On 1/28/07, Keith Goodman <[EMAIL PROTECTED]> wrote:
On 1/28/07, Keith Goodman <[EMAIL PROTECTED]> wrote:
> On 1/28/07, Fernando Perez <[EMAIL PROTECTED]> wrote:
> > On 1/28/07, Keith Goodman <[EMAIL PROTECTED]> wrote:
> > > On 1/28/07, Fernando Perez <[EMAIL PROTECTED]> wrote:
> > > > [snip] T
On 1/28/07, Keith Goodman <[EMAIL PROTECTED]> wrote:
> On 1/28/07, Fernando Perez <[EMAIL PROTECTED]> wrote:
> > On 1/28/07, Keith Goodman <[EMAIL PROTECTED]> wrote:
> > > On 1/28/07, Fernando Perez <[EMAIL PROTECTED]> wrote:
> > > > [snip] The test code uses double arrays, and SSE2 has double
> >
On 1/28/07, Fernando Perez <[EMAIL PROTECTED]> wrote:
> On 1/28/07, Keith Goodman <[EMAIL PROTECTED]> wrote:
> > On 1/28/07, Fernando Perez <[EMAIL PROTECTED]> wrote:
> > > [snip] The test code uses double arrays, and SSE2 has double
> > > precision support but it's purely 64-bit doubles. SSE is
On 1/28/07, Keith Goodman <[EMAIL PROTECTED]> wrote:
> On 1/28/07, Fernando Perez <[EMAIL PROTECTED]> wrote:
> > [snip] The test code uses double arrays, and SSE2 has double
> > precision support but it's purely 64-bit doubles. SSE is
> > single-precision only, which means that for a double compu
On 1/28/07, Charles R Harris <[EMAIL PROTECTED]> wrote:
> But how come it isn't consistent and seems to depend on timing? That is what
> makes me think there is a race somewhere in doing something, like setting
> flags . I googled yesterday for floating point errors and didn't find
> anything that
On 1/28/07, Charles R Harris <[EMAIL PROTECTED]> wrote:
> It is strange, isn't it. I'm still thinking race condition, but where? I
> suppose even python could be involved someplace.
>
> BTW, your algorithm sounds like a great way to expose small descrepancies.
> There's a great test for floating po
On 1/28/07, Fernando Perez <[EMAIL PROTECTED]> wrote:
> [snip] The test code uses double arrays, and SSE2 has double
> precision support but it's purely 64-bit doubles. SSE is
> single-precision only, which means that for a double computation,
> ATLAS isn't used and the Intel FPU does the computa
On 1/28/07, Keith Goodman <[EMAIL PROTECTED]> wrote:
On 1/28/07, Fernando Perez <[EMAIL PROTECTED]> wrote:
> On 1/28/07, Charles R Harris <[EMAIL PROTECTED]> wrote:
>
> > > The problem goes away if I remove atlas (atlas3-sse2 for me). But
that
> > > just introduces another problem: slowness.
> >
On 1/28/07, Fernando Perez <[EMAIL PROTECTED]> wrote:
On 1/28/07, Charles R Harris <[EMAIL PROTECTED]> wrote:
> > The problem goes away if I remove atlas (atlas3-sse2 for me). But that
> > just introduces another problem: slowness.
> >
> > So is this something to report to Clint Whaley? Or does
On 1/28/07, Fernando Perez <[EMAIL PROTECTED]> wrote:
> On 1/28/07, Charles R Harris <[EMAIL PROTECTED]> wrote:
>
> > > The problem goes away if I remove atlas (atlas3-sse2 for me). But that
> > > just introduces another problem: slowness.
> > >
> > > So is this something to report to Clint Whaley?
On 1/28/07, Charles R Harris <[EMAIL PROTECTED]> wrote:
> How come your computation is so sensitive to these small effects?
I don't know.
The differences I am seeing are larger than in the test script---but
still of the order of eps. Each time step of my simulation selects a
maximum value and the
On 1/28/07, Charles R Harris <[EMAIL PROTECTED]> wrote:
> > The problem goes away if I remove atlas (atlas3-sse2 for me). But that
> > just introduces another problem: slowness.
> >
> > So is this something to report to Clint Whaley? Or does it have to do
> > with how numpy uses atlas?
>
>
> Inter
On 1/28/07, Keith Goodman <[EMAIL PROTECTED]> wrote:
On 1/27/07, Keith Goodman <[EMAIL PROTECTED]> wrote:
> On 1/27/07, Fernando Perez <[EMAIL PROTECTED]> wrote:
> > It's definitely looking like something SMP related: on my laptop, with
> > everything other than the hardware being identical (Lin
On 1/28/07, Keith Goodman <[EMAIL PROTECTED]> wrote:
> The problem goes away if I remove atlas (atlas3-sse2 for me). But that
> just introduces another problem: slowness.
>
> So is this something to report to Clint Whaley? Or does it have to do
> with how numpy uses atlas?
I was wondering if atla
On 1/27/07, Keith Goodman <[EMAIL PROTECTED]> wrote:
> On 1/27/07, Fernando Perez <[EMAIL PROTECTED]> wrote:
> > It's definitely looking like something SMP related: on my laptop, with
> > everything other than the hardware being identical (Linux distro,
> > kernel, numpy build, etc), I can't make i
I believe that I didn't communicate my problem completely. Your suggestion
to flatten the matrix should help develop a solution. Now as I progress
down the sorted list I need to detect when I have 25 complete solution sets.
My goal is to find a set of 25 where the highest collective covariance
(
16 matches
Mail list logo
