Ringold, you have risen some important points. But...
It hard to compare success in term of credit generated since:
1. credit is still not very comparable among projects (you can get +50% on some projects than others with same HW) or at least haven't been in past (we have seen SETI giving x5 extra credit which made a lot of bad blood among SETI and BOINCers in general).
2. SETI is an old-timers project comparing to Eintein and Rosetta for example. It's user base was enourmous (I would call that a succcess) and some users were lost during classic -> BOINC transition as you noted. (CPDN has also classic Windows client, but user base was small during the time as HW demands were quite high, user were impatient while waiting to complete a climate model and the transition to BOINC took place quite soon).
I would not judge project by credit generated (see points above).
A number of active participants/machines is a good measurement of project success or let's say 'popularity'.
The problem is - it says nothing about project results it term of science or it's goals hence nothing about it's effeciency (effort and money put in, results out). It this term only, SETI does not do very well, IMNSHO.
Yet it serves pretty well in getting 'masses' into DC and computer aided science.
BOINC development has a separate funding of SETI, even some developers are/were active in both. I known I would be disliked if I say SETI is a beta for BOINC, but there are and were many features that comes from SETI specific needs and some general features were developed later.
It would be far from accurate to put SETI developers in one with BOINC developers; on the other hand, some BOINC projects added a lot to BOINC (generation of 4.x BOINC was mainly driven by CPDN needs when BOINC lacked some basic functions needed by CPDN, BURP gives a lot to BOINC forum code etc. etc.).
I think there is a good cooperation of BOINC developement effort and particular BOINC projects; SETI has no privileged position there.
It remains a question how to bring the spirit of classic SETI back.
Only a very few participants know something about science behind SETI (proportion of those interested in science being undertaken on other projects is considerable higher) and credit issue is dominating SETI forum as never seen on any other project.
I personally lack science papers published by SETI..perhaps there are not many results to discuss so far. Hence were are back to what you cited in your post - cost is a bit high when in comes to results to my taste. But you have also provided a solution so only time will tell...