Yeah, on one hand you have to be at least slightly sympathetic toward PG - because virtually any change gets a significant portion of the user-base red-faced & hollering. On the other, when you're asking people to freely donate both time and money to your project, you really need to go the extra mile to address concerns and needs. And this is where I see PG continuing to fall flat.
I don't mind retiring old systems - its been years since I ran the client on older P4 hardware - it wasn't that their PPD was ever reduced, but simply that they didn't have the HP to push beyond the low triple digits - making the cost / benefit ratio completely untenable. Add the too often seen client instability, and it simply wasn't worth the effort anymore. But those machines were outdated by years when they were no longer project relevant.
While this decision isn't the first of its kind, remember BA already was limited to 8+ cores, this invalidates existing high end systems that currently and easily achieve preferred deadlines. We're used to seeing our systems slowly age out, but this feels arbitrary and is not inline with actual performance. Reading through that FC thread, and probably echoed in team F@h forum threads across the web, you see examples of people who elected to go out and purchase the premium priced 12 threaded processors, or HT versions of more midrange 8 threaded models, specifically because it allowed access to BA WUs and the associated point production. While perhaps a good reminder that one should never buy based on F@h performance (how many times have we been bit now? lol), its a pretty tough pill to swallow. I would be PISSED if I were a F@h contributor and I'd just upgraded to a new 12 threaded proc. Instead, I just get to monitor my existing 8 threaded systems still achieve a sufficient cost/perf ratio to remain in the project.
Though perhaps not at the same level as many, I've supported and participated in the F@h project for years. I am tired of seeing decisions made that reduce enthusiasm and participation, tired of seeing dedicated users throw in the proverbial towel because of unresolved issues (the now standard client beta status and absolute disregard toward the deployment needs of super-folders being the biggest and most disappointing). I'm not throwing in the towel yet, but these decisions continue to curb my enthusiasm, and make me really question PG's decision-making process.
(non server speaking)
