I wonder how rigid the shooting conditions need to be to extract all those 150MP. Low ISO, lots of light, fast shutter speed, and large (diffraction avoiding) aperture - all rigidly bolted to a tripod / studio stand ?
Anything else and my guess is that the theoretical MP count will soon start to drop - e.g. a movement of just 8/100mm during exposure and you're already at a maximum of 75MP of real data.
These crazy high resolution backs really deserve some form of IB(ack)IS methinks.
I think it's pretty easy to overstate the effects of diffraction/movement for these new backs.
150mp backs have 25% more pixels in the vertical direction. On paper* that means you need a quarter stop faster shutter speed to hand hold, as compared to 100mp. Given that the IQ4 150mp spec is for another stop worth of ISO (which we will be testing in practice via ISO ramps) compared to the IQ3 100mp (max ISO of 25,600 rather than ISO 12,800) it should actually be hand holdable in
more circumstances than the 100mp.
Even compared to a 36 megapixel small format camera the pixel size implies only a one stop difference in minimum shutter speed for a given lens. Because of the larger mirror and focal plane shutter you might need one stop beyond that for a total of two. Two stops can be quite significant, but it's also not debilitating. Will you use the IQ4 150mp hand held with a long lens in twilight (as is possible with some micro 4/3rds cameras with 5-axis IBIS and low pixel counts)? No; at least not with perfectly sharp pixels at 100%. Will it likely be excellent for a very large range of hand held applications? Yes.
Diffraction wise these pixels are ~3.8 microns. We will do the aperture sweep testing we've always done when we receive our first prototype, but I strongly suspect f/8 will be the sweet spot, and f/11 with diffraction correction will look very good in print.
Moreover in both cases (diffraction and hand held speed) if you use the same rules of thumb as a 100mp back you won't get less detail; you just won't get as much
additional detail. It's like owning two cars, one of which is faster on the straightaways, but both of which have to decelerate to roughly the same speed on a sharp corner; the faster car is never
slower, but sometimes it's advantage is muted by circumstance.
Also, remember that IBIS and OIS only address camera movement. They don't address subject movement. Obviously in some circumstances it's quite nice (aesthetically) to have a slightly or very unsharp subject because that subject is moving. But when you want the subject to be sharp and the subject is moving IBIS/OIS does nothing to help that. I have many images from my Fuji XH-1 with surprisingly decent overall frame sharpness (due to it's very good IBIS) at surprisingly slow hand held shutter speeds, but with a subject where their face is ghosted between a smile and not, or a soft face because they were (only slightly) moving their head.
Finally, I have (personally) many medium format images that I treasure despite not being perfectly pixel-sharp at 100% review. Some images really benefit from extreme resolution (e.g. landscapes with textured subject matter throughout the composition). With others, it's really not that important or can even be detrimental. I own Fuji X cameras/lenses and still choose to use Phase One in a lot circumstances that others would pull out the Fuji, even when I know it will be challenging to get perfectly sharp images (at 100%), because I love the color, the lens look, the color, the tonality, the color, and the type/look of the grain/noise of Phase One. Did I mention the color? Torturing the car metaphor above, it's nice to drive in a luxury convertible on the way to Key West, even though the speed limit is (if I remember from years ago) 45mph most of the way. Cars and cameras often get talked about by their headline features (resolution and max speed) but the experience of using either is far deeper than one spec.
*We will be conducting practical tests in the coming weeks, not just assuming the math bears out in practice.