>I partially agree with his statements. I do notice that I >shoot faster shutter speeds to get razor sharp images with my >D800 (or use a tripod). But if you shoot the D800 identically >to how you'd shoot a 12MP camera, then any apparent loss of >sharpness would be equivalent to the 12MP camera if you >downres the image to 12MP. Just don't expect to get super >sharp 36MP images with anything short of super fast shutter >speeds or strict adherence to proper shooting techniques. The >D800 magnifies errors, so what looks acceptable with a 12MP >camera may not look OK when you have 3x the pixels magnifying >your shortcomings during shooting.
What I am saying here applies to noise too. Now, let's compare a mouse to an elephant. If we have super-duper electronic microscope and be able to go on atomic level we would find out that mouse's atoms are exactly the same as elephant's. Yet one is many thousands times larger. Same thing here applied to the cameras. If you have a minute camera shake on 36mp you will see it a little more than on 12mp if you zoom in to pixel level but the beauty of it is that if you look at 12mp photo @ 100% you only be looking 2.5x smaller size from 36mp and thus will Not see the same minute camera shake because you are not looking @ 100% but only about 50%. And no, my calculations are not wrong.