Intel Corp (INTC)

[Ixse]

Improvement K8L over rev F:

0. Native quad core
1. Hypertransport up to 5.2GT/s [HTT version 3]
2. Better coherency
3. Private L2, shared L3 cache that scales up.
4. Separate power planes and pstates for north bridge and CPU [allows higher frequency / lower power]
5. 128b FPUs - see 14,15 [probably better performing than NGA]
6. 48b virtual/physical addressing and 1GB pages
7. Support for DDR2, eventually DDR3
8. Support for FBD1 and 2 eventually
9. I/O virtualization and nested page tables
10. Memory mirroring, data poisoning, HT retry protocol support
11. 32B instead of 16B ifetch [Better prefetch, see also 12]
12. Indirect branch predictors
13. OOO load execution - similar to memory disambiguation
14. 2x 128b SSE units
15. 2x 128b SSE LDs/cycle [bit of an advantage over NGA as well]
16. Several new instructions

Coprocessors:
media processing
JVM/CLR acceleration
TOE, XML or SSL processing

The above and more can be found here: http://www.realworldtech.com/forums/index.cfm?action=detail&id=67239&threadid=67239&room....

Oversimplified K8L is to K8 what Merom is to Yonah.

And yes I think K8L improved enough that the dual core version of it will outdo NGA on most loads. This is provided Intel doesn't walk away frequency wise. Equally the quad core version will I believe outperform a native quad core version of NGA. I believe this because some people that know way more than me and that I've learned to rely upon up to a point are making similar sounds.

See also the Inq for a more detailed explanation: http://www.theinquirer.net/?article=31761

Regards,

Rink