Originally posted by Glorfindelrb You don't need the latest sensor available either. A camera is a complex equilibrium between a sensor, a processor, a bus to transfer data and mechanical parts, you cannot drop any sensor in it as you were dropping film. It's possible but you will have 1 maybe 2 upgrades before needing a new processor. I don't think it's worth the cost in development constraints and size.
I buy 1 of 3 Pentax flagship DSLR up to now K20D --> K5IIs --> K-1 ? and I don't feel like my sensor is lagging behind that much (ok there is a trick, the K5IIs is an exceptional camera but still).
Your points are doubtless valid, for currently implemented designs.
However, I was indulging in some "blue sky" thinking. That is, I suggest a start with a clean piece of paper and to design for modularity. Where there is a will there is a way. However, maybe the market does not want it, only crazy old people like me.
---------- Post added 01-19-16 at 03:21 PM ----------
Originally posted by RonHendriks1966 Basicly K-5 was the big step forward and that sensor is good for a decade.
That sounds dubious to me. Who knows when the next sensor breakthrough will occur?
It seems to me, as an engineer, that if the mechanical, optical and electronic interfaces were standardised appropriately (c.f. K-mount, still going strong after 5 decades or so) sensor interchangeability could be achieved. It seems to me that many people seem to want to change their cameras with the same frequency and regularity that they change cell phones, or socks.
However, I defend everyone's right to spend their lucre as they wish