On Fri, Dec 11, 2009 at 3:13 AM, Oliver Kullmann
<O.Kullmann at swansea.ac.uk>wrote:
> Hi,
>
> I've built it successfully from sources, using Ecl version 9.10.2.
> On my 64-bit machine the single regression test failure was the usual
>
> Running tests in rtest14:
> ********************** Problem 144 ***************
> Input:
> closeto(AS_10_4_17(- 10), 1.e-15)
>
>
> Result:
> 3/2 3/2
> 2 2 10 2 2 10
> 10 (bessel_j(- -, -------) - bessel_j(-, -------))
> 3 3 3 3
> -------------------------------------------------- + airy_dai(- 10)
> 3
>
> This differed from the expected result:
> true
>
closeto(AS_10_4_17(- 10), 1.e-15) is a numerical test, but the failure
result is symbolic. I am overseas on vacation for a few weeks, so I can't
follow this up at present.
David