[SciPy-user] 64 bit and 32 bit on Ubuntu

[Kurt Smith]

On Mon, Oct 13, 2008 at 10:25 AM, Anne Archibald
<peridot.faceted at gmail.com> wrote:
> 2008/10/12 Kurt Smith <kwmsmith at gmail.com>:
>
>> I've recently acquired a new Ubuntu system, and unfortunately some 3rd
>> party software I'd like to use won't work under the 64 bit version of
>> the OS, yet.  It's not a deal breaker, however, and I'm willing to
>> work around it as long as there are compelling reasons to use 64 bit
>> for numerical work, specifically under scipy, etc.  I'm not well
>> versed in the low-level pros and cons of 64 bit vs 32 bit, esp. as to
>> how it affects scipy performance.  Can you enlighten me?  I"m on an
>> Intel Core 2 Duo, 3.16 GHz, E8500, 4 GB RAM.  Will using 64 bit yield
>> great benefits running scipy/numpy code automatically, or will it
>> require fine-tuning on my part?  Will 64 bit yield benefits besides
>> having a larger address space (besides the fact that my 4 GB RAM is
>> still addressable by 32 bits, anyway)?
>
> I think memory is the only major reason to prefer 64 bits. The speed
> difference will probably be minimal. But having a 64-bit address space
> means that if your applications get too big for RAM they just slow
> down, as opposed to crashing and forcing you to rewrite them. You will
> also find, I think, that running in 32-bit mode you cannot access the
> last gigabyte or so of physical RAM. This can make a big difference
> too.

Thanks to all for your answers; since posting I've worked out some
more issues and 64 bit looks better and better.  Indeed, it is at
least 2x "cooler."  ;-)  I'll do more benchmarks of my own, but so far
things look pretty good.

Kurt