Archived posting to the Leica Users Group, 2003/04/26

[Author Prev] [Author Next] [Thread Prev] [Thread Next] [Author Index] [Topic Index] [Home] [Search]

Subject: [Leica] Digital slightly OT
From: Frank Dernie <Frank.Dernie@btinternet.com>
Date: Sat, 26 Apr 2003 12:10:33 +0100

If Olympus are right, and their statement in
http://www.four-thirds.com/index_01.htm
seems plausible to me, maximising the performance of a digital sensor 
may require hardware redesign.
If the depth of the chip filter sandwich is significant rays impinging 
on the sensor at an angle will always have a different effect to those 
impinging normal to the sensor. This will obviously be most severe the 
nearer the rearmost lens element is to the sensor, making M lenses less 
suitable than R for digital work.
At first I thought it would be worst in the corners and for lenses with 
small diameter rear elements because in this case none of the rays 
impinge normal to the sensor. In fact however large rear elements will 
suffer also because whilst some corner rays may impinge almost normally 
the diagonally opposite rays will come from the opposite side of a 
large rear element at a large angle of incidence.
Erwin has lectured us on the incompatibility of the Hexar RF with Leica 
lenses due to a possible (depending on individual tolerance buildup) of 
the film to flange distance. The error he considers unacceptable is a 
max of 0,09mm (0.0035") in the worst case. His Leica acceptable error 
is 0,05mm (0.002). I hope I am not misinterpreting in assuming from 
this that unless the effective thickness of the sensor/filter sandwich 
is considerably less than this Leica M lenses can not achieve their 
potential. Clearly this problem is more significant the larger the 
sensor.
Olympus propose their new E system standard with a smaller chip and 
bigger lens throat with lenses designed for parallel rays at the sensor 
(as they say they do in their E10 and E20 digital cameras).
I have 2 questions.
1. If this is the case how can "parallel" rays ever be in focus at the 
sensor surface?
2. If not is there an optimum sensor size beyond which the reduction in 
performance in the corners due to optical and sensor thickness effects 
is more than the benefit of sensor size?

I obviously have not got enough to occupy me this morning!

frank

- --
To unsubscribe, see http://mejac.palo-alto.ca.us/leica-users/unsub.html

Replies: Reply from Dave Rodgers <drodgers1@lightcurves.com> (RE: [Leica] Digital slightly OT)