FierceKatana

I think I have a pretty good understanding of x64 technology so here goes.

I understand that x64@3.0ghz is NOT equivalent to x86@6.0ghz for the simple reason that x86 programming cant just magically change to do more calculations in one cycle...

BUT, could a line be drawn down the center of a 3.0ghz x64 cpu to effectively create two x86 'lanes' (cores) of an equal 3.0ghz clock speed?

Is it a motherboard support issue or could it possibly be done with some vmware-esque software?

Motivation: I have an ancient 2005 1.8ghz Turion 64 laptop that id like to milk for a few more years and im hardly using its 64 bit capabilities.

Thanks

shuuhen

Well, they definitely aren't equivalent. Clock speed is less of an issue. 64-bit has more to do with bus sizes, address sizes, register sizes, etc. Each processor model will function differently from a different internal design. This could be from differences in the Arithmetic Logic Unit (ALU) or other circuitry. Older ALU's took a much longer time on operations (multiplication and division being the popular examples). I think multiplication could take something like 30 clock cycles on some processors, while now they should be somewhere around 3 clock cycles.


Not quite. Multicore processors allow multiple programs to execute at the same time, but each core can only execute instructions from one process at a time, regardless of if it is 8, 16, 32 or 64 bit.


If Windows isn't performing well enough, you could look into running Linux on the laptop. You can run a 64-bit distribution, but there's a good chance you won't notice much of a performance difference between 32 and 64 bit operating systems (this can depend on what you use the machine for). Most of the performance increase you'd get would be from running Linux (especially a lightweight one). You can still get plenty of use out of that laptop.

I'm probably skipping over some details, but I think I hit all the major ones.