Re: [Leica] 2x extender R disappointment

[Dominique PELLISSIER <pelliss@droit-eco.u-nancy.fr>]

>From: "Henning J. Wulff" <henningw@archiphoto.com>
>Date: Tue, 4 Aug 1998 16:11:02 -0700
>Subject: Re: [Leica] 2x extender R disappointment
>
>Dominique wrote:
>
>>RE : the extender is a divergent system, more precisely a Barlow lens
>>according to Leica (Gunter Osterloh, angewandte leica technik, p.157 :"=
d=3D
>ie
>>Barlow-linse ist eine starke negativ-linse").
>>Suppose your lens has an angle of view of 27=3DB0 (summicron 90). With =
the
>>2X-extender behind the lens the angle of view remains the same but the
>>divergent system doubles the size of the projected image on the focus. =
S=3D
>o
>>the resolution of the image is divided by 2. If your lens at f:2 has a
>>resolution of, say, 20 lp/mm ; with a perfect extender, we'll have exac=
t=3D
>ly
>>10 lp/mm. BTW as the size of the projected image is multiplied by 2, th=
e
>>surface is multiplied by 4 and then the aperture is reduced by 2-stops.
>>Q.E.D.
>
>Dominique, in this case Eric is basically right. Adding a complex optica=
l
>system to another complex optical system results in a third complex opti=
c=3D
>al
>system, which can be less, or even possibly more than the sum of its par=
t=3D
>s.
>A system can be optimized for lens plus converter, so that removing the
>converter will actually reduce performance. If there were a converter wh=
i=3D
>ch
>just 'doubled' everything, without other regard to optical properties, y=
o=3D
>ur
>conclusion might be justified, but adding optical systems such as a
>converter can actually improve certain aspects of optical performance, a=
n=3D
>d
>actual performance losses are often a lot less than you surmise. A good
>converter often causes a loss of only 10 percent in lines per millimeter.
>Since there are a lot more things to optical performance than lines per
>millimeter, the overall loss can be greater, but hard to quantify. In th=
e
>case of short, fast teles used wide open, flare and loss of contrast are
>common in all lens plus converter combinations I have tried.
>
>As the converter in question is not part of an optimized system but a
>general purpose converter, the quality of lens plus converter will usual=
l=3D
>y
>be less than that of the lens alone, which is an optimized system on its
>own. That is about all you can predict in general; the degree of
>degradation is not fixed.
>
>
>   *            Henning J. Wulff
>  /|\      Wulff Photography & Design
> /###\   mailto:henningw@archiphoto.com
> |[ ]|     http://www.archiphoto.com

Comment :
I'm not an engineer in optics. I only remember simple experiments with a
divergent lens behind a convergent lens (Barlow principle).I think i'm
right but i'm going to put the question to God (i.e.Leica).
You write :"adding optical systems such as a
converter can actually improve certain aspects of optical performance, an=
d
actual performance losses are often a lot less than you surmise. A good
converter often causes a loss of only 10 percent in lines per millimeter".
When the french review Chasseur d'images(n=B0 116) tested the apo 2.8/280
with the extenders, the results were deceptive.At 2.8 without extender
:very good (center), good (edge). With the apo 1.4-extender : good,
fair;with the non apo 2x-extender :fair, poor.
I do think that the resolution is mathematically cut in two with the
2x-extender.Osterloh in Leica Angewandte leica technik is not clear :"Bei
den im Prinzip ahnlich aufgebauten echten tele-objektiven erfolgt die
Kompensation der Restfehler des negativen Hintergliedes durch
entgegengesetzte Fehler, die der Optik-Rechner bewusst im positiven
Vorderglied des Objektivs belasst.Das ist bei der Kombi Objektiv+extender
naturgemass nicht gegeben und die Bildqualitat wird dadurch mehr oder
weniger negativ beeinflusst."(p.159)

Dominique