ComputerBase released some Yonah and Sossaman benchmarks. If you're interested you can find them Here.
Though they only did Cinebench, it does show the scaling compared to Dothan's and single core Sossamans. Overall I'd say it looks pretty good. We see very close to a 2x performance increase between a Dothan 2.13 and a yonah 2.0 (which will probably be really really close to exactly 2x between equally-clocked cpus). Sossaman, on the other hand, looks rather weak since it will require new infrastructure, etc and WILL NOT support 64bit, which is much quite good on the server side of things and will, ultimately be replaced by Whitefield (or whatever that's called) next year. Power consumption, of course, is excellent (95.6W with two Sossaman 1.5s is truly great).
Though they only did Cinebench, it does show the scaling compared to Dothan's and single core Sossamans. Overall I'd say it looks pretty good. We see very close to a 2x performance increase between a Dothan 2.13 and a yonah 2.0 (which will probably be really really close to exactly 2x between equally-clocked cpus). Sossaman, on the other hand, looks rather weak since it will require new infrastructure, etc and WILL NOT support 64bit, which is much quite good on the server side of things and will, ultimately be replaced by Whitefield (or whatever that's called) next year. Power consumption, of course, is excellent (95.6W with two Sossaman 1.5s is truly great).