Went from native dual-core at 90nm to native quad-core at 65nm and increased the frequency apparently, AMD have still to manage this, so no small feat.
As x86 get more and more RAS features , the pie IPF has to content with shrinks rapidly.
IPF's pie is pretty static, HP Enterprise Servers and proprietary operating systems. It's biggest rivals at the moment are the legacy processors it is earmarked to replace and which it has to make look obsolete to trigger upgrades. By the time it has finished doing that then 32nm Poulson should be here and be more competitive against Power/Rock/x86 to maintain/increase share going forward.
Why weren't they able to fix Foxton and make Tukwilla rock ?
Foxton is probably permanently screwed, a bright idea that just didn't work.
Market Data 
Markets 
