I agree Redmint that you have to have some sort of an eye on what's coming out soon. Upgrading just before Christmas, for example, would have been madness. There's a world of difference between that and sitting here when p67's just launched and trying to decide whether to wait though. By the time z68 comes out AMD's Llano will be just around the corner. I'm not saying Llano will be better, but another argument could be made to wait for proper reviews of it to come out before upgrading. This cold continue ad infinitum.
Meh, Llano shouldn't really be in the same class as Sandy bridge, better IGP no doubt, worse CPU no doubt, though depending on just how many changes there are to the cpu it might be better than people think.
Here's the problem though if you want an overclocked Sandybridge you have one mobo option and it disables the IGP, and quicksync which is by FAR the most important improvement in Sandybridge, quicksync is by FAR the biggest performance improvement in Sandybridge and you simply will not get it with any currently available P67 board.
The p67 board it turns off the best advantages, and has you paying for silicon you can't use, waiting for a mobo that doesn't waste the chip isn't a small issue.
Anyway by all accounts Bulldozer is out before Llano by quite a few months, and Z68 chipset might not be far before Bulldozer at all and that is most certainly worth waiting for before upgrading.
If you can get a 8 core bulldozer, with a cheaper mobo, that allows full overclocking that is marginally slower per core but has 100% more cores, Bulldozer could entirely wipe the floor with the current Sandybridge.
Its surprising the overclocking you're allowed to do on H67, but its severely limited in that it only works with Turbo still so you can't actively have a 4.2Ghz core all the time, if you're using a game loading 4 cores you'll have significantly less than that.
I really have no clue why Intel have done it to themselves. Also supply looks very poor for an Intel launch as their production capacity is just silly.
It actually, marginally reads to me that they brought forward Sandybridge because they don't have a hope in hell of beating a 8 core Bulldozer, and they just couldn't get the main chipset done in time, which again suggests the potential for a good reason to wait.
Lastly, afaik end of the year we'll be seeing 6-8 core i7 Sandy's, not 12 cores. At a guess we'll see 12 cores when 22nm is available and they'll be Ivy bridges, and they should be quite ridiculous chips by then. Though I believe they'll be fighting 16 core Bulldozers by then aswell.
