Originally posted by GlassJunkie If you can manage pixel "crosstalk" (the extra light bouncing around pixels from the pixel next door), I would always take the higher res
I would always take the higher resolution file, too, but you have one big IF in there, and there's another - light hitting the 'wiring', that microlenses attempt to fix but do not do so perfectly.
File size and file transfer time, buffer time, etc would be adversely affected.
Those are some of the minuses.
The pluses? Better resolution, sure, but on a lp/mm basis the Q lenses are already the highest testing lenses out there to my knowledge. I doubt they'd improve much when you triple resolution, especially since, among lenses out there, they're already at their best wide open - i.e. diffraction is causing a significant portion of the softness.
If there were an 8mm F/0.2 or F/0.5 released, though, of course it would no longer be diffraction limited.
It's a case of diminishing returns, and on a smaller sensor with equivalently-slow lenses, you're close to the limit already.
---------- Post added 09-03-14 at 11:00 AM ----------
Originally posted by GlassJunkie It is getting closer. I don't know where it will end up.... Where the diminishing returns are with all of the variables:
- Marketing
- pixel pitch
- sensor size
- image circle
- human resolving need by media (what is it worth/ cost for print/ video/pc/phone)....
- in-camera processing
- aberration management (in-camera/ post processing) algorithms
- storage
- battery life/ power drain....
- portability
- ruggedness...
- optics... Glass will be a constraint(thresholds/ constraints on deliverable resolution to the sensor)
all are part of the equation.
How could you forget the most important reason? Fixed that for you....