CPPSRB has other "fairness" attributes too. With backpay, you have the potential to earn full PPS for every share you've ever submitted. PPLNS throws away shares after a given time frame; you will never earn anything for them. I suppose that if the "N" in PPLNS is large enough, you get kind of the same thing (since shelved shares that are old enough are unlikely to ever be paid out).
No, even if "N" is relatively small, you
cant[/s] still expect to be paid the same as PPS for every share. A larger "N" just reduces variance. I'm not sure what "N" would equate with CPPSRB in terms of variance.
Over time both CPPSRB and PPLNS will converge to the expected value per share. I'm not sure what the difference in variance between the two is.
I didn't mean that it was the same as PPLNS, I just meant that CPPSRB is somewhat similar to PPLNS with a large "N", in the way that they both could pay for shares that are relatively "old".
Also, as un_ordinateur pointed out, CPPSRB could reasonably be portrayed as PPLNUS. Looks at CPPSRB from another angle.
I don't know if this makes a difference to you, but my original post contained a bizarre typo which changed the whole meaning of what I wrote. Fixed in the quote above.
Well, kind of yes, kind of no (in terms of it making a difference). Of course in the long run, PPLNS and PPS converge to equal the same amount. I have never stated otherwise.
But each system has its "intuitive" appeal to people who don't fully understand/believe in the Law of Large Numbers. Some people like everything to be accounted for, and to have steady, predictable earnings under PPS. Others like to gamble, to have the idea that they can earn more than their "fair share" by having shares double-paid under PPLNS.
Of course this will not be the only thing that is taken into consideration when choosing a pool. I think that Eligius gets a first glance from most miners due to its 0% fees.