Bring on the Megapixels!
As regards resolution – more MPs are better. Current sensor pixel density is still limiting resolution if you're using good glass and are not stopped down too far into diffraction territority.
As regards noise, with the current high fill-factors (micro-lenses; backside illumination), with the same sensor technology, noise is mostly dependent on the total amount of light falling on the total sensor area which should be the same for similar APS-C format sensors. Take a 10MP & a 40MP APS-C format sensor. The 40MP version would have sensels (the photosensitive part of a pixel) that are ¼ the area of the 10MP version (assuming a fill-factor approaching 100%). However the same amount of total light would fall on the two APS-C sensors, and if the resultant picture is printed out the same size, both images would appear to have the same noise level at mid-ISO levels, but the 40MP version should be sharper, unless either on-sensor NR (if CMOS) or PP NR was strong enough to wipe out the resolution advantage.
There is a noise difference between the two sensors which see-saws at low & high ISOs as mentioned in the summary below.
See
http://theory.uchicago.edu/~ejm/pix/20d/tests/noise/index.html,
particularly
http://theory.uchicago.edu/~ejm/pix/20d/tests/noise/noise-p3.html#pixelsize
His summary:
Among the important measures of image quality are signal-to-noise ratio of the capture process, and resolution. It was shown that for fixed sensor format, the light collection efficiency per unit area is essentially independent of pixel size, over a huge range of pixel sizes from 2 microns to over 8 microns, and is therefore independent of the number of megapixels. Noise performance per unit area was seen to be only weakly dependent on pixel size. The S/N ratio per unit area is much the same over a wide range of pixel sizes. There is an advantage to big pixels in low light (high ISO) applications, where read noise is an important detractor from image quality, and big pixels currently have lower read noise than aggregations of small pixels of equal area. For low ISO applications, the situation is reversed in current implementations — if anything, smaller pixels perform somewhat better in terms of S/N ratio (while offering more resolution)
...
Rather than having strong dependence on the pixel size, the noise performance instead depends quite strongly on sensor size — bigger sensors [i.e. larger than APS-C]
yield higher quality images, by capturing more signal (photons).
This recent, interesting paper looks at the optical challenges of going below 2μm pixels. Sony has already introduced one of the recommendations: Back-Side Illumination. While increasing sensitivity, BSI also reduces the dielectric stack height, paving the way for more pixels per mm2. The paper can't be hyper-linked to directly. You can get it by googling for
""Microlens performance limits in sub-2μm pixel CMOS image sensors".
It's at the Optic InfoBase site and there's a link in the Abstract section to a full text pdf.
Dan.