I don't think you need to apologize and defend your position. Folding@Home is always a volunteer project and you are free to join and leave at any time. Whatever computing cycles you can spare.
Usacomp2k3 wrote:5/ I have been burned out by the lackluster clients that have been released. (and I know I'm not alone in this). If Vijay would actually release a functional SMP client, i'd be much more likely to promote it. However, and I know I've said this before, it seems like he doesn't care at all about the people that are making his work possible. We are lab rats that get to live with the frustration caused by him putting his time elsewhere. Sure, the ps3 client is nice, but it's really hard to start a farm with those. How long after CUDA was released did it take him to finally port a client over to nVidia hardware?
They initially worked with ATI's stuff so it was difficult to port to CUDA, they basically had to recruit a parallel team (of 2 I think) to build this out. They are not a software shop, it's not like (good) developers fall off from some tree anytime they want. Quite frankly I would rather Vijay do more with the data and release more results that people can see than to open up that many fronts of different clients. They have to think about some sort of support structure. Looks like they managed to muddle through and I give them props for that.
Usacomp2k3 wrote:The biggest one, though, is why hasn't there been made available a business-friendly client? I predict that we could easily quadruple the worldwide production if it were easy to install and manage folding clients in a business environment. Be it Active Directory, or even something less complicated, if one could deploy the client remotely, How??? wouldn't have left (the flakiness of the client was his main reason.
Don't be so sure. Businesses look at the bottom line, and electricity costs are not insignificant. If you talk to beancounters they won't necessarily like the idea. A boss (or bosses) that wants to be good corporate citizens and/or publicity may participate, but he will still be looking at the cost and benefit for this.
Usacomp2k3 wrote:He didn't have time to babysit, and that's what the FAH project has turned into in the last couple years.). That was a huge blow to the program, and one that is still being felt.
Back when the single core client was in its early stages, and then there was Genome@Home, the program was an even bigger mess. I remembered seeing all those posts on the [H] and it wasn't a pretty sight. UD at the time was pretty "hands-off" so I stuck with that. The single core client has really gone a long way and all the rest are relatively new so I'm willing to give them time to mature.
The Model M is not for the faint of heart. You either like them or hate them.
Gerbils unite! Fold for UnitedGerbilNation, team 2630.