Archived posting to the Leica Users Group, 2006/06/15
[Author Prev] [Author Next] [Thread Prev] [Thread Next] [Author Index] [Topic Index] [Home] [Search]Don - I get what you are saying - but doesn't applying a mathematical shift to the number come to essentially the same thing just in the digital domain vs. the analogue? - the question is about the noise in the low bits of the signal - it should be fairly easy to simulate with an existing camera - under expose by a stop and 'push' the processing in PS RAW. You won't see any effect at ISO100, but at 400 or 800 I'd hazard a guess that it will be noticeable. Presumably the RD1 has the same issue, so it would be interesting to see actual results. I guess we will see when Leica start to show images. Mike -----Original Message----- From: lug-bounces+michael.francis=gs.com@leica-users.org [mailto:lug-bounces+michael.francis=gs.com@leica-users.org] On Behalf Of Don Dory Sent: Thursday, June 15, 2006 1:40 PM To: Leica Users Group Subject: Re: [Leica] Digital Leica M Mike, No. You are not correcting the signal by selectively amplifying certain pixels. You are using a math algorythum that calculates what the pixel should be based on location, lens, and aperture. There will be somewhat more noise in some situations because of light fall off but no more than would be in a deep shadow in the center of the frame. Don don.dory@gmail.com On 6/15/06, Francis, Michael <michael.francis@gs.com> wrote: > > Doesn't correcting for light fall off imply that the information at > the periphery of the image will be at a higher effective ISO and hence > far more noisy than the central part? So unless Leica is able to > squeeze lower noise out of the sensor (assuming a 1 stop falloff) an > ISO 400 image would only be ISO 400 in the middle and ISO 800 at the > periphery ... and even the Canon's struggle with noise when they are > that slow. I suspect that new 'digital optimized' lenses will be a > requirement for high ISO shooting at wide angles. Time will tell and I > hope they get it right as I'd love to be able to use my M Leica stuff for digital. > > Mike > > > -----Original Message----- > From: lug-bounces+michael.francis=gs.com@leica-users.org > [mailto:lug-bounces+michael.francis=gs.com@leica-users.org] On Behalf > Of Henning Wulff > Sent: Thursday, June 15, 2006 1:05 PM > To: Leica Users Group > Subject: Re: [Leica] Digital Leica M > > At 2:43 PM +0100 6/15/06, Peter Dzwig wrote: > >I can't help feeling that these dots and in-camera s/w adjustments > >smack more than a little of a job not done properly. An attempt to be > >"better" than the RD-1/1s that hasn't come off? Does the need for > >adjustment hint at some basic incompatibility between Leica's design > >and the M lenses? If Epson can build the RD-1 without this "dotting" > >why do Leica need it and a new mount. > > > >Sorry to say that I have this feeling that the "M-8" may only be a > >half-way house to something else. > > > >Peter Dzwig > > If the 8 bits can trigger certain software (or firmware) algorithms to > correct for falloff due to the angle of incidence on the sensor, and > possibly also to correct CA and other lens issues, then Leica has done > something useful. With the RD-1, you do this in post processing by > manually selecting options in the Epson software (which is poor like > most of the camera manufacturer's efforts) or Photoshop. > > Epson didn't do a 'better' job; they just avoided the issues, as you > can do with the Leica by not getting the dots put on. No big deal. > > Note that my assumptions about correcting for fall-off and CA are just > that; there could be other things that the dots are for but those two > are two of the most obvious. In any case, just because Leica does this > and Epson did not doesn't imply a failing on Leica's part, it just > implies that Leica is trying to make this same correction issue more > convenient. It also doesn't mean that shooting will be slower; image > processing has come a long way since the RD-1 came out, and that > camera certainly wasn't at the sharp end of development with respect > to processing speed. > > Maybe the Leica won't be better than the Epson; most likely it will. > The dots or lack thereof imply neither. > > Justifiying the price is another matter and has little to do with the > above. > > -- > * Henning J. Wulff > /|\ Wulff Photography & Design > /###\ mailto:henningw@archiphoto.com > |[ ]| http://www.archiphoto.com > > _______________________________________________ > Leica Users Group. > See http://leica-users.org/mailman/listinfo/lug for more information > > > _______________________________________________ > Leica Users Group. > See http://leica-users.org/mailman/listinfo/lug for more information > _______________________________________________ Leica Users Group. See http://leica-users.org/mailman/listinfo/lug for more information