random in and out of focus on something that really should be relatively easy in good light, pelicans gliding at a constant speed.

Yes, it's a BIF but there's bird in flight and then there's BIRD IN FLIGHT.

tarn
Benedict Slotte said:
All right, I decided to test AI servo using the 50 mm f/1.4 on the
1Ds mk3, after reading in some places that a lens is supposed to
remain still if the camera and the target are static (so far I have
expected a slight occasional movement to be normal, since the
tracking can never be 100% perfect, and the camera AI servo cannot
"know" that no movement will happen at all).

No, it doesn't stay perfectly still. If I keep shooting, maybe 20% of
the shots are a bit out of focus. (Test setup: white page with black
text at about 50 cm distance, AI servo enabled, AF-ON button kept
pushed in, shooting frames one by one with a few second intervals,
lens kept at f/1.4, both camera and target kept steady on a table.)

The 50 mm f/1.4 moves randomly back and forth by a small amount.
Looking at the distance scale, I would say it shifts left and right
by about + - 1 mm, and rarely (maybe just 3 times during a minute) it
might occasionally move up to 2 mm and then back again.

I have the center focusing point activated, not the "ring of fire".

However, with smaller-aperture lenses (such as the 24-105 mm) there
is much, much less movement, and focus is OK on every shot.

I think I remember that my 1Ds mk2 showed similar behaviour, i.e.
large-aperture lenses were less stable in AI servo even with static
targets, but I could be wrong... I cannot test because I don't own
the 1Ds mk2 anymore.

When I read threads like the following, I get worried that this is
not how things should be (but I cannot repeat that specific test
exactly since I don't own the same lens):
http://forums.dpreview.com/forums/read.asp?forum=1032&message=27988910

I don't have samples to post yet, but maybe later.

--
B. Slotte
Turku, Finland
http://bslotte.smugmug.com
Click to expand...
 
The 50 1.4 is knowed to be a bit of a "hunter" , can you repeat your test with another lens ? My 1DsIII focus is perfect (though I have not tried my 50 1.4 on it)!!

--
JClaude
http://www.opusmeum.com
Personal Creative

 
I'm curious. Are white pages with black text your typical photographic subject?

Put another way, did you run this test because you've had an actual problem with your camera during practical applications (shooting photographs), or are you just testing in this manner because it was just something to do?

--
"Passion will make you crazy, but is there any other way to live?" —Kara Saun
 
Benedict Slotte said:
The 50 mm f/1.4 moves randomly back and forth by a small amount.
Looking at the distance scale, I would say it shifts left and right
by about + - 1 mm, and rarely (maybe just 3 times during a minute) it
might occasionally move up to 2 mm and then back again.
Click to expand...
Canon's 1D-Mk3 and 1Ds-Mk3 cameras need to come equipped with a Laser Range finder...
;) :o
 
Osiris Ani said:
I'm curious. Are white pages with black text your typical
photographic subject?

Put another way, did you run this test because you've had an actual
problem with your camera during practical applications (shooting
photographs), or are you just testing in this manner because it was
just something to do?
Click to expand...
That's awfully rude.

The gentleman is from Finland, but his English is perfectly clear.

Which part of this don't you understand?

"All right, I decided to test AI servo using the 50 mm f/1.4 on the 1Ds mk3, after reading in some places that a lens is supposed to remain still if the camera and the target are static..."

His photographic subjects obviously run the gamut:

http://bslotte.smugmug.com/

While yours is quite limited:

http://www.oqlus.com/

So cut people some slack as they explore their camera's strengths and weaknesses.

We all can benefit from such explorations.

That is those of us who have more than a casual interest in photography.

tarn
 
I accept this as normal behavior for the camera and lens under certain conditions. The variables seem to be the aperture and distance to subject. At anywhere from 1.2 to somewhere around 2.0 in aperture and with distances of just a couple of feet the lens will move frequently in AI Servo. It will do the same on my 20D. I'm not surprised. For example at close to the minimum focus disctance of just over 1 foor and at an f stop of 1.2 there is basically no DoF. Even in single focus mode under those conditions it is a challenge to get what you expect to be in focus in focus all the time. And it's not the "focus shift issue."
 
Tarn Tantikij said:
Osiris Ani said:
I'm curious. Are white pages with black text your typical
photographic subject?
Click to expand...
Click to expand...
Tarn Tantikij said:
So cut people some slack as they explore their camera's strengths and
weaknesses.
Click to expand...
I agree, cut the guy some slack. Had he posted say a AF test using red on green or red on blue here (or some more elaborate color scheme) test for focusing, numerous people here would be crying FOUL. I can just imagine the pages of -- Insufficient contrast, you should have used B/W!!!
 
Tarn Tantikij said:
That's awfully rude.
Click to expand...
Thank you for your response, Tarn, which was as useful as any of yours have ever been.

I'll await his reply, because you, not being him, can't answer my question.

--
"Passion will make you crazy, but is there any other way to live?" —Kara Saun
 
1JB said:
I accept this as normal behavior for the camera and lens under
certain conditions. The variables seem to be the aperture and
distance to subject. At anywhere from 1.2 to somewhere around 2.0 in
aperture and with distances of just a couple of feet the lens will
move frequently in AI Servo. It will do the same on my 20D. I'm not
surprised. For example at close to the minimum focus disctance of
just over 1 foor and at an f stop of 1.2 there is basically no DoF.
Even in single focus mode under those conditions it is a challenge to
get what you expect to be in focus in focus all the time. And it's
not the "focus shift issue."
Click to expand...
Your statement on aperture and distance to object may be correct but it's definitely incomplete when talking about DOF.

You must also consider the focal length.

I experienced same with the 400/5.6L near its closest focusing distance.

Which could be a lot closer one would wish.

The DOF would be shallow but not diminished to zero.

This certainly highlighted the weakness of this camera's AF system with this lens as I reported here earlier.

Perhaps it has to do with the f/5.6 and the AF's radically scaled down capability. Which was a real pity considering abundance contrast and light.

Which led me to question Canon's scheme:

http://forums.dpreview.com/forums/read.asp?forum=1032&message=27995289

Meanwhile if it was a flat piece of paper and some text on it, there could only be ONE plane of focus if he took the time to parallel his target to the sensor.

In which case the camera either achieved focus or failed.

There would be no different focus points available to it as an excuse.

We see attempts at such excuse before:

http://forums.dpreview.com/forums/read.asp?forum=1032&message=27802377

Meanwhile this strange AF system doesn't like white sign and black text according to this report:

http://forums.dpreview.com/forums/read.asp?forum=1032&message=27802864

Though I doubt that the target is glaringly bright here, we've not seen Mr. Benedict Slotte's sample photographs.

tarn
 
1JB said:
I accept this as normal behavior for the camera and lens under
certain conditions. The variables seem to be the aperture and
distance to subject. At anywhere from 1.2 to somewhere around 2.0 in
aperture and with distances of just a couple of feet the lens will
move frequently in AI Servo. It will do the same on my 20D. I'm not
surprised. For example at close to the minimum focus disctance of
just over 1 foor and at an f stop of 1.2 there is basically no DoF.
Even in single focus mode under those conditions it is a challenge to
get what you expect to be in focus in focus all the time. And it's
not the "focus shift issue."
Click to expand...
It is clear from some post read, that a few people think they know more about the Canon 1D(s) Mark III series cameras than the people who created them.

They set up test scenarios that are in conflict with the camera’s 1D(s) Mark III manual, and present the results as a problem or possible problem with the camera.
There is movement on my 20D, and my old 1Ds Mark II as well.
--
I am out to take the perfect picture, if it exits! :)
 
Osiris Ani said:
Osiris Ani said:
That's awfully rude.
Click to expand...
Thank you for your response, Tarn, which was as useful as any of
yours have ever been.

I'll await his reply, because you, not being him, can't answer my
question.
Click to expand...
The usefulness of answer is not just in the answer but the mind of the perceiver.

If your intention was honorable I could even provide a useful answer to your seemingly inane question.

While white pages with black text may or may not be a typical photographic subject, although it certainly is to one who photographs documents.

My trial attorney friend is in the habit of photographing his receipts. Although with not the top of the line Canon cameras as it were but with cell phone cams.

He complains that the Apple iPhone takes lousy pictures of receipts.

Should I tell him to not fret because the same is true with an $8000 professional camera?

For us photographers we frequently pick out the highest contrast subject in the same plane of focus to aid in our focusing. White signs with black text would be the natural targets and the equivalent of white pages with black text of your ridicule.

Hope this helps.

tarn
 
Thank you for your response, Tarn, which was as useful as any of yours have ever been.

I'll await his reply, because you, not being him, can't answer my question.

--
"Passion will make you crazy, but is there any other way to live?" —Kara Saun
 
WalterSrChat said:
1JB said:
I accept this as normal behavior for the camera and lens under
certain conditions. The variables seem to be the aperture and
distance to subject. At anywhere from 1.2 to somewhere around 2.0 in
aperture and with distances of just a couple of feet the lens will
move frequently in AI Servo. It will do the same on my 20D. I'm not
surprised. For example at close to the minimum focus disctance of
just over 1 foor and at an f stop of 1.2 there is basically no DoF.
Even in single focus mode under those conditions it is a challenge to
get what you expect to be in focus in focus all the time. And it's
not the "focus shift issue."
Click to expand...
It is clear from some post read, that a few people think they know
more about the Canon 1D(s) Mark III series cameras than the people
who created them.
They set up test scenarios that are in conflict with the camera’s
1D(s) Mark III manual, and present the results as a problem or
possible problem with the camera.
There is movement on my 20D, and my old 1Ds Mark II as well.
--
I am out to take the perfect picture, if it exits! :)
Click to expand...
Walter, we have yet to see any of your close to perfect pictures.

All we've seen is that you've memorized the manual particularly pages dealing with the AF.

I don't know if the socalled some posters know more about the 1D/Ds3 cameras than Canon.

But I am sure some do know more about photography where the 1D/Ds3 cameras are supposed to be the aid and not a detraction in the normal scheme of operation.

That is to say, if you have to enter a car from the bottom it is not a good design.

tarn
 
Osiris Ani said:
Thank you for your response, Tarn, which was as useful as any of
yours have ever been.

I'll await his reply, because you, not being him, can't answer my
question.

--
"Passion will make you crazy, but is there any other way to live?"
—Kara Saun
Click to expand...
Well, that proves one thing, you have a one track mind so your lines don't vary much.

Good luck getting an answer with such insult, picking on people whose English is not even their first language.

tarn
 
WalterSrChat said:
They set up test scenarios that are in conflict with the camera’s
1D(s) Mark III manual, and present the results as a problem or
possible problem with the camera.
Click to expand...
Can you post examples (links) of the "in conflict" test scenarios?
 
WalterSrChat said:
astirusty said:
WalterSrChat said:
They set up test scenarios that are in conflict with the camera’s
1D(s) Mark III manual, and present the results as a problem or
possible problem with the camera.
Click to expand...
Can you post examples (links) of the "in conflict" test scenarios?
Click to expand...
Click to expand...
...
Member said:
Start with this one:
http://forums.dpreview.com/forums/read.asp?forum=1032&message=27998391
Click to expand...
Point out specifically where the test scenario is in conflict with the camera's manual.
 
Thank you for your response, Tarn, which was as useful as any of yours have ever been.

--
"Passion will make you crazy, but is there any other way to live?" —Kara Saun
 
You must log in or register to reply here.

Keyboard shortcuts

f
Forum
m
My threads
Mobile site
About
Editorial content
Cameras & Lenses
Community
All content, design, and layout are Copyright © 1998–2025 Digital Photography Review All Rights Reserved.
Reproduction in whole or part in any form or medium without specific written permission is prohibited.
When you use DPReview links to buy products, the site may earn a commission.
©GPS Media - Guides, Products, Services.
Back
Top