On May 2, 2007, at 7:45 AM, William Robb wrote: >>> If that were true, then TTL metering in the general case wouldn't >>> work consistently for different lenses in any exposure mode at >>> different lens openings. >>> >> When lenses are wide open, it's less of a point source though, >> right? > > I don't think so, the lens is still covering the full format, but > the light > is dimmer. Large format lenses often increase the size of the image > cirgle > as the lens is stopped down.
I don't know how the notion of whether a lens appears to be a point source or not relates to a sensor reading light emanating from a diffuse imaging screen. Once light hits the screen's focusing surface, is scattered then collimated by the screen's built-in fresnel lens, whether the lens is acting as a point or diffuse source is of little concern. From the point of view of the sensor or film, yes: as a lens is stopped down, it becomes more of a point-source illuminator. This is why dust on the sensor of a digital camera is more apparent at small lens openings, the edges of the dust blob's shadow are better defined with a point source illumination. >> In any even, I'm talking out my ass... I don't know much about >> optical lens design. I could also be confusing the metering >> inaccuracy >> I've seen with my *ist-DS with the 3rd-party, split-prism focus >> screen. I >> don't recall if I had issues with metering inaccuracy with K/M >> lenses and >> the stock screen... been too long since I used it. > > I'm thinking that the metering is just plain innacurate as the > light levels > to the sensor drops closer to the sensitivity cutoff. I presume you mean the metering sensor, not the imaging sensor. I don't think this is the case. In the series I presented the other day, the problem becomes apparent well above the point where attenuation by a small lens opening in sunlight should reach the sensor's bottom threshold limit. I'm as certain as I can be without actually seeing the firmware source code that it is a calibration issue for the stop-down metering mode. One further test will confirm this definitively: if I replace the K10D's original focusing screen with the *ist DS models focusing screen, I would expect that the metering calibration will be on the money for the stop-down metering mode and will be off for the open- aperture mode with A-series and later lenses. That will prove the theory. Haven't had time to do the test yet. Godfrey -- PDML Pentax-Discuss Mail List [email protected] http://pdml.net/mailman/listinfo/pdml_pdml.net
