Thanks for letting us know, that's a relief. (My 2.4 Gb test file was
just the same 64 Mb file repeated 72 times, so it was hard to be sure it
was actually working properly.)
Wayne
On Mon, 26 Jan 2009, Andrew Fowler wrote:
> Hi Wayne,
>
> I stand corrected on my statement that "everyone" should probably move to
> v.2...
>
> More importantly, I applied the patches to v.1. My "monster" 2.7 gig NOESY
> opened, and did it relatively quickly. I've yet to decide whether to use it
> or the two subspectra I created as a workaround last week, but the
> functionality is definitely there.
>
> Andrew
>
>
> On 1/25/09 3:35 PM, "Wayne Boucher" <[log in to unmask]> wrote:
>
> > Hello,
> >
> > I've just updated the v1 code to deal with the large data files. I've
> > also meanwhile slightly updated the v2 code so that it includes
> > <sys/types.h> rather than <unistd.h> (and the v1 code does the same).
> > That would probably have been a more sensible choice in the first place
> > for getting at the definition of the off_t data type and avoids the name
> > conflict with bacus.c (which has now been avoided in v2 in any case now).
> >
> > Let me know of any problems.
> >
> > Wayne
> >
> > On Fri, 23 Jan 2009, Wayne Boucher wrote:
> >
> >> Right, Brian thinks it's worth updating the v1 code so I will do that once
> >> we see if the dust settles on this change in the v2 code (that readlink
> >> error was a bit of a surprise!).
> >>
> >> And Brian will point out (because he did last week in Cambridge!) that
> >> there are two sorts of people who might want to stick with v1: those who
> >> are close to finishing a project (in particular Ph.D. students) and those
> >> people who are close to doing structure calculations with ARIA, because
> >> the current publicly available ARIA does not work with v2 (that will
> >> hopefully be sorted soon).
> >>
> >> Wayne
> >>
>
|