MGrayson
Subscriber and Workshop Member
Jack,And that's due to the oversampling effect I was discussing above. Matt, I do not believe this is in direct conflict with your comments, but oversampling in digital has a tangible benefit if sampled at Nyquist (2x base resolution) or above. Without going into the math, we basically impart an interpolated artifact at the midpoint (½ Nyquist); and voila, more perceived resolution... In Pegelli's example, it was enough to clarify the data accurately on the 6 to 7 line pairs -- though note we have a hint of a false artifact about ¾ way up that line pair progression in a thicker or darker line. I would further guess that the 7-8 line pairs and above are below Nyquist for that sensor. Would a 200MP sensor generate realistic detail between the 7-8 pairs? I have no idea, but suspect they might, though probably not as accurately as they did for 6-7...
Oh, I'm not addressing oversampling at all. I'm assuming that the sensor has a proper AA filter, which they never do, these days, which surprises the heck out of me. But:
In theory, there's no difference between theory and practice. In practice, there is.
Matt