The last sentence that you quoted is what was missed. It states that XP2 is an exception to the norm. Why; I do not know.
An increase in granularity is what should be expected when density is increased for classic BW films, those who form an image with metallic silver and it isn't bleached - fixed at some stage. Colour negative films on the other hand have the opposite behaviour and granularity is decreased when density is increased. Kodak has published granularity - density charts for their ECN negative films and an example can be found here. Scroll down to "rms Granularity".
Overexposure increases graininess, since graininess is tied to density. IME, normal exposure and normal processing together are the best way to minimize grain in C-41 films.
Pro 400H is just not a grainy film, and especially not in 120. The OP must be printing up thin negs, which makes the grain in the film more apparent.
In low light situations, whatever advantages in grainlessness might be gained over the 400H by using new Portra 400 (and how big could they really be, given that 400H is not grainy?) would probably be overshadowed by the fact that the four layer technology of the Fuji makes it so much easier to color balance shots made in common low light color temperatures. I love Kodak's films, but Fuji's just plain balance easier when shot in foul lighting.
I agree with these points entirely, I would still be curious to know what size prints the O.P. is experiencing unacceptable grain on ?, is it home or trade processed ?, and what is he rating it at ?
I've always considered that over exposing a C41 film will decrease grain, and when shooting Portra I very much find that to be the case. I have found the Fujifilm C41 products to be a smidge grainy compared to the Kodak counterparts, this is based on getting them processed at some pro labs. It probably would not hurt for the OP to try shooting a frame or two of 400H at 320 ISO, and see if he prefers the results.
Interesting. I always overexpose my C-1 film when I want more grain. It works great. I do use grainy films to begin with when doing this, however (such as Fuji Press 800). Since the dyes ride on the silver that is bleached, it makes sense to me that the characteristics of the silver will determine the characteristics of the dyes. That should include grain. In the graph of the ECN film, the granularity appears relatively constant throughout the density range. I am no expert in the technical aspects of photo emulsions, but I do know via experience that overexposing color film is a way to get more grain out of it.
"Truth and love are my law and worship. Form and conscience are my manifestation and guide. Nature and peace are my shelter and companions. Order is my attitude. Beauty and perfection are my attack."
2F, it has been said previously by PE that C41 films are "coupler limited/starved". So, while increased exposure will increase the size of grain during development, the resulting dye clouds won't follow this behaviour. Since all silver must be removed, what matters is the behaviour of the dye clouds.
The bit about pushing the film three stops without changes in grain, saturation or contrast was interesting:
Also the most amazing thing about this film is it can be pushed up to 3200 with no grain increase, color shifts, or lack of contrast and saturation. I habe not even tried 6400 which i am goong to assume that it is 100% acceptable as when it was pushed to 3200 there was no visible grain.