One trend that has been positive for full frame, but negative for crop sensor is what I call the notion of scope of development. In this article I'm going to apply that primarily to image sensors.
Let's start with crop sensor (APS-C/DX), because there we see a clear problem.
Historically, crop sensor went from 6mp to 8/10mp, 12mp, 16mp, 20/24mp, and most recently 28/32mp. But you'll notice that most APS-C/DX cameras are "stuck" at 20/24mp.
Nikon's DX DSLR lineup basically shows the problem: D3500 and D5600 are 24mp, D7500 and D500 (and mirrorless Z50) are 20mp. So, by buying a higher priced camera, you're not getting any improvement or change happening at the sensor, particularly in terms of image quality or image capability. Realistically, one would expect the low end camera to be 20mp, the middle range cameras to be 24mp, and the top camera to be 28/32mp, but that didn't happen in Nikon's lineup.
Why?
Unfortunately, the market collapse coupled with the cost of sensor development has forced companies to re-use sensors. Nikon, in particular, targeted fixed price points with their four DX DSLRs, and to hit those, they not only had to re-use sensors, but they had to make other cuts, as well (the D7500 being a primary example of small cost cutting that took away functionality).
This is what I'd call "developing too narrowly." I'll throw a straw-man proposal out there: if Nikon had pushed out a D500 replacement at 32mp at a US$300 higher price (to recover higher development costs), it would sell as well or better than the current model. Why? Because of the customer perception that this pushes the high-end product higher. (I'm assuming here that Nikon could make a 32mp sensor that improves on a number of image impacting bits, as Canon has.)
The current DX DSLR price points are: US$400 (w/o lens), US$600, US$900, US$1500. That's a 3x price spread without a real change in image quality or image capability. Indeed, many complain that the two higher models are only 20mp instead of 24mp. Hey, Nikon, are those two cameras selling as well as you expected them to? Nope, didn't think so. You developed too narrowly.
In full frame things are a bit different. We have 24mp and 45mp cameras (D780 and D850) that are more broadly different (developed "widely"). Yet, again we have the top camera (D6) sitting at the lowest pixel count (20mp)—the same pixel count as its predecessor—and not selling as broadly well as it could.
Moreover, I've long lambasted Nikon for not making what would today be a D850h and D850x and a D6h and D6x (they stopped the h/x thing with the D3 model). The development costs of doing so are nil. Yes, there are stocking unit (SKU) issues to deal with, and pricing gets a little tricky, but again you have the very top customer being denied something while trying to keep a fixed set of price points. So, again, narrow development.
Narrow development always comes back to haunt you. If you don't extend, your competitors will.
Unfortunately, I don't see this trend to develop narrowly changing in DSLRs (it might in mirrorless). The dollars are no longer there to support it, at least to the Nikon corporate beancounters. While a wider development push in the next generation of DSLRs might actually sell more cameras than a narrow development push, the problem is that we're now down to such a low number that the dollars out (wider development) won't match up well the dollars in (increased sales to customers).
With no real increase in sensor-produced image quality/capability, this puts the onus on future development of DSLRs elsewhere. Neither Canon nor Nikon have shown that they're willing to do that, unfortunately. Their best teams appear to working on mirrorless now.
Sadly, whether or not we continue to get DSLR iterations in the future, the time periods between them are going to increase and the differences between them are going to go down, I suspect. I hope I'm wrong, but all the clues currently point this direction.
