85mm f1.4 vs 80-200mm f2.8

Paul Furman wrote:

> The reason given why entry level bodies don't work with cpu-less lenses
> is along these lines though; so the vignetting can be figured from the
> actual focal length & max aperture (and supposedly it's a cost saver not
> to include the capability of entering the data manually). That seems a
> pretty weak excuse but oh well.

The reason entry-level bodies don't work with non-CPU lenses is that they
lack the mechanical coupling to tell how far the lens is stopped down.
The mechanical coupling is what adds cost.

The thing to set the focal length and max aperture on the camera is just
to make the other electronic stuff work properly, like EXIF data. Metering
and shooting works fine if you set that wrong, or don't set it at all.

--
Jeremy Nixon | address in header is valid
(formerly jeremy.DeleteThis@exit109.com)
 >> Stay informed about: 85mm f1.4 vs 80-200mm f2.8 

Jeremy Nixon wrote:
> Paul Furman wrote:
>
>> The reason given why entry level bodies don't work with cpu-less lenses
>> is along these lines though; so the vignetting can be figured from the
>> actual focal length & max aperture (and supposedly it's a cost saver not
>> to include the capability of entering the data manually). That seems a
>> pretty weak excuse but oh well.
>
> The reason entry-level bodies don't work with non-CPU lenses is that they
> lack the mechanical coupling to tell how far the lens is stopped down.
> The mechanical coupling is what adds cost.

Hmm, I thought that was part of the same tab that stops it down during
taking. I think you may be thinking of even older pre-AI lenses which
had (various?) mechanisms for registering the aperture range by turning
through the full range after mounting, I'm fuzzy on that.


> The thing to set the focal length and max aperture on the camera is just
> to make the other electronic stuff work properly, like EXIF data.

I recall reading that was the reason. Perhaps I misunderstood. It sort
of makes sense that a fast lens would be a bit wonky wide open compared
to what you'd get stopped down.


> Metering
> and shooting works fine if you set that wrong, or don't set it at all.

I agree it's not a big deal, like if the 85/1.4 is a tad off. I do like
my exif correct though and the D200 works a whole lot better for this
than the D700. On the D200 I can dial in any aperture & FL on the fly
with the extra button on the front (if it's set that way), the D700
makes me menu dive & change one of the 9 presets I don't know how
the D300 & D3 work.

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

all google groups messages filtered due to spam
 >> Stay informed about: 85mm f1.4 vs 80-200mm f2.8 

M-M wrote:
> In article ,
> Wolfgang Weisselberg wrote:

>> Blaming the lens instead of the body on exposure compensation
>> problems is a new one

> you may have a point there.

What if there were something wrong with the iris control machinery in
the lens so it wasn't doing exactly what the camera was telling it to
do?

--
Chris Malcolm
 >> Stay informed about: 85mm f1.4 vs 80-200mm f2.8 

This message is not archived
 >> Stay informed about: 85mm f1.4 vs 80-200mm f2.8 

Paul Furman wrote:

> Hmm, I thought that was part of the same tab that stops it down during
> taking.

No, it's a separate function.

> I think you may be thinking of even older pre-AI lenses which
> had (various?) mechanisms for registering the aperture range by turning
> through the full range after mounting, I'm fuzzy on that.

There is no mechanical mechanism for the camera to know the absolute
aperture; that is only accomplished electronically. The camera only knows
how far the lens is stopped down from its maximum aperture, not what that
aperture actually is. That's why you have to set the max aperture manually
on the modern cameras: so they can display the aperture, set the EXIF data,
and do whatever modern electronic hand-waving they do for program modes so
they can add more items to feature lists.

The manual setting of the lens aperture and focal length, however, has no
bearing on the accuracy of the metering. The camera doesn't need to know
what aperture the lens is set for in order to meter correctly. You can
set the max aperture to something completely wrong and exposure will still
be correct.

> I do like my exif correct though and the D200 works a whole lot better
> for this than the D700. On the D200 I can dial in any aperture & FL on
> the fly with the extra button on the front (if it's set that way), the
> D700 makes me menu dive & change one of the 9 presets

Eww. You're kidding? A menu dive for that?

I was thinking a D700 would make a fine next camera. I hate menus.
No photographic function should ever require a menu, only the digital
computer stuff.

I do wish Nikon would make me a DSLR that is a camera first, and a computer
only incidentally. Digital FM3A, is my dream, only I'll concede that it
would have to require batteries.

--
Jeremy Nixon | address in header is valid
(formerly jeremy.RemoveThis@exit109.com)
 >> Stay informed about: 85mm f1.4 vs 80-200mm f2.8 

Jeremy Nixon wrote:
> Paul Furman wrote:
>
>> Hmm, I thought that was part of the same tab that stops it down during
>> taking.
>
> No, it's a separate function.
>
>> I think you may be thinking of even older pre-AI lenses which
>> had (various?) mechanisms for registering the aperture range by turning
>> through the full range after mounting, I'm fuzzy on that.
>
> There is no mechanical mechanism for the camera to know the absolute
> aperture; that is only accomplished electronically. The camera only knows
> how far the lens is stopped down from its maximum aperture, not what that
> aperture actually is. That's why you have to set the max aperture manually
> on the modern cameras: so they can display the aperture, set the EXIF data,
> and do whatever modern electronic hand-waving they do for program modes so
> they can add more items to feature lists.
>
> The manual setting of the lens aperture and focal length, however, has no
> bearing on the accuracy of the metering. The camera doesn't need to know
> what aperture the lens is set for in order to meter correctly. You can
> set the max aperture to something completely wrong and exposure will still
> be correct.

I just tested a 50/1.2 on the D200 at f/8 at a white ceiling:
f/1.2 1/2.5 sec
f/3.5 1/2 sec
35mm f/1.4 with same settings:
f/1.4 1/25 sec
f/3.5 1/15 sec

This is not a huge problem but appears to be a factor. My test might be
flawed somehow (I think shifting clouds made the two lenses different),
but the idea is; vignetting wide open on a very fast (or wide?) lens
will throw off the metering because the corners are dark and slower
lenses change less. I guess you could assume that's responsible of them
for a film camera but on digital it's so easy to chimp & correct or fix
in post-processing.

http://home19.inet.tele.dk/ne/nikon4.htm
[Auto Indexing Takes Over
AI type
Introduced in 1977, representing Nikon's first major change to the F
mount since it was introduced. It has a new method of coupling with the
meter called Aperture Indexing (AI). Previously when a lens was mounted,
the camera had to be indexed by manually setting lens's maximum
aperture, or by turning the aperture ring to the the smallest aperture,
then the largest aperture. The new system automatically indexes the
camera, enabling much easier and quicker lens changes.

AI-S.. 1981...
With the introduction of AF lenses a few years later, maximum aperture
and focal length information were transmitted electronically (more
reliably and accurately) so the mechanical linkages of AI and AI-S types
are something of a dead end.]

So Ai lenses did this mechanically and automatically without a CPU
during the act of mounting a new tab indicates max aperture. Pre-AI
apparently you had to work the aperture ring both ways to tell the
camera. Before that you had to just stop-down meter. That's what AI
means: Aperture Indexing.

Hmm, here's a theory for why the 85/1.4 overexposes: the camera uses a
canned formula for correcting assuming all fast lenses vignette but the
85 doesn't vignette much at all so it has out-performed the camera's
expectations. I think that's it because it doesn't appear to overexpose
wide open. Anyways it's not a big problem.


>> I do like my exif correct though and the D200 works a whole lot better
>> for this than the D700. On the D200 I can dial in any aperture & FL on
>> the fly with the extra button on the front (if it's set that way), the
>> D700 makes me menu dive & change one of the 9 presets
>
> Eww. You're kidding? A menu dive for that?

If you are happy with only 9 presets, you only have to menu dive once to
set them up. If you need 10 it's pretty awful. The extra button can
still be used to select the presets (if you set it that way). With the
D200 you use the front button plus front & rear dials for f/l & max
aperture, on the D700 only the rear dial with front button to choose a
preset. That's about my only complaint with the D700 but it does irk me.
I emailed Nikon suggesting a firmware update & got no response so I
assume it's intentional and the D3 doesn't have this limitation. Maybe
calling would be more effective.


> I was thinking a D700 would make a fine next camera. I hate menus.
> No photographic function should ever require a menu, only the digital
> computer stuff.
>
> I do wish Nikon would make me a DSLR that is a camera first, and a computer
> only incidentally. Digital FM3A, is my dream, only I'll concede that it
> would have to require batteries.


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

all google groups messages filtered due to spam
 >> Stay informed about: 85mm f1.4 vs 80-200mm f2.8 

Paul Furman wrote:

> I just tested a 50/1.2 on the D200 at f/8 at a white ceiling:
> f/1.2 1/2.5 sec
> f/3.5 1/2 sec
> 35mm f/1.4 with same settings:
> f/1.4 1/25 sec
> f/3.5 1/15 sec

There's just no way those readings match up, and there is no way that
corner falloff / vignetting is responsible. The lighting has to be
different there. The corner falloff on the 50/1.2 is not great enough
to matter here; I don't have a 35/1.4 myself, but it's next on my list.

> So Ai lenses did this mechanically and automatically without a CPU
> during the act of mounting a new tab indicates max aperture. Pre-AI
> apparently you had to work the aperture ring both ways to tell the
> camera. Before that you had to just stop-down meter. That's what AI
> means: Aperture Indexing.

I honestly know almost nothing about pre-AI; I've never used it or
really seen it.

But when you put an AI or AI-S lens on a modern camera, the only way
it knows the maximum aperture of the lens is because you tell it.
If you tell it the lens is f/4, it will believe you. It will still
expose correctly, but the camera will display f/4 instead of f/1.2
(or whatever) and so will the EXIF.

> Hmm, here's a theory for why the 85/1.4 overexposes: the camera uses a
> canned formula for correcting assuming all fast lenses vignette but the
> 85 doesn't vignette much at all so it has out-performed the camera's
> expectations. I think that's it because it doesn't appear to overexpose
> wide open. Anyways it's not a big problem.

I don't have an 85/1.4, either, but if the camera's meter is trying to
compensate for corner falloff, which would be bizarre, you could factor
that out by switching the meter to the old-fashioned center-weighted
mode.

("3D Color Matrix Metering"... feh. Kids these days, and all that.
You're choosing a shutter speed, not sending a probe to Mars.)

Using the matrix meter, you can remove any influence of the lens aperture
(which may well be part of the program) by lying to the camera about the
max aperture of the lens. Put the 50/1.2 on the camera and say it's f/5.6.

> If you are happy with only 9 presets, you only have to menu dive once to
> set them up.

Oh! I get it; you set up the presets in the menu, but don't need the
menu to switch among them. That's not nearly as bad as I thought you
meant. I feel much better now.

--
Jeremy Nixon | address in header is valid
(formerly jeremy DeleteThis @exit109.com)
 >> Stay informed about: 85mm f1.4 vs 80-200mm f2.8 

Jeremy Nixon wrote:
> Paul Furman wrote:
>
>> I just tested a 50/1.2 on the D200 at f/8 at a white ceiling:
>> f/1.2 1/2.5 sec
>> f/3.5 1/2 sec
>> 35mm f/1.4 with same settings:
>> f/1.4 1/25 sec
>> f/3.5 1/15 sec
>
> There's just no way those readings match up, and there is no way that
> corner falloff / vignetting is responsible. The lighting has to be
> different there. The corner falloff on the 50/1.2 is not great enough
> to matter here

I just checked the 50/1.2 on a D700 (a lot more vignetting than DX
crop). It gives a full stop brighter exposure if I tell it I'm using a
500mm f/4.5 lens, whether I shoot wide open or f/16. OK I went into the
darn menu & set it to 50mm f/4.5 & the meter gave almost 2 stops
slower/brighter. Center weighted versus 3D color matrix doesn't make
much difference. I don't really know if vignetting is the core reason or
what assumptions they make but something happens where specifying too
slow of a max aperture makes it overexpose.

Then I checked the 85/1.4 against the 35/1.4 (with correct lens data)
wide open and at f/8, the 85 overexposed 1/2 stop in both cases, none of
the comparable exposures matched. Stopping down really changes things
for both, the histogram looks all different for these 4 test shots and
this is a white ceiling with hardly any little wispy clouds to change
the light. I'm not much of a scientist but there is something going on
here. Active d-lighting (highlight recovery/fill light effect) could
interfere some too... don't make me menu dive to change that <g>. It
shouldn't matter with a white wall. I didn't have shades on any of these
and they all have big exposed glass on front so flare could be interfering.

>> So Ai lenses did this mechanically and automatically without a CPU
>> during the act of mounting a new tab indicates max aperture. Pre-AI
>> apparently you had to work the aperture ring both ways to tell the
>> camera. Before that you had to just stop-down meter. That's what AI
>> means: Aperture Indexing.
>
> I honestly know almost nothing about pre-AI; I've never used it or
> really seen it.

Me either... just reading that Nikon history page... & something else I
recall reading which said this.


> But when you put an AI or AI-S lens on a modern camera, the only way
> it knows the maximum aperture of the lens is because you tell it.

Modern cameras are missing the AI linkage apparently. They use the cpu
instead so you have to enter it manually.


> If you tell it the lens is f/4, it will believe you. It will still
> expose correctly, but the camera will display f/4 instead of f/1.2
> (or whatever) and so will the EXIF.

I agree it's not a big deal. Matrix metering is just nutty anyways,
there's no way to guess what it'll come up with but it's usually good.
In practice it's unlikely you'd have AI lenses vastly different like my
1.4 to 4.5 example.


>> Hmm, here's a theory for why the 85/1.4 overexposes: the camera uses a
>> canned formula for correcting assuming all fast lenses vignette but the
>> 85 doesn't vignette much at all so it has out-performed the camera's
>> expectations. I think that's it because it doesn't appear to overexpose
>> wide open. Anyways it's not a big problem.

Here's one more theory why the OP could be getting blown highlights with
the 85/1.4: more contrast from a better lens could make highlights a bit
brighter. Flare could effect things too.


> I don't have an 85/1.4, either, but if the camera's meter is trying to
> compensate for corner falloff, which would be bizarre, you could factor
> that out by switching the meter to the old-fashioned center-weighted
> mode.
>
> ("3D Color Matrix Metering"... feh. Kids these days, and all that.
> You're choosing a shutter speed, not sending a probe to Mars.)
>
> Using the matrix meter, you can remove any influence of the lens aperture
> (which may well be part of the program) by lying to the camera about the
> max aperture of the lens. Put the 50/1.2 on the camera and say it's f/5.6.
>
>> If you are happy with only 9 presets, you only have to menu dive once to
>> set them up.
>
> Oh! I get it; you set up the presets in the menu, but don't need the
> menu to switch among them. That's not nearly as bad as I thought you
> meant. I feel much better now.
>


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

all google groups messages filtered due to spam
 >> Stay informed about: 85mm f1.4 vs 80-200mm f2.8 

Paul Furman wrote:

> I just checked the 50/1.2 on a D700 (a lot more vignetting than DX
> crop). It gives a full stop brighter exposure if I tell it I'm using a
> 500mm f/4.5 lens, whether I shoot wide open or f/16. OK I went into the
> darn menu & set it to 50mm f/4.5 & the meter gave almost 2 stops
> slower/brighter. Center weighted versus 3D color matrix doesn't make
> much difference. I don't really know if vignetting is the core reason or
> what assumptions they make but something happens where specifying too
> slow of a max aperture makes it overexpose.

Well, there you go. That deserves further investigation. I don't have
a full-frame digital, but I'll try some things with the D2x when I get
a chance. You might be onto something.

If it's trying to outsmart a center-weighted meter reading, though, I
would be pretty annoyed. Not that I've actually noticed any kind of
pattern of error in actual use, but I use center-weighted specifically
because it's predictable and I can figure out what it's going to do
just by looking at the scene.

--
Jeremy Nixon | address in header is valid
(formerly jeremy.RemoveThis@exit109.com)
 >> Stay informed about: 85mm f1.4 vs 80-200mm f2.8 

Jeremy Nixon wrote:
> Paul Furman wrote:
>
>> I just checked the 50/1.2 on a D700 (a lot more vignetting than DX
>> crop). It gives a full stop brighter exposure if I tell it I'm using a
>> 500mm f/4.5 lens, whether I shoot wide open or f/16. OK I went into the
>> darn menu & set it to 50mm f/4.5 & the meter gave almost 2 stops
>> slower/brighter. Center weighted versus 3D color matrix doesn't make
>> much difference. I don't really know if vignetting is the core reason or
>> what assumptions they make but something happens where specifying too
>> slow of a max aperture makes it overexpose.
>
> Well, there you go. That deserves further investigation. I don't have
> a full-frame digital, but I'll try some things with the D2x when I get
> a chance. You might be onto something.
>
> If it's trying to outsmart a center-weighted meter reading, though, I
> would be pretty annoyed. Not that I've actually noticed any kind of
> pattern of error in actual use, but I use center-weighted specifically
> because it's predictable and I can figure out what it's going to do
> just by looking at the scene.
>

The metering is read through the focusing screen. The screen does not
accurately transmit any aperture wider than f4.5, as you can confirm by
making manual exposure readings and closing the aperture between f4.5
and f1.2 (even better, rig up a meter to read through the eyepiece, and
you will see how seriously 'adjusted' by the focusing screen the faster
apertures are).

The aperture you need to set manually (in absence of proper couplings)
will be specific to any one lens, and it will apply to any lens over a
certain aperture - f2.8 or faster I'd guess.

If you want to reduce the problem, fit a matt screen intended for wide
aperture or long lenses (for more accurate focusing). I don't know what
type this would be with Nikon but they should have one. It would need to
be a screen which needed either a user-entered or factory adjustment for
exposure once fitted.

David
 >> Stay informed about: 85mm f1.4 vs 80-200mm f2.8 

Chris Malcolm wrote:
> M-M wrote:
>> Wolfgang Weisselberg wrote:

>>> Blaming the lens instead of the body on exposure compensation
>>> problems is a new one

>> you may have a point there.

> What if there were something wrong with the iris control machinery in
> the lens so it wasn't doing exactly what the camera was telling it to
> do?

In this case you'd have a broken lens[1], and it probably wouldn't
show wide open.

-Wolfgang

[1] comparing with broken lenses isn't a good idea, unless you
are prepared to claim all lenses of that make are broken.
 >> Stay informed about: 85mm f1.4 vs 80-200mm f2.8 

This message is not archived
 >> Stay informed about: 85mm f1.4 vs 80-200mm f2.8