Hi,
I have read and seen at several places that it is regarded that 180 PPI is needed for an excellent print. That figure is probably based on 20/20 vision and 50cm or 20" viewing distance. It essentially corresponds to normal vision at that distance.
Jeff Schewe discusses that a bit in the second edition of "Image Sharpening" uses the figure 172 PPI at 20" on page 77 and suggests 180 PPI for large prints on page 83.
But, we need to keep in mind that those figures are for resolution, not for fine detail contrast. Fine detail contrast is the amount of contrast a system can transfer for a given size of detail. A more technical version is called MTF. There are many nice properties of MTF, and one is that it is multiplicative. So you can take the MTF of the lens, the MTF of the diffraction at shooting aperture, the MTF of the OLP filter (if present) and the MTF of the sensor. Oddly enough, the MTF of sensor depends only on the size of the pixel aperture.
The other interesting thing with MTF is that it is low and medium frequency modulation that affects apparent sharpness and not high frequency detail. So, it may be argued that sharpening for print should be oriented towards low frequencies, instead of focusing sharpening actual pixel view on screen.
Now, there are more things than resolution and MTF. DR has gotten famous as it often was stated that some sensors have a great advantage over others, but DR essentially just says how much you can lift the darks, sometimes it is very relevant but mostly not very so. Just check how many great images were shot on Canons.
Another factor is noise. At normal exposures that is dominated by "shot noise", which is just natural variation of photons hitting the pixels. A larger sensor can detect more photons than a small sensor, so all things equal it will have less noise. Lower noise is good because it gives more liberty for math.
The ability of pixels to collect photons is normally measured in FWC, and the usual measure is number of electron charges. A sensor with larger FWC will have lower native ISO, as it takes more photons to fill it up and will have less noise and better DR. It is feasible that the new CMOS sensors have FWC a bit on the large size.
To that comes CFA design, some designs may be optimised for low ISO work and some may be more permissive.
The final and perhaps dominant part is raw conversion and processing. Modern gapless microlenses may help the demosaicer by offering less of point sampling and more of areas sampling. Colour is mostly decided by camera profiles.
So, there are a lot of parameters. It is feasible that the new sensors can deliver even better image quality than the old ones. But you can say that the old ones were "good enough". And you can also say that the new ones "do not improve the quality of my work".
Best regards
Erik
Slightly off topic of your issue, but I regularly print 36x48 from my IQ180 files at about 209 dpi on my Z5200. And they look fantastic with almost a 3D look and super fine detail. And that's on canvas.
I don't believe the 300 dpi hype. I think it'a a creation of sellers of more megapixels.