How about penryn? they already have benchmark here http://www.anandtech.com/cpuchipsets/intel...doc.aspx?i=3069 , performance about 5% faster than equal speed 65nm core 2 overall. the only thing they haven't tested is SSE4 optimized software, maybe it should work best on divx encoding, about 110% faster i'm sure. Amd's offering will have to be as fast as penryn otherwise they will be in trouble,amd also have SSE4a instruction on their K10, amd's most advantages comes from their native quad core , unlike intels offering, where they glue together 2x dual core processor and named it core 2 extreme while amd natively made for quad core so it should work differently since it doesn't have to access the memory controller at the chipset, they have some compilers of their own processor as well , maybe applications needed to be recompile to work best on K10. Just wondering how will the L3 cache work

This post has been edited by Creative: Aug 23 2007, 01:51 PM