Re: a big limit of mathematica?

On 13 Sep 2005 11:45:35 -0700, "Daniel Lichtblau" <danl@xxxxxxxxxxx>
wrote:


>>
>> Just for this motive I have not marveled when the calculation is
>> arrested on Mathematica 5.1 + XP(32Bit)
>> I have marveled instead when the calculation is arrested on
>> Mathematica 5.2 + WinXP(64Bit) 2GB Ram 400GB HD
>
>As I noted, we are able to do the computation on 64 bit platforms. One
>thing to note: I am told that if you are using a beta version of 64-bit
>WinXP then your Mathematica installation will most likely be for the
>32-bit version.
>

Mathematica 5.2 from you distributed in version try has limitations?
How do I do to know if the installation of Mathematica 5.2 on Win XP
64 are really working to 64 bit?


>>
>>
>> Personally I have experimented in positive sense using Gnu gcc +
>> UbuntuLinux 5.04 (64Bit) + ACML (amd) (Blas+Lapack) 2GB RAM+ 400GB HD
>> +Athlon64
>
>We tested one of the Ma... languages you had mentioned. We were unable
>to verify your claim that it handled a 10^4 x 10^4 matrix determinant
>in any way other than to run out of memory.
>
>

>
>(1) We were unable to create a random matrix of dimension 11000 x 11000
>in the language we tried. Let alone 10^5 x 10^5.
>
>(2) Why do you specifically fault Mathematica on memory capacity for
>failing to find the determinant, if Matlab and Maple cannot find it
>either? You rather strongly imply that the memory restriction you
>encounter is specific to Mathematica. As best we can verify, this is
>not correct. At least one of those others, on our installation, got
>swamped even earlier (below 8000 x 8000).
>
I have never said this, I have never said to have resolved the
calculation of the determinant with Matlab and I have specified this
more times. I have said instead that in a lot of occasions in which I
had interruptions of numerical calculation in Mathematica I succeeded
in Matlab or C in resolving the problem.
However the problem is not this, I have observed more times that I
love Mathematica and really for this I wanted to look for remedies to
be able to use it to fund
>
>It is not terribly difficult to make memory estimates involving packed
>arrays in Mathematica. The only information not documented is total
>array size of 2^30 elements. You certainly know that, when packed, they
>occupy 8 bytes each for machine doubles. You can assume that a
>workspace copy will be required for matrix manipulations that require
>level 3 BLAS. There might be other considerations related to your
>machine, installation, availability of contiguous memory, etc. But most
>such issues will not be specific to Mathematica.
>
>
Thanks so many as soon as the time to my disposition will allow me to
deepen these matters and to apply these jewels recommends I will
certainly do it.
So many graces for the jewel help

Roberto

.

Relevant Pages

  • Re: a big limit of mathematica?
    ... One is that most 32 bit Mathematica platforms have a 2 ... Just for this motive I have not marveled when the calculation is ... > The determinant computation will require at ... >upshot is it is not hard to see memory being an issue. ...
    (sci.math.symbolic)
  • Re: Mathematica. A big limit of mathematica !!!???
    ... > Mathematica kernel has shut down. ... m will require at least several Gigabyte memory. ... Probably not since Mathematica already ... > scandalized me if in such problems the time of calculation lengthens ...
    (sci.math.num-analysis)
  • Re: a big limit of mathematica?
    ... This behavior is predictable knowing as mathematica kernel is ... This problem on my hardware is interrupted for lack of memory (as ... With Matlab or Maple on Same Hardware it is resolved (Ok!! ... Matlab performs this calculation because, ...
    (sci.math.symbolic)
  • Re: a big limit of mathematica?
    ... >as Paul Abbott and me, ... Mathematica could have a big defect of planning respect the management ... if to the goals of the calculation nothing changes having set as ... precision but it is slow in numerical calculations. ...
    (sci.math.symbolic)
  • Re: a big limit of mathematica?
    ... > Mathematica kernel has shut down. ... > problems that really make unusable Mathematica for my purposes. ... > scandalized me if in such problems the time of calculation lengthens ... did) and it is often easy to see simple relations (like Peter and Paul ...
    (sci.math.symbolic)
Loading