[Leica] Noctilux DOF Film vs. M8

[pklein at 2alpha.net (Peter Klein)]

Adam: It isn't a question of microlenses. Focus shift as one stops down is 
an optical property of many fast lenses, including the double gauss and 
Sonnar families.

Mark R: The Noctilux focus shift is well-known and even documented by 
Leica. It happened visibly even on film. It isn't stupidity, or even 
hydrogen, it's physics. A newer design could minimize the issue--floating 
elements, aspherics, etc.

The sensor simply has less tolerance than film before an object appears 
out of focus.  So we notice it more with the M8.

In a way, a lot of this is a product of how good Leica lenses are, and the 
fact that the M8 has no anti-alias filter.  The AA filter in many DSLRs 
obscures minor focus shift issues, broadening the "peak" of apparent 
in-focus tolerance. The M8 gives you more, but is more unforgiving in 
return.

--Peter

Adam wrote:
> So I'm still not sure I understand everything I know about what's going 
> on with the Nocti and the M8. I'm assuming that what we're seeing is 
> some effect caused by the micro-lenses on the sensor that are causing 
> the focus shift problem? That's the only thing I can imagine that's 
> doing it. Everything else is a purely mechanical issue which Leica has 
> solved a generation or so ago.

I'm just trying to understand what part is really broken before I try to 
wrap my head around what a fix might look like. Because if it's either 
some dimensionality of the sensor, or of the micro-lenses, then the "fix" 
would seem to be that you have an M8 mated to a Nocti and you just leave 
the two together after getting the pair adjusted to work. Nothing else 
will match, you'll just come arbitrarily close.