Archived posting to the Leica Users Group, 2014/03/21
[Author Prev] [Author Next] [Thread Prev] [Thread Next] [Author Index] [Topic Index] [Home] [Search]Well now Howard that is quite a topic you have broached ;-) Before Dr Ted reminds me that only content matters ultimately (as is true of course) I shall dive in. For some attempt at clarity (post brevity not being my strength!) I will just talk about the specific hardware you mentioned, although the principles are more generally relevant. Wide open the lens you mentioned has 2.5 stops of vignetting. How much of that do you consider ought to be corrected out? There's no wrong answer of course just preference.That is what you got with positive film previously though. The camera corrections are also non-aperture dependent. That is to say that a single (compromise aperture value I guess) (less vignetting when stopped down) is corrected for because neither the M9 nor the M (typ 240) can reliably determine the exact aperture used due to the legacy designs. A new system (S & X for example) and I guess T? is not so limited. By brightness sensor value comparison estimate the M full frames might be within say a stop/stop and a half or two at worst. If the (single) correction value per lens was set at that for the worst case (wide open) you would get over-correction at smaller apertures. Actually odd lighter corners and at the expense of increased noise/ more loss of dynamic range there to do so. All correction is a compromise with some loss of quality in those corners. That may or not matter at all or be noticed. The camera is also making significant correction for every image for basic homogeneity because the 1954 fundamentals were just never designed for optimum use with a sensor.That includes asymmetric colour shift which is an optical reality with all systems more or less (Italian Flag) as well as so called red edge syndrome. That fundamental is why M digital sensors have their unique microlens arrangements in the first place and why the legacy wides in particular are compromised when adapted to other systems' sensor (Sony being the current prominent example). Phew, that ought to kick the discussion off, or get filtered out because my name is on the top! ;-) Cheers Geoff http://www.pbase.com/hoppyman On 21 March 2014 21:50, Howard Ritter <hlritter at bex.net> wrote: > In doing some preliminary exploratory shooting with my new M240 and the > previous-generation 35mm Summilux ASPH, I encountered the inevitable severe > fall-off of illumination at the corners, as I expected. What I did not > expect was that the M's built-in lens correction feature would reduce this > by only a subjective 50% or so, leaving a prominent and very disappointing > degree of vignetting still to be seen. > > I realize that this can be easily corrected in post-processing, e.g. > Lightroom, PS, and DxO, but my question is WHY? Why would Leica engineers, > after recognizing the problem, creating a software correction to it, and > deciding to incorporate that correction into the FF M digital camera, then > proceed to implement it in such a half-assed fashion? Clearly a full > correction is straightforwardly implementable in post-processing, so why > not write the firmware to accomplish it rather than hobble it to perform a > half-correction? > > Anybody know the reasoning behind this? Or am I missing some feature that > would actually give full correction? And when correcting for this in > Lightroom etc., what do most of you do? Let the camera do its bit and then > finish it, or simply dispense with the built-in correction and do > everything in LR? Will LR and the other software suites with built-in > corrections for various lens and body combinations even perform properly > with the M's built-in correction applied? > > Thanks for any suggestions. > > --howard > > _______________________________________________ > Leica Users Group. > See http://leica-users.org/mailman/listinfo/lug for more information >