Yeah, I'm somewhere in the middle. I run some projects for pure profit. On the other hand, I also like crunching the "17 or bust" subproject on PrimeGrid, which has poor conversion into magnitude. The mathematical problem behind "17 or bust" fascinates me so much that I can't help but crunch on it.
N.b. I think this is an interesting dimension of Gridcoin which should be studied/discussed in more depth. To what extent might Gridcoin influence a redistribution of computational power on the BOINC network? Or is this offset by the fact that Gridcoin is attracting new users and/or encouraging existing users to contribute more computational power? What you describe here does indeed provide good evidence that the redistribution of computational power might not be so drastic. Many of us are, after all, science geeks, technologists, and altruists :D
But yeah, at some point if the future all of this may merit a more formal "study".
This is an interesting question. I'll think about, maybe there is a simple way to show it based on available historical RAC information. A topic for one of my next articles?? Thanks.