In message <f0d25a7f52.old_coaster@old_coaster.yahoo.co.uk>
Tony Moore <[email protected]> wrote:
> On 11 Apr 2012, george greenfield <[email protected]>
> wrote:
>> In message <0326ac7e52.old_coaster@old_coaster.yahoo.co.uk>
>> Tony Moore <[email protected]> wrote:
>
> [snip]
>
>> > Interesting. For you, RPCEmu runs about twice as fast as for me,
>> > which reflects the relative speeds of the Win7 processors.
>>
>> But we're both trumped by Chris Gransden's results for the BB-XM and
>> Pandaboard ...:-)
>
> ... which have processor speeds of only 1.0GHz and 1.5GHz. I wonder why
> RPCEmu should run so fast on slower hardware.
>
If 'slower hardware' refers to the BB-XM and Pandaboard, perhaps it's
because they are running RISC OS natively, and not under emulation.
Also, the latest ARM chips can issue more than one instruction per
cycle, I believe (which wasn't the case with chips up to and including
that in the Iyonix), meaning more work gets done per clock cycle.
(But take the above with a pinch of salt: I'm no expert).
George
--
george greenfield
_______________________________________________
Rpcemu mailing list
[email protected]
http://www.riscos.info/cgi-bin/mailman/listinfo/rpcemu
