A question about GMPs performance on the Intel Core i3 & i5 processors

[Torbjorn Granlund]

Will Galway <galway at math.uiuc.edu> writes:

  I'm considering getting a new computer using, most likely, either an
  Intel i3-530 processor (my preference) or an Intel i5-570 processor.
  In terms of the performance of GMP, is there any particular reason I
  should prefer the latter processor?  (The i5-570 has 8MB of level-3
  cache, vs the i3-530's 4MB, but I rarely use GMP for "large" operands,
  rather I tend to work mostly with 128..256-bit operands.)
  
  Thanks for any advice that readers might have.
  
The GMPbench reslts page at

   http://gmplib.org/gmpbench.html
  
might be useful reading.

A more technical page, https://gmplib.org/devel/asm.html might also
provide some insights.

AMD processors are the far fastest for GMP work currently, at least
until operands get really huge.  Intel's newest processors have a better
memory subsystem (better prefetch, more bandwidth) so they catch AMD's
procssors when we perform the largest FFTs.

The main reason behind AMD's performance is the 64 x 64 -> 128 bit
multiply accumulation performance.  AMD can approach one such operation
each 2.25th cycle; Intel core iX can approach one such operation each
3.75th cycle.

I am not aware of any differences between core i3, i5, i7 wrt GMP
performance.

-- 
Torbjörn