Shocking insight: AF is not a closed feedback loop!

Started Jul 18, 2003 | Discussions
OP Mishkin™ Contributing Member • Posts: 917
to sooth you...

the one that broke had miscalibrated drive quantity detector. It deserved its fate

David G wrote:

Buying a 50mm 1.8 & trashing it that is.. It fell out of my bag
and broke in two.

Replaced it the next day with a new one. Gotta love those $70
disposable lenses..

-David

-- hide signature --

Mishkin

AdamT
AdamT Forum Pro • Posts: 58,985
Re: incompatibility with 10D?

distant scale. Guess what? It was jumping as much as on 10D when
pressing AF multiple times on the same subject. Focus lock - 10ft!

the 24-85 does this also at 24mm especially at infinity in fact you can see it in the viewfinder most of the time and unless you're careful and know what you're doing OOF shots can abound, this has been mentioned before, it would seem that the magic number here is "24", All lenses can Jitter around on ANY body from a 10 to a 10D, the wider they are, the worse it is depending on what you're aiming at and that includes the D60 and 1D also, the AF sensors are massive compared to the detail they're trying to focus on at infinity though that doesn't explain why they are OOF if ALL the detail under the sensor area is at infinity..

I'm just saying that it's not a "10D" thinng and not solely with the 24-70L

-- hide signature --

Please ignore the Typos, I'm the world's worst Typist

..Zero..to..Shot..Taken..in..16,000th..of..a..second----> EOS 1D

The No1 Dedicated 1D forum in the UK -------->

http://www.1dforum.co.uk/php/phpBB2/

 AdamT's gear list:AdamT's gear list
Canon PowerShot G1 X Sony RX100 III Sony RX10 III
OP Mishkin™ Contributing Member • Posts: 917
for now that you don't have 1Ds ;)

...you just should remember that with 10D you may get SOME of the shots blurred by up to 4 pixels at f/2.8 (100% of 0.03mm CoC) and proportionally more at faster apertures (e.g. 8 pixels at f/1.4). I don't know the statistical distribution within this 0-100% acceptable accuracy. Hopefully, it has maximum at 0% (perfect focus, 0-pixel blur) and only a small tail at 100%. But some of the shots are almost quaranteed to be as blurry as described. And it's within AF spec!

With 1D/1Ds, theoretically things should improve by 3x. There still could be problems in f/1.0-f/2.0 range.

It's best just to know these limitations and learn how to deal with them. I for myself decided what to do when shooting in the 1.4-4 range - shoot a lot of frames while moving to your subject (when it's close) or rotating focus ring (when subject is too far). I'm going to use this technique extensively when shooting available light and not worry about AF giving me blurry images.

And L lenses are still great. Use them now (with knowledge), and enjoy your investment in the future when you will have a camera with much better AF.

Larry H. Smith wrote:
Up-front: I haven't read (and probably couldn't have followed) all
the tech-discussion on the AF "issue'"in the various threads.

Does this new understanding (Thanks mishkin!)make it possible to
put into some reasonable number of words , a "best" way for the
user to maximize the AF potential of his rig, ...or a best way to
test/adjust "whatever" for that "best" performance?

(I mean when using whatever f-stop the photog WANTS to use, for
image reasons other-than-focus.)

This whole "what the camera can/can't do FOR you" thing is the main
reason I won't be happy till I have a pro-level camera that will
let ME do what I want, with a viewfinder that CLEARLY shows me what
is happening.

Thanks,

Larry

Mishkin2 wrote:

...a lotta stuff!

-- hide signature --

Mishkin

OP Mishkin™ Contributing Member • Posts: 917
Re: Hang on...

yada10d wrote:

I would think the recomendation should be to get a fast prime and
stop it down, not put a possibly less capable lens on the camera to
'hide' the problem.

It's a typical recommendation coming from pros to use fast primes and shoot available light without flash. Available light forces one to open up the lens and misfocusing (within the spec!) can bring a lot of disappointment to newbies who heard about primes being incredibly sharp.

-- hide signature --

Mishkin

DavidP Forum Pro • Posts: 29,088
but how do it know?

As Mishkin described it, he said the body tells the lens "go 20 steps thata way", and the lens goes 20 steps that away, and it's done. No feedback at all? No double-check? As in, "OK, I went 20 steps, now waddya think, big boy?"

If it's only a single step, the body would have to know what size steps the lens takes, right? That would mean the lens has to know that info, and tell it to the body somehow, right?

Jason Hutchinson wrote:

Loop stability is the most likely one, but what about the response
time of the older motors? They're clearly not designed as well as
modern USM motors for the small repetitive motions that'd be
required for that. Who knows.

-- hide signature --

The Lowest Paid Concert Photographer Around
http://www.neonlightsimaging.com/artshow/final.htm
Photography -- just another word for compromise

'Since we can't keep crime in check, why don't we legalize it and tax it out of business?' -- Will Rogers

DavidP Forum Pro • Posts: 29,088
a simple rule

Buy fast primes, but stop them down.

Problem solved.

Larry H. Smith wrote:
Up-front: I haven't read (and probably couldn't have followed) all
the tech-discussion on the AF "issue'"in the various threads.

Does this new understanding (Thanks mishkin!)make it possible to
put into some reasonable number of words , a "best" way for the
user to maximize the AF potential of his rig, ...or a best way to
test/adjust "whatever" for that "best" performance?

(I mean when using whatever f-stop the photog WANTS to use, for
image reasons other-than-focus.)

This whole "what the camera can/can't do FOR you" thing is the main
reason I won't be happy till I have a pro-level camera that will
let ME do what I want, with a viewfinder that CLEARLY shows me what
is happening.

Thanks,

Larry

Mishkin2 wrote:

...a lotta stuff!

-- hide signature --

The Lowest Paid Concert Photographer Around
http://www.neonlightsimaging.com/artshow/final.htm
Photography -- just another word for compromise

'Since we can't keep crime in check, why don't we legalize it and tax it out of business?' -- Will Rogers

DavidP Forum Pro • Posts: 29,088
does anybody concur?

I'll have to admit I didn't quite follow your "why f/1.4 is worse than f/4 max aperture" when it comes to AF accuracy.

Did anybody else (Jason?) follow and agree with what he said in that other thread?

I certainly feel like I can MANUALLY focus better with a f/1.4 lens stopped down to f/4 than just using an f/4 lens wide-open.

Mishkin2 wrote:

It's a typical recommendation coming from pros to use fast primes
and shoot available light without flash. Available light forces one
to open up the lens and misfocusing (within the spec!) can bring a
lot of disappointment to newbies who heard about primes being
incredibly sharp.

-- hide signature --

The Lowest Paid Concert Photographer Around
http://www.neonlightsimaging.com/artshow/final.htm
Photography -- just another word for compromise

'Since we can't keep crime in check, why don't we legalize it and tax it out of business?' -- Will Rogers

OP Mishkin™ Contributing Member • Posts: 917
thanks, Adam

it's just very exhausting that I went through FIVE 24-70's and they all can't consistently shoot sharp pictures at 24mm f/2.8 - with camera which doesn't have any significant issues with my other 4 lenses.

I understand now that I need to send my camera and lenses to Canon, but I'm afraid to get screwed up and then to beg for weeks for special attention from Canon.

Adam-T wrote:

distant scale. Guess what? It was jumping as much as on 10D when
pressing AF multiple times on the same subject. Focus lock - 10ft!

the 24-85 does this also at 24mm especially at infinity in fact you
can see it in the viewfinder most of the time and unless you're
careful and know what you're doing OOF shots can abound, this has
been mentioned before, it would seem that the magic number here is
"24", All lenses can Jitter around on ANY body from a 10 to a 10D,
the wider they are, the worse it is depending on what you're aiming
at and that includes the D60 and 1D also, the AF sensors are
massive compared to the detail they're trying to focus on at
infinity though that doesn't explain why they are OOF if ALL the
detail under the sensor area is at infinity..

I'm just saying that it's not a "10D" thinng and not solely with
the 24-70L

-- hide signature --

Mishkin

DavidP Forum Pro • Posts: 29,088
what you're telling me

What you're telling me is that I'm attempting to do the impossible. You know that, right? LOL.

Mishkin2 wrote:

It's a typical recommendation coming from pros to use fast primes
and shoot available light without flash. Available light forces one
to open up the lens and misfocusing (within the spec!) can bring a
lot of disappointment to newbies who heard about primes being
incredibly sharp.

-- hide signature --

The Lowest Paid Concert Photographer Around
http://www.neonlightsimaging.com/artshow/final.htm
Photography -- just another word for compromise

'Since we can't keep crime in check, why don't we legalize it and tax it out of business?' -- Will Rogers

yada10d Regular Member • Posts: 279
Re: Hang on...

Mishkin2 wrote:

yada10d wrote:

I would think the recomendation should be to get a fast prime and
stop it down, not put a possibly less capable lens on the camera to
'hide' the problem.

It's a typical recommendation coming from pros to use fast primes
and shoot available light without flash. Available light forces one
to open up the lens and misfocusing (within the spec!) can bring a
lot of disappointment to newbies who heard about primes being
incredibly sharp.

Yes, I realise that, but the low light photo's are going to be out of focus in more ways with the slow lens:

  • It will have more trouble locking on to low contrast subjects

  • If the newbie shoots wide-open in low light, camera shake will kill the photo's on the slow lens even if they are in focus, and the fast prime will still be better if it manages to get a good focus. (which it will, some of the time)

I think the fast prime recomendation should stand. We just need a newbie FAQ to refer people to when they come (and they will) There is a pretty good thread over at Rob Galbraith's I think that could be referred to as a FAQ...

Y.

-- hide signature --

Mishkin

OP Mishkin™ Contributing Member • Posts: 917
it's all programmed in CoC, I believe

the signal from phase detection AF unit is proportional to the diameter of the circle of confusion as currently projected by the lens on sensor focal plane (or equivalent of this plane as reflected by secondary mirror). Plus sign, e.g. AF processor tells the lens: "hey, you're misfocused by -12.5 CoC's!" (CoC being 0.03mm). Lens "drive quantity detector" also is calibrated in CoC's. "Yes, Sir, I moved -12.5 CoC's!" - "Good boy! Have a rest for now."

This talk may happen several times during focusing procedure, though - in case, for example, camera moved since the focusing began but before it locked. This also helps AF sensor to gather more data and tell the lens: "Did I say -12.5? I meant -12.54" at the last moment before it shouts "Stop!".

DavidP wrote:
As Mishkin described it, he said the body tells the lens "go 20
steps thata way", and the lens goes 20 steps that away, and it's
done. No feedback at all? No double-check? As in, "OK, I went
20 steps, now waddya think, big boy?"

If it's only a single step, the body would have to know what size
steps the lens takes, right? That would mean the lens has to know
that info, and tell it to the body somehow, right?

Jason Hutchinson wrote:

Loop stability is the most likely one, but what about the response
time of the older motors? They're clearly not designed as well as
modern USM motors for the small repetitive motions that'd be
required for that. Who knows.

-- hide signature --

Mishkin

yada10d Regular Member • Posts: 279
Agreed. (nt)

DavidP wrote:
Buy fast primes, but stop them down.

Problem solved.

Larry H. Smith wrote:
Up-front: I haven't read (and probably couldn't have followed) all
the tech-discussion on the AF "issue'"in the various threads.

Does this new understanding (Thanks mishkin!)make it possible to
put into some reasonable number of words , a "best" way for the
user to maximize the AF potential of his rig, ...or a best way to
test/adjust "whatever" for that "best" performance?

(I mean when using whatever f-stop the photog WANTS to use, for
image reasons other-than-focus.)

This whole "what the camera can/can't do FOR you" thing is the main
reason I won't be happy till I have a pro-level camera that will
let ME do what I want, with a viewfinder that CLEARLY shows me what
is happening.

Thanks,

Larry

Mishkin2 wrote:

...a lotta stuff!

OP Mishkin™ Contributing Member • Posts: 917
Re: a simple rule

or ask Wilkinsons to move forth and back while you're firing a burst of 21 frames.

Them sitting in the chairs with guitars is no good for fast primes

DavidP wrote:
Buy fast primes, but stop them down.

Problem solved.

Larry H. Smith wrote:
Up-front: I haven't read (and probably couldn't have followed) all
the tech-discussion on the AF "issue'"in the various threads.

Does this new understanding (Thanks mishkin!)make it possible to
put into some reasonable number of words , a "best" way for the
user to maximize the AF potential of his rig, ...or a best way to
test/adjust "whatever" for that "best" performance?

(I mean when using whatever f-stop the photog WANTS to use, for
image reasons other-than-focus.)

This whole "what the camera can/can't do FOR you" thing is the main
reason I won't be happy till I have a pro-level camera that will
let ME do what I want, with a viewfinder that CLEARLY shows me what
is happening.

Thanks,

Larry

Mishkin2 wrote:

...a lotta stuff!

-- hide signature --

Mishkin

DavidP Forum Pro • Posts: 29,088
sorta closed-loop ?

Doesn't this, in effect, make it a semi-closed loop?

I simply can't imagine the lens being told "move 13 steps" and it then moves 13 steps, without a final "OK, now how far should I move?" step in there.

Mishkin2 wrote:

This talk may happen several times during focusing procedure,
though - in case, for example, camera moved since the focusing
began but before it locked. This also helps AF sensor to gather
more data and tell the lens: "Did I say -12.5? I meant -12.54" at
the last moment before it shouts "Stop!".

-- hide signature --

The Lowest Paid Concert Photographer Around
http://www.neonlightsimaging.com/artshow/final.htm
Photography -- just another word for compromise

'Since we can't keep crime in check, why don't we legalize it and tax it out of business?' -- Will Rogers

OP Mishkin™ Contributing Member • Posts: 917
"Amanda, can you please move from -100% DOF to +100% DOF..."

"...while I hold the shutter button?"

;)))

Mishkin2 wrote:
or ask Wilkinsons to move forth and back while you're firing a
burst of 21 frames.

Them sitting in the chairs with guitars is no good for fast primes

DavidP wrote:
Buy fast primes, but stop them down.

Problem solved.

Larry H. Smith wrote:
Up-front: I haven't read (and probably couldn't have followed) all
the tech-discussion on the AF "issue'"in the various threads.

Does this new understanding (Thanks mishkin!)make it possible to
put into some reasonable number of words , a "best" way for the
user to maximize the AF potential of his rig, ...or a best way to
test/adjust "whatever" for that "best" performance?

(I mean when using whatever f-stop the photog WANTS to use, for
image reasons other-than-focus.)

This whole "what the camera can/can't do FOR you" thing is the main
reason I won't be happy till I have a pro-level camera that will
let ME do what I want, with a viewfinder that CLEARLY shows me what
is happening.

Thanks,

Larry

Mishkin2 wrote:

...a lotta stuff!

-- hide signature --

Mishkin

DavidP Forum Pro • Posts: 29,088
I can do that

And here I was thinking that I should sit perfectly still while taking the pictures. Now I know I should move back and forth. But only along the lens axis, otherwise I'll get motion blur.

Mishkin2 wrote:

or ask Wilkinsons to move forth and back while you're firing a
burst of 21 frames.

Them sitting in the chairs with guitars is no good for fast primes

-- hide signature --

The Lowest Paid Concert Photographer Around
http://www.neonlightsimaging.com/artshow/final.htm
Photography -- just another word for compromise

'Since we can't keep crime in check, why don't we legalize it and tax it out of business?' -- Will Rogers

DavidP Forum Pro • Posts: 29,088
newbie recommendation

Buy a fast prime, or the fastest zoom you can afford.

This doesn't mean you have to shoot it wide-open, though. In fact, you shouldn't, unless you need to, or want the effect.

Hey, that applies to ALL of us, newbies or not.

-- hide signature --

The Lowest Paid Concert Photographer Around
http://www.neonlightsimaging.com/artshow/final.htm
Photography -- just another word for compromise

'Since we can't keep crime in check, why don't we legalize it and tax it out of business?' -- Will Rogers

Larry H. Smith Veteran Member • Posts: 6,270
Thanks Mishkin...

That's the kind of thing i come here hoping to learn

Larry

Mishkin2 wrote:
...you just should remember that with 10D you may get SOME of the
shots blurred by up to 4 pixels at f/2.8 (100% of 0.03mm CoC) and
proportionally more at faster apertures (e.g. 8 pixels at f/1.4). I
don't know the statistical distribution within this 0-100%
acceptable accuracy. Hopefully, it has maximum at 0% (perfect
focus, 0-pixel blur) and only a small tail at 100%. But some of the
shots are almost quaranteed to be as blurry as described. And it's
within AF spec!

With 1D/1Ds, theoretically things should improve by 3x. There still
could be problems in f/1.0-f/2.0 range.

It's best just to know these limitations and learn how to deal with
them. I for myself decided what to do when shooting in the 1.4-4
range - shoot a lot of frames while moving to your subject (when
it's close) or rotating focus ring (when subject is too far). I'm
going to use this technique extensively when shooting available
light and not worry about AF giving me blurry images.

And L lenses are still great. Use them now (with knowledge), and
enjoy your investment in the future when you will have a camera
with much better AF.

Larry H. Smith wrote:
Up-front: I haven't read (and probably couldn't have followed) all
the tech-discussion on the AF "issue'"in the various threads.

Does this new understanding (Thanks mishkin!)make it possible to
put into some reasonable number of words , a "best" way for the
user to maximize the AF potential of his rig, ...or a best way to
test/adjust "whatever" for that "best" performance?

(I mean when using whatever f-stop the photog WANTS to use, for
image reasons other-than-focus.)

This whole "what the camera can/can't do FOR you" thing is the main
reason I won't be happy till I have a pro-level camera that will
let ME do what I want, with a viewfinder that CLEARLY shows me what
is happening.

Thanks,

Larry

Mishkin2 wrote:

...a lotta stuff!

OP Mishkin™ Contributing Member • Posts: 917
yes, but...

it's still subject to body or lens miscalibration. The AF sensor perhaps rectifies the position several times, but it still suffers from miscalibration.

DavidP wrote:
Doesn't this, in effect, make it a semi-closed loop?

I simply can't imagine the lens being told "move 13 steps" and it
then moves 13 steps, without a final "OK, now how far should I
move?" step in there.

Mishkin2 wrote:

This talk may happen several times during focusing procedure,
though - in case, for example, camera moved since the focusing
began but before it locked. This also helps AF sensor to gather
more data and tell the lens: "Did I say -12.5? I meant -12.54" at
the last moment before it shouts "Stop!".

-- hide signature --

Mishkin

OP Mishkin™ Contributing Member • Posts: 917
harmonic oscillator?

make sure this motion does not develop into habit when you don't hold the camera

DavidP wrote:
And here I was thinking that I should sit perfectly still while
taking the pictures. Now I know I should move back and forth. But
only along the lens axis, otherwise I'll get motion blur.

Mishkin2 wrote:

or ask Wilkinsons to move forth and back while you're firing a
burst of 21 frames.

Them sitting in the chairs with guitars is no good for fast primes

-- hide signature --

Mishkin

Keyboard shortcuts:
FForum MMy threads