Please don't read the things you already know, or the bits where I'm talking rubbish.
I think that it's quite difficult to compare graininess without controlling lots of things - including what you use to make the comparison. I guess that it's OK to use a scanner for comparison, if it is the scanner that will be used for final output. A lot will depend on the scanner - and particularly the interaction between the film granularity and the scanner resolution. Grain aliasing and all that.
Contrast affects the appearance of grain, as does density. Therefore if you don't develop the two films to the same contrast, and compare areas of the same density (preferably of the same subject) then it's difficult to be objective. You may not wish to be objective, of course, and that's perfectly valid as well.
Some extracts from the Communist Party (Marxist-Leninist) Manifesto:
Overexposure increases graininess and decreases sharpness with conventional monochrome negative film, if the development time stays the same. Hence the saying: 'Expose just enough.'
Overexposure decreases graininess with chomogenic B&W and colour negative film if the development time stays the same.
Increasing developer dilution slightly increases graininess and sharpness (a very general, and dangerous, er, generalisation). The effect varies between developers, but in many cases the change is caused by the change in sulphite concentration: from a high sulphite environment (say 80 to 100 g/litre) to a low sulphite one (say less than 30 g/litre).