Just to expand on what Donald said regarding plotting the characteristic curve--there are two different approaches floating around in this discussion. Either one is valid, depending on what you want to find out.
His suggestion, following more closely to what you've outlined as your plan, is that you determine your development time for a normal Zone VIII density and then use that time to develop the roll that you will plot from 0-X. This will give you the characteristic curve of the film at "N" development.
My original suggestion was to shoot, say, four rolls with exposures from 0-X, develop them at four different development times, and plot each one of them on the same graph. This will show you how the curve changes with changes in development time.
Yet another option would be to procede as Donald has suggested, and then do a series of rolls with exposures from 0-X, adjusting the development time by N-40%, N-20%, N+20%, and N+40%, and plot those along with the N curve, and that should get you in the ballpark for finding N-2, N-1, N+1, and N+2 development times, and will also show how the curve changes with development time.
This all can sound like a lot of testing, but you do it once, and you'll really get more out of the film you use. Sometimes when I'm traveling, for instance, and I don't want to carry too much equipment, but I know that I'll be moving between various lighting situations, I'll carry a medium format camera with one moderately wide lens and three backs, rather than carrying three lenses and one back, because it's easier to crop in the darkroom to improve composition than it is to massage a bad negative to get the right tonality.