dolfin team mailing list archive

[dolfin@xxxxxxxxxx]

Commit from johanjan (2006-01-09 02:44 CET)
--------------------

ODE:

Progress on conservative modification for multi-adaptivity:
implemented simple condition which seems to work well generally.

Added results for "domain" version of reaction benchmark. The
multi-adaptive error is now below the mono-adaptive error. The
multi-adaptive gain is very large (a factor 10+, looks like 20 in the
limit), so these results are good.

  dolfin  src/kernel/ode/MultiAdaptiveTimeSlab.cpp     1.54
  dolfin  src/kernel/ode/MultiAdaptivity.cpp           1.25
  dolfin  src/kernel/ode/dolfin/MultiAdaptivity.h      1.18
  dolfin  src/demo/solvers/ode/reaction/bench.log      1.17
+ dolfin  src/demo/solvers/ode/reaction/domain.log     1.1
  dolfin  src/demo/solvers/ode/reaction/run-bench.py   1.6
  dolfin  src/demo/solvers/ode/reaction/run-domain.py  1.2