Pixel density and noise?

I think you are thinking too much about this. Both are great cameras that will produce equally good photos. The 5D Mark II and 50D have other big differences either then MP and sensor size. If you need speed; get the 50D.

tonylong wrote in post #7751265
Many full frame users comment on the fact that their images straight from the camera have more "pop", or "clarity" or "richness" than they were used to with the crop camers, and that consequently led to less post processing.

I agree that full frame images have more pop/clarity/richness, even when used at the same depth of field. I call it contrast.

Your assumption is that the difference is due to pixel size, but in reality it is the sensor size and perhaps other factors that are the cause for the difference.

Again, if pixel size was the reason, then a FF35 image cropped down to 1.6X size would still have superior contrast, pop, clarity, and richness. But it doesn't. The sensor size, not pixel size, is the reason for the improvement.

Contrast can be quantified in the form of MTF. The biggest reason for the improvement is because Canon lenses (and most lenses) have higher MTF at lower spatial frequencies. For a given print size, FF35 samples lower spatial frequencies than 1.6X. In order for FF35 and 1.6X to have the same image, the "crop factor" needs to apply to MTF. If it doesn't, then the image quality wont be the same.

So in order for an EF-S f/2.8 lens to have the same resolution and contrast as a FF35 f/4.5 lens, the MTF must be the same at 1.6X higher frequencies. That is, if the f/4.5 EF lens has MTF80 at 19 lp/mm and 14mm image height, then a EF-S lens must have MTF80 at 30 lp/mm and 9mm image height.

EF lenses do not have the same MTF at higher spatial frequencies (at best they're only the same), so EF lenses will always have more "pop/clarity/richness" on FF than crop.

tonylong wrote in post #7751265
A lot of people question this, suggesting it's psychological,

It's not psychological, it's real.

tonylong wrote in post #7751265
and more point out that you can even things out with post processing,

That's certainly true in some circumstances, but for me the full frame advantage is still tremendous.

tonylong wrote in post #7751265
but really there is logic to this perception. Each pixel has more light gathered leading to that higher signal to noise ratio, and meaning that the "raw material" is purer and richer than that gathered by the more compressed sensor.

That's only true for different spatial frequencies. When you normalize the spatial frequencies, the S/N is the same, not higher. This is explained and demonstrated in the links above. The per-pixel spatial frequency (Nyquist) of different sizes is at different spatial frequencies. Comparing S/N at different spatial frequencies is wrong.

It's like saying an engine with 20 horsepower per cylinder is better than one with 10 horsepower per cylinder. The horsepower per cylinder doesn't matter. All that matters is the total horsepower of the engine. For that, you need to consider the number of cylinders.

Smaller pixels may have lower S/N per pixel, but their S/N at any fixed spatial frequency is as good or better than larger pixels.

tonylong wrote in post #7751265
No one is suggesting that the 30D/40D/50D cameras are bad, but merely that the images from the bigger sensor do have an inherent quality boost.

Agreed.

I like my 5D and like FF, but other than the DOF/subject isolation "advantage" of FF (which is the major difference, IMO and the VF - had to put that in there for Perry) the actual IQ differences between FF and 1.6X crop get way overblown on these forums.

Here's an example that I've posted previously....

Same scene, same lens (24-105), different body (5D vs 40D)....(both shot RAW and processed with the same settings in DPP)...

5D, 47mm, f/11...