Transactional Synchronization Extension coming to Intel processors

I found this illustration on his blog easier to follow: <a href="http://software.intel.com/en-us/blogs/2012/02/07/coarse-grained-locks-and-transactional-synchronization-explained/" rel="nofollow">http://software.intel.com/en-us/blogs/2012/02/07/coarse-grai...</a>

How much of a performance increase can be seen with transactional processing?

I'd except PyPy guys are keeping close eye on this.