From: Wilba on
Paul Furman wrote:
> Wilba wrote:
>>
>> The Canon EF 50mm f/1.8 II "Nifty Fifty" has a reputation for two
>> shortcomings, 1) softness at wide apertures (OK from f/2.8), and 2)
>> erratic focus under difficult conditions (low light, shallow DOF).
>>
>> Many people claim that 2) is a result of the crudeness of the cheap
>> focussing motor and electronics in the lens, that those components are
>> not able to provide the required accuracy and control of motion of the
>> focus ring.
>>
>> But I wonder if 2) is actually a result of 1) - if the AF sensors have
>> fuzzy images to work with, how /could/ the system nail the focus in
>> difficult conditions?
>>
>> It would be interesting to see what happens when the AF sensors have
>> sharper images to work with (e.g. at f/2.8 or f/4), but my 450D refuses
>> to AF when the DOF preview button is pressed, so I can't test that.
>> External aperture perhaps?
>>
>> Any ideas for how these competing hypotheses could be tested? Is there a
>> consequence of either hypothesis that could be disproved empirically?
>
> If you are using the center AF point, sharpness shouldn't be a problem.

I can interpret that to mean several different things. What do you mean
exactly?

> In fact the shallower DOF should improve accuracy.

I think that's part of why the "crude mechanism" theory is so popular, but I
can't think of any way to test it. I'm not interested in theory or
speculation except that which leads to an experiment which proves something.

> One factor could be the phenomenon where focus shifts when stopping down,
> but I suspect they've included corrections for that in the firmware.

If that were a factor I would expect to see a consistent mis-focus, but
that's not what I get.

> There could also be issues with curvature of field, if you are concerned
> about something off to the side, etc.

I'm not. :- )


From: Paul Furman on
Wilba wrote:
> Paul Furman wrote:
>> Wilba wrote:
>>> The Canon EF 50mm f/1.8 II "Nifty Fifty" has a reputation for two
>>> shortcomings, 1) softness at wide apertures (OK from f/2.8), and 2)
>>> erratic focus under difficult conditions (low light, shallow DOF).
>>>
>>> Many people claim that 2) is a result of the crudeness of the cheap
>>> focussing motor and electronics in the lens, that those components are
>>> not able to provide the required accuracy and control of motion of the
>>> focus ring.
>>>
>>> But I wonder if 2) is actually a result of 1) - if the AF sensors have
>>> fuzzy images to work with, how /could/ the system nail the focus in
>>> difficult conditions?
>>>
>>> It would be interesting to see what happens when the AF sensors have
>>> sharper images to work with (e.g. at f/2.8 or f/4), but my 450D refuses
>>> to AF when the DOF preview button is pressed, so I can't test that.
>>> External aperture perhaps?
>>>
>>> Any ideas for how these competing hypotheses could be tested? Is there a
>>> consequence of either hypothesis that could be disproved empirically?
>> If you are using the center AF point, sharpness shouldn't be a problem.
>
> I can interpret that to mean several different things. What do you mean
> exactly?

Sharpness shouldn't be a problem in the center. I was doing some lens
tests with macro stacking software and just for curiosity tried some
really fast primes, much closer than they were designed for, not
reversed or anything... and was amazed that the center performance was
excellent wide open. That quickly degraded to 'hideous' outside the
middle third though.
http://www.flickr.com/photos/edgehill/4212576776/sizes/o/in/set-72157603231101723/

If you have extension tubes or a bellows, try it at extreme closeup.
That's a good way to make flaws much more obvious. It may not apply to
normal operation though.

>> In fact the shallower DOF should improve accuracy.
>
> I think that's part of why the "crude mechanism" theory is so popular, but I
> can't think of any way to test it. I'm not interested in theory or
> speculation except that which leads to an experiment which proves something.

Lots of ways to test. Do a series at the same spot but defocus manually,
first several starting from infinity, then several from closest focus,
then from just slightly off in each direction. Do that for something up
close then something 30 feet away.

You could try your stopped down theory by putting a piece of black paper
with a hole punched on the front, but might try testing first that it
doesn't make things less sharp due to non-optimal aperture placement.
Put the paper aperture in the right place in back and the lens becomes
telecentric, meaning light rays travel straight out the front and the
field of view is no wider than the front element.


>> One factor could be the phenomenon where focus shifts when stopping down,
>> but I suspect they've included corrections for that in the firmware.
>
> If that were a factor I would expect to see a consistent mis-focus, but
> that's not what I get.

The pattern could be complex though... depending how much you are
stopped down, maybe depending if it's close up or infinity...


>> There could also be issues with curvature of field, if you are concerned
>> about something off to the side, etc.
>
> I'm not. :- )
>
>


--
Paul Furman
www.edgehill.net
www.baynatives.com

all google groups messages filtered due to spam
From: Wilba on
Paul Furman wrote:
> Wilba wrote:
>> Paul Furman wrote:
>>> Wilba wrote:
>>>>
>>>> The Canon EF 50mm f/1.8 II "Nifty Fifty" has a reputation for two
>>>> shortcomings, 1) softness at wide apertures (OK from f/2.8), and 2)
>>>> erratic focus under difficult conditions (low light, shallow DOF).
>>>>
>>>> Many people claim that 2) is a result of the crudeness of the cheap
>>>> focussing motor and electronics in the lens, that those components are
>>>> not able to provide the required accuracy and control of motion of the
>>>> focus ring.
>>>>
>>>> But I wonder if 2) is actually a result of 1) - if the AF sensors have
>>>> fuzzy images to work with, how /could/ the system nail the focus in
>>>> difficult conditions?
>>>>
>>>> It would be interesting to see what happens when the AF sensors have
>>>> sharper images to work with (e.g. at f/2.8 or f/4), but my 450D refuses
>>>> to AF when the DOF preview button is pressed, so I can't test that.
>>>> External aperture perhaps?
>>>>
>>>> Any ideas for how these competing hypotheses could be tested? Is there
>>>> a consequence of either hypothesis that could be disproved empirically?
>>>
>>> If you are using the center AF point, sharpness shouldn't be a problem.
>>
>> I can interpret that to mean several different things. What do you mean
>> exactly?
>
> Sharpness shouldn't be a problem in the center. I was doing some lens
> tests with macro stacking software and just for curiosity tried some
> really fast primes, much closer than they were designed for, not reversed
> or anything... and was amazed that the center performance was excellent
> wide open. That quickly degraded to 'hideous' outside the middle third
> though.
> http://www.flickr.com/photos/edgehill/4212576776/sizes/o/in/set-72157603231101723/

No, it is a problem for the nifty fifty. You can see it in the shots I
posted a while back -
http://www.users.on.net/~alanw/Usenet/f1.8LiveViewFullsize.jpg (3.6MB)
http://www.users.on.net/~alanw/Usenet/f2.8LiveViewFullsize.jpg (3.4MB)

> If you have extension tubes or a bellows, try it at extreme closeup.
> That's a good way to make flaws much more obvious. It may not apply to
> normal operation though.
>
>>> In fact the shallower DOF should improve accuracy.
>>
>> I think that's part of why the "crude mechanism" theory is so popular,
>> but I can't think of any way to test it. I'm not interested in theory or
>> speculation except that which leads to an experiment which proves
>> something.
>
> Lots of ways to test. Do a series at the same spot but defocus manually,
> first several starting from infinity, then several from closest focus,
> then from just slightly off in each direction. Do that for something up
> close then something 30 feet away.

What consequence of which theory would this test, and how?

> You could try your stopped down theory by putting a piece of black paper
> with a hole punched on the front, but might try testing first that it
> doesn't make things less sharp due to non-optimal aperture placement.

How could I do that?

> Put the paper aperture in the right place in back and the lens becomes
> telecentric, meaning light rays travel straight out the front and the
> field of view is no wider than the front element.

You lost me somewhere there. :- )

Put the external aperture _behind_ the lens? And light comes out the
_front_?! I don't get it. :- )

>>> One factor could be the phenomenon where focus shifts when stopping
>>> down, but I suspect they've included corrections for that in the
>>> firmware.
>>
>> If that were a factor I would expect to see a consistent mis-focus, but
>> that's not what I get.
>
> The pattern could be complex though... depending how much you are stopped
> down, maybe depending if it's close up or infinity...

Yeah, let's just stick with what happens wide open.


From: Paul Furman on
Wilba wrote:
> Paul Furman wrote:
>> Wilba wrote:
>>> Paul Furman wrote:
>>>> Wilba wrote:
>>>>
>>>>> The Canon EF 50mm f/1.8 II "Nifty Fifty" has a reputation for two
>>>>> shortcomings, 1) softness at wide apertures (OK from f/2.8), and 2)
>>>>> erratic focus under difficult conditions (low light, shallow DOF).
>>>>>
>>>>> Many people claim that 2) is a result of the crudeness of the cheap
>>>>> focussing motor and electronics in the lens, that those components are
>>>>> not able to provide the required accuracy and control of motion of the
>>>>> focus ring.
>>>>>
>>>>> But I wonder if 2) is actually a result of 1) - if the AF sensors have
>>>>> fuzzy images to work with, how /could/ the system nail the focus in
>>>>> difficult conditions?
>>>>>
>>>>> It would be interesting to see what happens when the AF sensors have
>>>>> sharper images to work with (e.g. at f/2.8 or f/4), but my 450D refuses
>>>>> to AF when the DOF preview button is pressed, so I can't test that.
>>>>> External aperture perhaps?
>>>>>
>>>>> Any ideas for how these competing hypotheses could be tested? Is there
>>>>> a consequence of either hypothesis that could be disproved empirically?
>>>> If you are using the center AF point, sharpness shouldn't be a problem.
>>> I can interpret that to mean several different things. What do you mean
>>> exactly?
>>
>> Sharpness shouldn't be a problem in the center. I was doing some lens
>> tests with macro stacking software and just for curiosity tried some
>> really fast primes, much closer than they were designed for, not reversed
>> or anything... and was amazed that the center performance was excellent
>> wide open. That quickly degraded to 'hideous' outside the middle third
>> though.
>> http://www.flickr.com/photos/edgehill/4212576776/sizes/o/in/set-72157603231101723/
>
> No, it is a problem for the nifty fifty.

Huh? "No" what? It's not sharp in the center?


> You can see it in the shots I posted a while back -
> http://www.users.on.net/~alanw/Usenet/f1.8LiveViewFullsize.jpg (3.6MB)
> http://www.users.on.net/~alanw/Usenet/f2.8LiveViewFullsize.jpg (3.4MB)
>>
>> If you have extension tubes or a bellows, try it at extreme closeup.
>> That's a good way to make flaws much more obvious. It may not apply to
>> normal operation though.

This will tell you if it's sharp in the center.


>>>> In fact the shallower DOF should improve accuracy.
>>>
>>> I think that's part of why the "crude mechanism" theory is so popular,
>>> but I can't think of any way to test it. I'm not interested in theory or
>>> speculation except that which leads to an experiment which proves
>>> something.
>> Lots of ways to test. Do a series at the same spot but defocus manually,
>> first several starting from infinity, then several from closest focus,
>> then from just slightly off in each direction. Do that for something up
>> close then something 30 feet away.
>
> What consequence of which theory would this test, and how?

Testing the 'slop' theory. To see if it stops at different positions on
repeated runs of the same movement or different movements.


>> You could try your stopped down theory by putting a piece of black paper
>> with a hole punched on the front, but might try testing first that it
>> doesn't make things less sharp due to non-optimal aperture placement.
>
> How could I do that?

Just side by side comparison shots to see if the new aperture makes
things softer.


>> Put the paper aperture in the right place in back and the lens becomes
>> telecentric, meaning light rays travel straight out the front and the
>> field of view is no wider than the front element.
>
> You lost me somewhere there. :- )
>
> Put the external aperture _behind_ the lens? And light comes out the
> _front_?! I don't get it. :- )

Heh, sorry, off on a tangent. The idea is that adding an aperture like
that can do weird things, so maybe not sharper than stopping down the
lens' own aperture and who knows, maybe softer than wide open.


>>>> One factor could be the phenomenon where focus shifts when stopping
>>>> down, but I suspect they've included corrections for that in the
>>>> firmware.
>>>
>>> If that were a factor I would expect to see a consistent mis-focus, but
>>> that's not what I get.
>>
>> The pattern could be complex though... depending how much you are stopped
>> down, maybe depending if it's close up or infinity...
>
> Yeah, let's just stick with what happens wide open.

But you wanted to test if things improved stopped down. If I understand
correctly, the view in the optical viewfinder and presumably what the AF
sensors see, is already stopped down to maybe f/2.8; it doesn't benefit
from faster lenses, due to obstructions I guess. So there is already
some potential for focus shift there... but as you say, that should be
consistent, so again this points toward slop in the mechanism. If you
tested various final taking apertures, at least that's some kind of data
pointing to whether that might be playing a role.

The simplest test is that focus chart manually focused then stop down to
various amounts & see if the focus shifts. It will get wider DOF but
might not be centered if you count from each side. Then see if the AF
places the focus differently from the manual fixed focus test at
different apertures. If there is focus shift and the camera firmware is
correcting for it, there should be little differences in where the
camera decides to focus.

PS I still don't get how the AF calibration thing works on the cameras
that have that. I can see one master set screw for coordinating the
distance of the AF sensors to match the distance to the picture sensor
and or viewfinder ground glass but calibrating differently for different
lenses would seem to me to have to be relying on this idea of focus
shift when stopping down. If so, that's going to have to be checked at
various apertures to establish a curve for the correction, not just one
test shot. Does that make sense?

--
Paul Furman
www.edgehill.net
www.baynatives.com

all google groups messages filtered due to spam
From: Wilba on
Paul Furman wrote:
> Wilba wrote:
>> Paul Furman wrote:
>>> Wilba wrote:
>>>> Paul Furman wrote:
>>>>> Wilba wrote:
>>>>>>
>>>>>> The Canon EF 50mm f/1.8 II "Nifty Fifty" has a reputation for two
>>>>>> shortcomings, 1) softness at wide apertures (OK from f/2.8), and 2)
>>>>>> erratic focus under difficult conditions (low light, shallow DOF).
>>>>>>
>>>>>> Many people claim that 2) is a result of the crudeness of the cheap
>>>>>> focussing motor and electronics in the lens, that those components
>>>>>> are not able to provide the required accuracy and control of motion
>>>>>> of the focus ring.
>>>>>>
>>>>>> But I wonder if 2) is actually a result of 1) - if the AF sensors
>>>>>> have fuzzy images to work with, how /could/ the system nail the focus
>>>>>> in difficult conditions?
>>>>>>
>>>>>> It would be interesting to see what happens when the AF sensors have
>>>>>> sharper images to work with (e.g. at f/2.8 or f/4), but my 450D
>>>>>> refuses to AF when the DOF preview button is pressed, so I can't test
>>>>>> that. External aperture perhaps?
>>>>>>
>>>>>> Any ideas for how these competing hypotheses could be tested? Is
>>>>>> there a consequence of either hypothesis that could be disproved
>>>>>> empirically?
>>>>>
>>>>> If you are using the center AF point, sharpness shouldn't be a
>>>>> problem.
>>>>
>>>> I can interpret that to mean several different things. What do you mean
>>>> exactly?
>>>
>>> Sharpness shouldn't be a problem in the center. I was doing some lens
>>> tests with macro stacking software and just for curiosity tried some
>>> really fast primes, much closer than they were designed for, not
>>> reversed or anything... and was amazed that the center performance was
>>> excellent wide open. That quickly degraded to 'hideous' outside the
>>> middle third though.
>>> http://www.flickr.com/photos/edgehill/4212576776/sizes/o/in/set-72157603231101723/
>>
>> No, it is a problem for the nifty fifty.
>
> Huh? "No" what? It's not sharp in the center?

Correct. You can see it in the shots I posted a while back -
http://www.users.on.net/~alanw/Usenet/f1.8LiveViewFullsize.jpg (3.6MB)
http://www.users.on.net/~alanw/Usenet/f2.8LiveViewFullsize.jpg (3.4MB)

>>>>> In fact the shallower DOF should improve accuracy.
>>>>
>>>> I think that's part of why the "crude mechanism" theory is so popular,
>>>> but I can't think of any way to test it. I'm not interested in theory
>>>> or speculation except that which leads to an experiment which proves
>>>> something.
>>>
>>> Lots of ways to test. Do a series at the same spot but defocus manually,
>>> first several starting from infinity, then several from closest focus,
>>> then from just slightly off in each direction. Do that for something up
>>> close then something 30 feet away.
>>
>> What consequence of which theory would this test, and how?
>
> Testing the 'slop' theory. To see if it stops at different positions on
> repeated runs of the same movement or different movements.

From previous tests for other purposes, I can say yes, it does exactly that.
But the problem is, that behaviour is just as well explained by the "soft at
f/1.8" theory - if the AF sensor doesn't have sharp fine detail to compare,
I would expect it to show that small-scale variability.

>>> You could try your stopped down theory by putting a piece of black paper
>>> with a hole punched on the front, but might try testing first that it
>>> doesn't make things less sharp due to non-optimal aperture placement.
>>
>> How could I do that?
>
> Just side by side comparison shots to see if the new aperture makes things
> softer.

Ah, that's what you mean, OK, got it. :- )

Yes, I might have a play with some external apertures today. We know that
f/2.8 gives a sharp image, so I will find an external aperture which gives
the same exposure, and if the image is as sharp with the lens on f/1.8, then
I can try focussing through it. Alternatively, if the "equal exposure"
method doesn't work, I could try to find an external aperture which gives
the same or more sharpness as the naked lens at f/2.8.

>>> Put the paper aperture in the right place in back and the lens becomes
>>> telecentric, meaning light rays travel straight out the front and the
>>> field of view is no wider than the front element.
>>
>> You lost me somewhere there. :- )
>>
>> Put the external aperture _behind_ the lens? And light comes out the
>> _front_?! I don't get it. :- )
>
> Heh, sorry, off on a tangent. The idea is that adding an aperture like
> that can do weird things, so maybe not sharper than stopping down the
> lens' own aperture and who knows, maybe softer than wide open.

OK, I'll try to erase all that from my mind.

>>>>> One factor could be the phenomenon where focus shifts when stopping
>>>>> down, but I suspect they've included corrections for that in the
>>>>> firmware.
>>>>
>>>> If that were a factor I would expect to see a consistent mis-focus, but
>>>> that's not what I get.
>>>
>>> The pattern could be complex though... depending how much you are
>>> stopped down, maybe depending if it's close up or infinity...
>>
>> Yeah, let's just stick with what happens wide open.
>
> But you wanted to test if things improved stopped down.

Just to make sure no-one misunderstands that statement, I would like to be
able to auto-focus with the aperture stopped down.

> If I understand correctly, the view in the optical viewfinder and
> presumably what the AF sensors see, is already stopped down
> to maybe f/2.8; it doesn't benefit from faster lenses, due to
> obstructions I guess.

Absolutely not. The aperture is wide open except for the duration of the
shutter opening. To see what I mean, in aperture priority mode press the DOF
preview button while you change the aperture.

> So there is already some potential for focus shift there...

I might come back to focus shift later if nothing else works. :- )

> PS I still don't get how the AF calibration thing works on the cameras
> that have that. I can see one master set screw for coordinating the
> distance of the AF sensors to match the distance to the picture sensor and
> or viewfinder ground glass but calibrating differently for different
> lenses would seem to me to have to be relying on this idea of focus shift
> when stopping down. If so, that's going to have to be checked at various
> apertures to establish a curve for the correction, not just one test shot.
> Does that make sense?

It does. There is a lot of mystery in this area.