I think the sudden awakening of this thread of nearly a decade old is connected with recent private conversation between Alan and myself, so I feel compelled to comment. The experiment above has one flaw in that it is more about a particular film, than it is about a particular property of a developing agent. It would be better to produce pairs of negatives, developed in Sease #1 (simple PPD-sulfite) and Rayco developer for different films. It is true, though, that the derivatives of PPD all produce very flat negatives on BW film, which, in my understanding, was exactly the intention. A seemingly Gargantuan effort to compare PPD/CD-4 on several films could be made simpler by making small volumes of developer and using a film patron as a tank, a technique suggested once in this forum by David Lyga,
On the theoretical side, in my understanding, the derivatives of PPD all differ from unsubstituted PPD in that they are only very weakly complexing metallic silver. Increase in the activity of the CDs is at least in part happening because this property is diminished. Also, sulfite is all but eliminated in a typical color developer, as it is a competitor for the dye production. I noticed that PPD-based developers, when used on currently available films, produce negative of specific appearance, featuring a strong sheen, which, however, is not a silver deposit that has to be wiped off, and not a typical dichroic fog either, both of which are uniform, while sheen is clearly proportional to the density. Substitution of CDs for PPD does not produce any such sheen on the film, or if it is present, it is very inconspicuous.
I do not have sufficient knowledge to further speculate on the mechanism, but I think that none of the CDs is an adequate substitute for PPD in the legacy formulas. The use of PPD in the legacy formulas is clearly based on the balance of a slow development with the main developing agent and the presence of PPD/sulfite, which acts as a grain reducer.
Again, claims that this happens due to the impact of physical development, such as made by Windisch and later, remain just a plausible speculation.