[WBEL-users] White Box wins election on Super Tuesday

[Bill Davidsen]

William Warren wrote:
> many many benchmarks show the Athlon 64 either neck and neck or 
> outpacing the HT p-4's.  The p-4's simply are too inefficient to run too 
> far ahead.  I ma not saying p-4's are not good cpu's..and they do have 
> application specific advantages..but right now in terms of 
> price..overall performance..and heat(Athlon 64's are at a max of 84 
> watts where the new prescott's coming out are rated at 120 watts) The 
> theory of faster interrupt handling is disproved by realistic 
> performance metrics that are all over the web..:)

I honestly don't see many benchmarks which are overly good at measuring 
interrupt latency, and fewer yet which use Linux to test, particularly a 
2.4 vs 2.6 on each bench. If I had the hardware set up, I would love to 
run a little test I wrote to measure IPC behaviour. Unfortunately unless 
I get a cure for sleep the hardware will sit in boxes for another month.

Note that in most applications this is a non-issue, but one 120w HT CPU 
has less heat than two 84w CPUs if you want to check real SMP operation. 
I'll build the dual Opteron sometime in the next year, but I haven't 
even ordered parts because I'm waiting for the O/S to be stable.
> 
> Bill Davidsen wrote:

	[... snip ...]
>> Obviously amost any decent Linux will run in 32bit mode, that's not 
>> interesting. I was planning to wait for the new Intel x86-64 CPU, 
>> because it has hyperthreading, which in theory will allow faster 
>> interrupt service.
>>
> 


-- 
    -bill davidsen (davidsen@tmr.com)
"The secret to procrastination is to put things off until the
  last possible moment - but no longer"  -me