[yt-users] covering grids, memory

[David Collins]

> It is, but it's a bit DIY and will require incompatible changes that
> then require recompilation.  The covering grid machinery is written in

Hmmm, sounds invasive and something a not-so-sharp cookie like myself
will screw up at least once.

I've been caching my blocks to disk at single, so I'll try just work
off the disk for now.

Thanks!
d.

> C, usign the Python/C API.  So you have to modify the initialization
> of the empty (zero) grids in Python code in
> yt/data_objects/data_containers.py, and then again in
> yt/utilites/data_point_utilities.c in the FillRegion section.
>
> It might be easier to do it piecewise over the domain and then
> down-convert manually.
>
> -Matt
>
> On Tue, Apr 5, 2011 at 4:58 PM, david collins <antpuncher at gmail.com> wrote:
>> Hi--
>>
>> I'm trying to work with come covering grids, extracted at the root
>> grid resolution of a 512^3 sim.  I'm on a somewhat memory bound
>> machine (6 gig total).
>>
>> My data is in double precision, but I'd like to be working in single,
>> as I'm getting memory bound by the things I want to be doing.  Is it
>> possible to get the covering grid machinery to force the work to be
>> done in single precision?
>>
>> Thanks,
>> d.
>>
>>
>> --
>> Sent from my computer.
>> _______________________________________________
>> yt-users mailing list
>> yt-users at lists.spacepope.org
>> http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org
>>
> _______________________________________________
> yt-users mailing list
> yt-users at lists.spacepope.org
> http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org
>



-- 
Sent from my Stone Tablet and carried by my Pterodactyl.