Re: New Lens Comparison: Nikon 50mm 1.4D vs. 50mm 1.4G
Get lost
> I had the opportunity to have in my possession both of these lenses- the
> older 50mm 1.4D and the new 50mm 1.4G.
>
> I took sample photos with both, as carefully controlled as I could. I
> think the results were quite surprising but remarkably consistent.
>
> I know which one is going back, for sure. The photos tell the story:
>
> http://home.comcast.net/~mhmyers/1.4dg.html
>
> --
> m-mhttp://www.mhmyers.com
G seems a little contrast challenged compared to the D, but overall
it's the better lens.
Thanks for the comparison.
Bruce
>I had the opportunity to have in my possession both of these lenses- the
>older 50mm 1.4D and the new 50mm 1.4G.
>
>I took sample photos with both, as carefully controlled as I could. I
>think the results were quite surprising but remarkably consistent.
>
>I know which one is going back, for sure. The photos tell the story:
>
>http://home.comcast.net/~mhmyers/1.4dg.html
An interesting test, thanks for posting.
How were the lenses focused? You have to be careful to ensure that you
focus manually, and extremely accurately, to make any comparison between
the optics a useful one. Then, you can do a second test, one of the
differences between the focusing systems.
The 100% crop images of "kitchen timer on/off" suggest the G lens gives
a less sharp image. But is it the optics, or is it an autofocus
problem? Only you can now that, and you either choose not to tell us,
or you don't know. If both shots were taken with autofocus, we will
never know.
Also, in the shots of the white vase, those taken at f/4 appear to show
an opposite conclusion to the one you have drawn about the G lens's iris
diaphragm. The differences in saturation appear to be differences in
exposure. Once again, were these shots manually focused?
Once again,if both shots were taken with autofocus, we will never know
whether any differences are down to the different optics, or an error in
the autofocus system.
That's the problem with tests that are not properly controlled.
Interesting all the same, and thanks for posting. I won't be
downgrading my 50mm f/1.4 AIS Nikkor to a G lens anytime soon, and the
nasty plastic D lens with its excessive backlash in the focusing
mechanism never attracted me either. ;-)
Bruce
>In article <jgs3s4p0583509s0u...@4ax.com>,
> Bruce <n...@nospam.net> wrote:
>
>> M-M <nospa...@ny.more> wrote:
>>
>> >I had the opportunity to have in my possession both of these lenses- the
>> >older 50mm 1.4D and the new 50mm 1.4G.
>> >
>> >I took sample photos with both, as carefully controlled as I could. I
>> >think the results were quite surprising but remarkably consistent.
>> >
>> >I know which one is going back, for sure. The photos tell the story:
>> >
>> >http://home.comcast.net/~mhmyers/1.4dg.html
>>
>>
>> An interesting test, thanks for posting.
>>
>> How were the lenses focused? You have to be careful to ensure that you
>> focus manually, and extremely accurately, to make any comparison between
>> the optics a useful one.
>
>I used aperture priority single area autofocus for all shots, aiming the
>focus point at the same spot for all comparisons- I felt that was the
>best way to ensure uniformity. Sometimes the G made a slight difference
>in shutter speed exposing 1/3 stop longer.
The only way to ensure uniformity is to use manual focusing and manual
exposure, applying the same shutter speed and aperture to both lenses.
Otherwise you are introducing variations in the autofocus and auto
exposure systems, which isn't what you are testing for.
Unfortunately, I don't think your tests tell us much, in spite of all
the effort you put in. Modern lenses such as the G lens are a nightmare
to bench test - which would otherwise have been a good route to take -
because of the difficulty of controlling the aperture. That makes it
all the more important to exercise control when testing on a camera.
John McWilliams
> In article <p384s4l3eopogpugq...@4ax.com>,
> Bruce <n...@nospam.net> wrote:
>
>> The only way to ensure uniformity is to use manual focusing and manual
>> exposure, applying the same shutter speed and aperture to both lenses.
>> Otherwise you are introducing variations in the autofocus and auto
>> exposure systems, which isn't what you are testing for.
>
>
> one lens wanted to up the exposure a bit for the same lighting
> conditions, this is important. I wasn't just testing the glass which is
> what your suggestions would do. I was testing the lens systems and how
> they react to similar conditions.
>
> I found the G quite often but consistently exposed about 1/3 stop
> longer. You would never know this if I used manual exposure settings.
But that part is the sensor and the camera's electronics reacting to the
perceived light strength to whatever formula that was selected (spot,
average, weighted ave, etc.) You'd expect differently internally
configured lenses to give different amounts of light to the sensor.
--
john mcwilliams
Bruce
>In article <p384s4l3eopogpugq...@4ax.com>,
> Bruce <n...@nospam.net> wrote:
>
>> The only way to ensure uniformity is to use manual focusing and manual
>> exposure, applying the same shutter speed and aperture to both lenses.
>> Otherwise you are introducing variations in the autofocus and auto
>> exposure systems, which isn't what you are testing for.
>
>
>The tests were designed to show the differences in the lenses, and if
>one lens wanted to up the exposure a bit for the same lighting
>conditions, this is important. I wasn't just testing the glass which is
>what your suggestions would do. I was testing the lens systems and how
>they react to similar conditions.
>
>I found the G quite often but consistently exposed about 1/3 stop
>longer. You would never know this if I used manual exposure settings.
>
>As for manual focusing, I aimed the focus point at the exact same spot
>for each comparison shot- again this is a further test of a lens system
>and how it works. The two models use different focusing motors- how
>could I ignore this when comparing lenses?
That depends whether you are testing the optics, the autofocus
mechanisms, or the auto exposure system. You have tried to test all
three at the same time; consequently the results tell us almost nothing
about the optics.
Forgive me, but it is the optics that matter above all. If the AF or AE
systems don't work properly, all you are doing is identifying faults in
those systems. Yet the optics make the images.
If you routinely use AF and/or AE, it is perfectly legitimate to test
those. But don't try to test two or three things together. The
objective of a properly controlled test is to isolate each factor as far
as practicable and test for it on its own without confusing it with any
other factors.
So is the G lens sharper than the D lens? We don't know, because you
used autofocus. Which has the higher saturation? We don't know,
because you used auto exposure.
Which lens should you send back? We don't know that either. You think
you do, but you don't, because any conclusions that you draw from this
uncontrolled test will be fundamentally unreliable.
Personally, I wouldn't bother with either of them. But that's because I
know that even the best autofocus systems cannot always be trusted, even
in optimal conditions.
Pboud
>
>
> That depends whether you are testing the optics, the autofocus
> mechanisms, or the auto exposure system. You have tried to test all
> three at the same time; consequently the results tell us almost nothing
> about the optics.
>
> Forgive me, but it is the optics that matter above all. If the AF or AE
> systems don't work properly, all you are doing is identifying faults in
> those systems. Yet the optics make the images.
>
> If you routinely use AF and/or AE, it is perfectly legitimate to test
> those. But don't try to test two or three things together. The
> objective of a properly controlled test is to isolate each factor as far
> as practicable and test for it on its own without confusing it with any
> other factors.
>
> So is the G lens sharper than the D lens? We don't know, because you
> used autofocus. Which has the higher saturation? We don't know,
> because you used auto exposure.
>
> Which lens should you send back? We don't know that either. You think
> you do, but you don't, because any conclusions that you draw from this
> uncontrolled test will be fundamentally unreliable.
>
> Personally, I wouldn't bother with either of them. But that's because I
> know that even the best autofocus systems cannot always be trusted, even
> in optimal conditions.
>
Better yet, he could create a hermetic room, with laser controlled
environmental monitors and do a full disassemble of the lenses to
independently test each element..
It's the only way to be sure.
P.
Bruce
>
>Better yet, he could create a hermetic room, with laser controlled
>environmental monitors and do a full disassemble of the lenses to
>independently test each element..
>
>It's the only way to be sure.
Now you're just being silly. I was merely pointing out that with so
many variables, and no attempt to control them, the comparison was
meaningless.
I do admire the OP for trying, though. At least he was sufficiently
interested to make the effort. Most people buy equipment and make no
attempt to test it, and instead make claims for its performance based on
some magazine review they read.
pboud
While I agree that your test would have been more comprehensive, what he
did was a fair low-end 'general lens' comparison. while it wouldn't
isolate any one item, it's (as you said) tons more than most people do.
P.
Bruce
>
>All right, you caught me :)
>
>While I agree that your test would have been more comprehensive, what he
>did was a fair low-end 'general lens' comparison. while it wouldn't
>isolate any one item, it's (as you said) tons more than most people do.
True - if it gave the OP peace of mind about his decision, it was worth
doing. What did seem clear is that, optically, the AF-S G lens was not
a significant advance on the AF-D lens. I already know that the AF-D
lens, and its AF (non D) predecessor were not a significant advance on
the AIS lens, so I am still happy with my AIS version.
The lens I would really like to see revised is the 35mm f/1.4 Nikkor. I
have the AI version, and it's a fine lens. The AIS version probably has
slightly better coatings than my AI. But there has never been an AF or
AF-D version, so the available AI and AIS versions are mostly well used.
This is a lens that would probably benefit more from updated coatings
than the 50mm f/1.4, so my wish would be for a 35mm f/1.4G AF-S Nikkor.
I would buy one in an instant.
"mcdonaldREMOVE TO...@scs.uiuc.edu
>
> I found that there is nothing improved in the G except for the iris
> diaphragm which only shows itself under certain conditions. I was hoping
> the lens coatings would fix the chromatic aberration of the D
Lens coating don't have any effect on chromatic aberration. Those
pictures showed a severe case of longitudinal chromatic aberration
(as opposed to lateral chromatic, which produces later blur
only at picture edges.) The purple fringe was very bad indeed.
Doug McDonald
Robert Coe
: In article <0kc4s49bv09sekpbl...@4ax.com>,
: Bruce <n...@nospam.net> wrote:
:
: > That depends whether you are testing the optics, the autofocus
: > mechanisms, or the auto exposure system. You have tried to test all
: > three at the same time; consequently the results tell us almost nothing
: > about the optics.
:
: The optics are about the same. I was not pixel-peeping but rather
: testing how well the lens takes pictures. There are plenty of
: pixel-peeping studies out there for these 2 lenses.
:
: I found that there is nothing improved in the G except for the iris
: just swapped purple for red and there is more red in the G than purple
: in the D. (Of course if you have a Nikon without a focus motor, you have
: little choice).
:
: It is also quite a bit larger and heavier and that is big to me. I am
: used to BIG telephotos and I really like to shoot narrow DOF like this:
:
: http://www.mhmyers.com/d80//DSC_3456dt.jpg
:
: (talk about creamy bokeh!)
:
: so I wanted a large aperture normal lens to get the same effect I get
: with the longer lenses but able to fit in my pocket.
:
: I returned the G today and I am satisfied with my decision.
I think there's only one sensible criterion for buying an f/1.4 lens: its
performance at maximum aperture. If you don't need the wide aperture (because
you don't shoot in low light and don't need critical control over DOF in
bright light), there are cheaper options that will do a better job. I'm not
sure you tested for low-light performance directly, but your tests do appear
to show that under comparable conditions the D lens gave sharper results than
the G lens at f/1.4. So given the limitations of the tests, I think you came
to the correct conclusion, i.e. that the D lens does its job better than the
G.
Bob