From: Wilba on
Paul Furman wrote:
>
> Looking at that again prompted me to try another paper cutout test. I cut
> two holes in a black paper on each side like that, a little bit larger
> than the scale of that drawing and it works as a phase detect AF system.
> The whole scene through the viewfinder turns into double-vision ghosts
> that split apart and overlap when out of focus but it all snaps together
> when in focus. Pretty cool.

I couldn't make it work. How big did you make the holes and how far apart?


From: Wilba on
Paul Furman wrote:
> Wilba wrote:
>> Paul Furman wrote:
>>> Wilba wrote:
>>>> Paul Furman wrote:
>>>>>
>>>>> The wider rays from f/1.8 won't be seen by the AF sensor. It's view
>>>>> comes through a system that only gathers narrower f/2.8 angled rays.
>>>>>
>>>>> The wider rays will get clipped, bouncing around side surfaces, never
>>>>> reaching the AF sensor.
>>>>
>>>> I guess you're saying that the effective size of an AF sensor's virtual
>>>> aperture doesn't change with the lens aperture, which is fair enough,
>>>> because that's part of the explanation of why there is no ARFD
>>>> occurring in the AF system. Still, my system's performance meets that
>>>> standard (from an initial near side focus), so I'm not concerned about
>>>> that aspect.
>>>>
>>>> It would be interesting to get a better idea of the effective size of
>>>> an AF sensor's virtual aperture (I think I read somewhere that it might
>>>> be something like f/11, but I can't recall where). As long as we're
>>>> clear that "f/2.8" doesn't refer to that. :- )
>>>
>>> Heh, now I don't know what you're talking about :-)
>>
>> Isn't this fun!? :- )
>>
>> See figure 13 of http://doug.kerr.home.att.net/pumpkin/Split_Prism.pdf
>> (but read your way down to it from the beginning so that it makes sense).
>>
>> The light that reaches the AF sensor effectively comes through a virtual
>> aperture in the lens ("virtual AF apertures" in figure 13), as we have
>> been discussing for quite some time. If we knew the diameter of one of
>> those virtual apertures, we could work out the f-number for it. IIRC, I
>> read somewhere that it is typically something like f/11. As long as we
>> are clear that the "f/2.8" in "f/2.8 high-precision AF sensor" does not
>> refer to the size of the virtual AF aperture.
>
> OK, I see what you are saying now. Those apertures exist for the purpose
> of selecting rays from opposite sides of the lens so that you have two
> images to compare for the phase detect.

I'm uncomfortable with the way that sounds. Those virtual apertures don't
exist as real holes in anything anywhere, they just define the areas on an
ideal lens through which passes the light that ends up reaching the AF
sensor. As I understand it, what light through yonder window breaks is
determined more by the AF prisms than anything else (see figure 9 and the
two paragraphs above it). You have to start at the sensor and work back, not
the other way around.

> Fig 13 shows an f/2.8 AF sensor setup, so it's picking rays from near
> the edges of the lens and ignoring the less accurate central part by
> blocking it out.

Again, those rays aren't blocked by an aperture as such - if they reach the
prisms they don't end up shining on the AF sensor.

> There is still some space outside the green virtual apertures though,
> and that's the f/1.8 region, so this system only sees the f/2.8 rays.

It makes more sense to me this way - the system only needs an f/2.8 exit
pupil to illuminate the "f/2.8" AF sensor. An f/1.8 lens has a larger exit
pupil than required, it has "excess capacity". :- )

> Yeah, I know I'm reading stuff into the diagram but that's the idea I've
> been trying to convey about how if the AF system is designed for f/2.8,
> it won't be helped by a faster f/1.8 lens.

Sure, still for me the preferred connotation is that the "f/2.8" AF sensor
_requires_ an f/2.8 (or better) lens for effective illumination, rather than
there being a sense of optimisation to work well only with lenses of exactly
f/2.8.

Did you read all the way through Kerr's article? The very last section sums
it up nicely. (And AFAIK, the 20D system is a direct ancestor of the 450D
system, so I would assume that the principles hold for both.)

> I don't even know if this is a problem, maybe those virtual apertures
> do reach up to the edge of faster glass and that's just simplification in
> the diagram but it could be an issue.

No, I'm sure they snuggle up to the size of an f/2.8 exit pupil. An f/1.8
exit pupil would simply be larger than that.


From: Chris Malcolm on
Chris Malcolm <cam(a)holyrood.ed.ac.uk> wrote:
> Wilba <usenet(a)cutthisimago.com.au> wrote:
>> Chris Malcolm wrote:
>>> Wilba wrote:
>>>> Paul Furman wrote:
>>>>>
>>>>> The wider rays from f/1.8 won't be seen by the AF sensor. It's view
>>>>> comes through a system that only gathers narrower f/2.8 angled rays.
>>>>>
>>>>> The wider rays will get clipped, bouncing around side surfaces, never
>>>>> reaching the AF sensor.
>>>>
>>>> I guess you're saying that the effective size of an AF sensor's virtual
>>>> aperture doesn't change with the lens aperture, which is fair enough,
>>>> because that's part of the explanation of why there is no ARFD
>>>> occurring in the AF system. Still, my system's performance meets
>>>> that standard (from an initial near side focus), so I'm not concerned
>>>> about that aspect.
>>>>
>>>> It would be interesting to get a better idea of the effective size of an
>>>> AF
>>>> sensor's virtual aperture (I think I read somewhere that it might be
>>>> something like f/11, but I can't recall where). As long as we're clear
>>>> that
>>>> "f/2.8" doesn't refer to that. :- )
>>>
>>> Depends what you take "effective aperture" to mean. It could be taken
>>> as effective diameter of lens exploited, which gives the numbers like
>>> f2.8 and f5.6 often mentioned. It could be taken to be effective light
>>> capture area of the AF sensor as a fraction of the lens aperture. But
>>> since it doesn't look at the whole image, that esttimate could in turn
>>> be adjusted by effective change of focal length to account for the
>>> reduction on field of view. And so on.
>>>
>>> It all depends on what you want "effective aperture" to mean. Once
>>> you've defined what you mean by "effective aperture" it's just
>>> arithmetic on well known numbers to calculate it.

>> Use the precise meaning I gave - the effective size of an AF sensor's
>> virtual aperture. If you can tell me the well known diameter of the AF
>> sensor's virtual aperture, we can work it out. :- )

> But you still haven't explained what you mean by "virtual aperture" of
> an AF sensor. The only clue you've given us is that you probably don't
> mean what camera makers usually mean by it, since I don't think anyone
> claims a virtual AF aperture as small as f11, and the f2.8 many do
> claim you tell us is definitely not what you mean.

> So what do you mean?

That document is the only one I've seen to use "virtual AF aperture"
with that meaning. But if that's what you want then the next question
is what useful information you want to derive from that
information. For example, if it's virtual aperture with respect to
comparable light gathering power then it will need to be adjusted for
the (very approximately) half silvered mirror loss.

--
Chris Malcolm
From: Chris Malcolm on
Wilba <usenet(a)cutthisimago.com.au> wrote:
> Paul Furman wrote:
>> Wilba wrote:
>>> Paul Furman wrote:
>>>>
>>>> The wider rays from f/1.8 won't be seen by the AF sensor. It's view
>>>> comes through a system that only gathers narrower f/2.8 angled rays.
>>>>
>>>> The wider rays will get clipped, bouncing around side surfaces, never
>>>> reaching the AF sensor.
>>>
>>> I guess you're saying that the effective size of an AF sensor's virtual
>>> aperture doesn't change with the lens aperture, which is fair enough,
>>> because that's part of the explanation of why there is no ARFD occurring
>>> in the AF system. Still, my system's performance meets that standard
>>> (from an initial near side focus), so I'm not concerned about that
>>> aspect.
>>>
>>> It would be interesting to get a better idea of the effective size of an
>>> AF sensor's virtual aperture (I think I read somewhere that it might be
>>> something like f/11, but I can't recall where). As long as we're clear
>>> that "f/2.8" doesn't refer to that. :- )
>>
>> Heh, now I don't know what you're talking about :-)

> Isn't this fun!? :- )

> See figure 13 of http://doug.kerr.home.att.net/pumpkin/Split_Prism.pdf (but
> read your way down to it from the beginning so that it makes sense).

> The light that reaches the AF sensor effectively comes through a virtual
> aperture in the lens ("virtual AF apertures" in figure 13), as we have been
> discussing for quite some time. If we knew the diameter of one of those
> virtual apertures, we could work out the f-number for it. IIRC, I read
> somewhere that it is typically something like f/11. As long as we are clear
> that the "f/2.8" in "f/2.8 high-precision AF sensor" does not refer to the
> size of the virtual AF aperture.

"As we have all been discussing for some time"?

Since that document is the only one I've ever seen which uses "virtual
AF aperture" in that sense, and all the other uses I've seen, by
camera makers, reviewers, etc., have used it in the other sense of the
width of lens the pair stretch across, I'm pretty sure everyone (except
you) in this discussion has been using it in that sense too.

--
Chris Malcolm
From: Wilba on
Paul Furman wrote:
> Wilba wrote:
>> Paul Furman wrote:
>>> Wilba wrote:
>>>>
>>>> What do you see in this arrangement, back-focus or front-focus?
>>>>
>>>> Camera Subject Plane of focus
>>>
>>> Back focus.
>>>
>>>> And therefore this is...?
>>>>
>>>> Camera Plane of focus Subject
>>>
>>> Front focus.
>>
>> Excellent. So here we go.
>>
>> C = Camera,
>> POF = Plane of Focus,
>> DOF = Depth of Field,
>> S = Subject.
>>
>> When the plane of focus is a long way (i.e. much greater than the DOF,
>> >>DOF) in front of the subject -
>>
>> C POF<------ >>DOF ------>S
>>
>> I call that "near focus". You can think of it as initial gross
>> front-focus.
>>
>> If I start like that, autofocus and beep focus both put the plane of
>> focus coincident with the subject. Lovely.
>>
>> When the plane of focus is a long way behind the subject -
>>
>> C S<------ >>DOF ------>POF
>>
>> I call that "far focus". You can think of it as initial gross back-focus.
>>
>> If I start like that, autofocus and beep focus both put the plane of
>> focus just outside the DOF (>DOF/2) on the _front_ side of the subject -
>>
>> C POF<-- >DOF/2 -->S
>>
>> Summary - initial gross front-focus results in optimal focus, and initial
>> gross back-focus results in _front-focus_.
>>
>> This outcome -
>>
>> C S POF
>>
>> never occurs with my gear in my tests using autofocus or beep-focus.
>>
>> Let me know if that doesn't make sense. :- )
>
> That makes sense but doesn't match your previous diagram:

Yeah, but that's _your_ diagram, which is wrong. :- )

Here's mine again and we'll go through it line by line. Constant width font
required for scale. Disregard any previous misunderstandings - these are the
results. :- )

--------------------------------------

Beep Test (focus by changing the sensor to subject distance until the system
beeps to confirm focus)

Sensor to plane of focus distance -
|< 430mm >|
(The lens remains focussed at 430mm through the beep test.)

Theoretical DOF -
| |< 2.2mm >|< 2.2mm >|
(The calculated DOF goes from 427.8 to 432.2mm.)

Sensor to subject distance you get with an initial near-side focus -
|< 430mm >|
(Start with the subject 440mm from the sensor and decrease that distance.
When the system beeps the subject is perfectly in focus.)

Sensor to subject distance you get with an initial far-side focus -
|< 433mm >|
(Start with the subject 420mm from the sensor and increase that distance.
When the system beeps the subject is 3mm further away than the plane of
focus, i.e. front-focus.)

The "beep band" -
| |< 3.0mm >|
(The system will confirm focus over this range of sensor to subject
distances, for a plane of focus 430mm from the sensor.)

Outside the DOF -
| |<0.8mm>|
(In the initial far-side focus case, the subject ends up 0.8mm further from
the sensor than the far limit of the theoretical DOF.)

--------------------------------------

AF Test

Sensor to subject distance -
|< 430mm >|
(The sensor to subject distance remains the same throughout the AF test.)

Sensor to plane of focus distance you get with an initial near-side focus -
|< 430mm >|
(Start with the lens focussed at 420mm from the sensor and invoke AF. When
the system beeps the subject is perfectly in focus.)

Sensor to plane of focus distance you get with an initial far-side focus -
|< 427mm >|
(Start with the lens focussed at 440mm from the sensor and invoke AF. When
the system beeps the plane of focus is 3mm short of the subject, i.e.
front-focus.)

Theoretical DOF for an initial far-side focus -
| |< 2.2mm >|< 2.2mm >|
(The calculated DOF goes from 424.8 to 429.2mm.)

The "beep band" -
| |< 3.0mm >|
(The system will confirm focus over this range of sensor to plane of focus
distances, for a subject 430mm from the sensor.)

Outside the DOF -
| |<0.8mm>|
(In the initial far-side focus case, the subject ends up 0.8mm further from
the sensor than the far limit of the theoretical DOF.)

--------------------------------------