Thanks,
that's the initial point in integration. My values there are
-- dstep: 9.81743e-17 3.08687e-48 9.07047e-10 -9.07047e-10 -9.07047e-10
9.07047e-10 -3.62750e-13 -2.20616e-27 -6.93678e-59 -1.81409e-16
-1.81409e-16 1.81409e-16 1.81409e-16 1.57079e-14 4.93899e-46
The differences appear to be related to very small numbers. Can someone
make some sense out of this?
On 12/01/2011 08:13 PM, Andreas Schwab wrote:
Tuomo Keskitalo<[email protected]> writes:
What's your architecture/platform, compiler& version?
gcc 4.6 on G5.
msbdf should never increase or decrease order by more than one, unless it
is reset, so there's something really wrong there. Please post all lines
until previous "msbdf_apply:". Is this coming from test_extreme_problems?
-- msbdf_reset called
msbdf_apply: t=0.00000e+00, ord=1, h=1.00000e-10, y:0.00000e+00 0.00000e+00
0.00000e+00 0.00000e+00 0.00000e+00 0.00000e+00 0.00000e+00 0.00000e+00
0.00000e+00 0.00000e+00 0.00000e+00 0.00000e+00 0.00000e+00 0.00000e+00
0.00000e+00
-- ord=1, ni=0, ordwait=2
-- ordprev: 1 1 1 1 1
-- errlev: 1.00000e-07 1.00000e-07 1.00000e-07 1.00000e-07 1.00000e-07
1.00000e-07 1.00000e-07 1.00000e-07 1.00000e-07 1.00000e-07 1.00000e-07
1.00000e-07 1.00000e-07 1.00000e-07 1.00000e-07
-- check_no_order_decrease 1, check_step_size_decrease 0
-- calccoeffs ordm1coeff=0.00000e+00 ordp1coeff=2.00000e+00
ordp2coeff=6.00000e+00 errcoeff=5.00000e-01
-- predicted y: 0.00000e+00 0.00000e+00 0.00000e+00 0.00000e+00 0.00000e+00
0.00000e+00 0.00000e+00 0.00000e+00 0.00000e+00 0.00000e+00 0.00000e+00
0.00000e+00 0.00000e+00 0.00000e+00 0.00000e+00
-- evaluate jacobian
-- update M, gamma=1.00000e-10
-- dstep: 9.81743e-17 -8.68549e-36 9.07047e-10 -9.07047e-10 -9.07047e-10
9.07047e-10 -3.62750e-13 -2.20616e-27 5.60182e-46 -1.81409e-16 -1.81409e-16
1.81409e-16 1.81409e-16 1.57079e-14 -1.96797e-42
I too bet on a compiler issue.
Definitely not.
Andreas.
--
[email protected]
http://iki.fi/tuomo.keskitalo
_______________________________________________
Bug-gsl mailing list
[email protected]
https://lists.gnu.org/mailman/listinfo/bug-gsl
