Polyhedron b'marks | Linux vs Windows

[Jerry DeLisle]

On 06/03/2014 07:36 PM, Steve Kargl wrote:
> On Tue, Jun 03, 2014 at 06:43:28PM -0700, Jerry DeLisle wrote:
>> On 06/01/2014 03:57 PM, Evan Cooch wrote:
>>> So, have a box running a new-ish, fairly high-end Intel multi-core chip
>>> (which one doesn't really matter here). Decided to see if the OS
>> ---- snip ----
>>
>> I do not remember all the details, but I did find that at least one of the
>> common libraries was implementing some of the basic trig functions, such as
>> sin(), with software only methods and was not using the floating point
>> co-processor capability in most modern CPUs.
>>
>> I attempted to get a maintainer to update this code and use the available
>> hardware.  I was met with a response more or less "Does it make a real
>> difference in a real application?"  That is a legitimate question and I don't
>> have an answer.
>>
> 
> If your modern hardware is using an Intel 387 coprocessor (or
> equivalent), you should run away as fast as possible from using
> fsin and friends.
> 
> OP needs to step back and benchmark the IO of the OSs to
> the underlying filesystems.  It would then be prudent to 
> learn how to use a profiler (i.e., gprof).
> 

Ha, I am very aware of the potential/real problems with some of the hardware
implementations. I suppose this is why some folks have chosen to be correct vs
fast. I was only trying to suggest that the speed differences being observed may
be related to these library differences.

And, as Tim mentioned, we don't know exactly which build of gfortran is being used.

gfortran -v might shed some light.

Jerry