I just read this.
Not that i am disapointed, but makes sense to me that something should be made.
IIRC, i did say when the hugeadv units started to be assigned that if Pande Group wanted them to be assigned to more powerfull hardware they should either increase the minimum requirements or impose a tighter deadline, or even a X6 like mine could do them. Things changed within 2 months and deadlines went from 7 days from 6903 and 10 days to 6904 to 4 and 5 days if i'm not mistaken. that put many X6 and 2500K out of the picture.
This said, i can't say i'm surprised that this will happen, and will throw many 2600K to regular SMP, as as i see it, there is no point in shortening the deadlines to more than 2 and 3 days to 6903s and 6904s respectively.
i do believe that with a core hack and Linux, the new 12 threaded Intel CPUs will crunch them within the deadline, but might be a little tight nevertheless.
i am quite pleased that the great majority of the bigadv folders in here are accepting this decision and remember than we're not into this for the points in the first place, we're here to help this project grow and do our part and our best to try to help Stanford getting their research results in order to help people, and i do believe that Stanford do know what is best to their project better than all the folding communities all over the world.
Points are fun, i like points, otherwise i would be folding under anonymous and to team 0, but i can live withe folding regular SMP units.
IIRC, my X6, at 3.8GHz and DDR2 800 can reach up to 24K PPD under SMP in Linux, with some 1xxx base points WUs, although i can't remember the specific project number, but from the top of my head i'll risk the 7200. It's not a bad number, considering that lately i would get 29-30K with regular bigadv. Also, i can aim to a higher OC as SMP isn't as demanding (in most cases though) as bigadv, meaning more points.
i will probably think a little more before upgrading, as maybe i won't buy a 2600K now. i may be looking for a multi-socket system now. Even a SR-2 with 2 Hexas (24 threads) will be able to do bigadv, and it's price won't be that crazy.
i can still remember a time when we needed to fire up 2 unicore clients when using a dual-core chip to obtain a little more PPD.
I can understand and relate to all of you that bought 2600Ks and Hexa-cores to dedicate to folding, but they can still beat our old-tech chips in SMP
Sorry for so much text