A 5K computer monitor is still just shy of 15 megapixels. Other than the idea of allowing the ability to crop in aggressively, there's not much benefit to the extra resolution for sharing images that will be displayed on monitors.
The real benefit is for printing... and printing big. But even that comes with some caution.
Any given lens has a limit to it's ability to resolve fine detail. Even when the sensor has the ability to resolve the detail, the lens does not necessarily have that capability. There's no point in using a 5Ds / 5Dsr with mediocre lenses.
But even the best lenses have limits to what they can do.... and it's not a problem of lens quality, it's a problem with the laws of physics. Light doesn't actually travel in a perfectly straight beam, it travels as a wave. The wave nature of light causes some interesting effects in physics when you really look closely. This was first noticed in astronomy because stars (while big) are so incredibly far away that they are effectively true "point" sources of light. They should not represent more than a single pixel... and yet they do. This isn't necessarily due to "pixel blooming" (that's a different problem) but due to the wave nature of light itself. The astronomer George Airy discovered that if you keep zooming in even with the best optical quality, what you get is a disk made of concentric rings (now called the "Airy Disk"). There's also the notion that whenever a "wave" hits any edge, it wants to "bend" around that edge (causing diffraction.)
This problem means that if you could have hypothetically "perfect" optics (if such a thing could exist) you would STILL be limited in the resolving power of the lens.
There is something called "Dawes' Limit" and the related Rayleigh Criterion. It turns out the resolving power of the lens (assuming perfect optics) is related to it's diameter. Lenses with a physically large optical diameter have the ability to resolve finer detail than a smaller diameter lens (even if other properties such as focal length are similar).
When I do astronomy outreach, I sometimes am asked if I can see the landing sites on the moon with my telescope with detail well enough to see things such as the flags, the rover, the lunar module (only the "descent stage" is left behind), etc. The answer is not only can my telescope not see it (and it's a big telescope)... you wouldn't be able to see it with any telescope on Earth. Nothing is large enough. By the physics of light, to see something as small as the lunar rover on the surface (just over 2 meters wide by about 3 meters long) you'd need a "lens" (telescope) with something slightly larger than a 25 meter diameter. The largest telescopes in the world are only about 10 meters (and it's quite an engineering challenge to build those.)
If you were to use a 5Ds or 5Dsr and expected to get the full value out of the 50 megapixel sensor, you need to stick with focal ratios of around f/5.6 or lower (assuming high quality lenses). At f/6.3 the camera will start to experience diffraction limits. It's heavily set in by f/8 and at f/11 there's literally zero benefit to the 50MP sensor (you may as well be shooting with a 16 MP camera).
This is why you see the comment from Bob that "if you need it, you'll know it". Most people don't need it.
Here's an article on lens diffraction by Cambridge in Color that's pretty useful: http://www.cambridgeincolour.com/tutorials/diffraction-photography.htm
Tim Campbell
5D III, 5D IV, 60Da
