Skip to main content
14 events
when toggle format what by license comment
Jul 18, 2019 at 5:56 vote accept Thangachelli Debopritama
Jul 16, 2019 at 9:33 answer added knl timeline score: 3
Jul 16, 2019 at 4:34 comment added user7440 -- A direct linear solver (like MUMPS) has no impact for the level of errors you are considering (i.e. ~ $10^{-3}$) -- When the problem is coercive (i.e. $\sqrt{ a(v,v) }$ is equivalent to a norm), using a series of nested meshes will guarantee that the error decreases with $h$... but not that you get a straight line.
Jul 16, 2019 at 4:26 comment added user7440 * You are correct that only the upper bound is a straight line.
Jul 14, 2019 at 3:40 comment added Thangachelli Debopritama @nicoguaro I will certainly try this, but is there a reason behind why this may cause such a behaviour?
Jul 14, 2019 at 1:25 comment added nicoguaro Have you tried refining the mesh instead, splitting by 4 every element every time, assuming it's a 2D problem.
Jul 14, 2019 at 1:13 comment added Thangachelli Debopritama @nicoguaro Yes, this is true. I am creating a new mesh every time.
Jul 13, 2019 at 20:39 comment added nicoguaro Judging from the plots, it does not seem like you are refining the mesh but creating a new one with different size every time, is that correct?
Jul 13, 2019 at 16:20 history edited Thangachelli Debopritama CC BY-SA 4.0
added 28 characters in body
Jul 13, 2019 at 10:44 history edited Thangachelli Debopritama CC BY-SA 4.0
added 355 characters in body
Jul 13, 2019 at 3:00 history tweeted twitter.com/StackSciComp/status/1149876303920848897
Jul 13, 2019 at 1:34 comment added Paul If you implement a method that converges with polynomial order N in the mesh size (and implement it correctly), that is what you should expect.
Jul 12, 2019 at 22:13 history edited nicoguaro CC BY-SA 4.0
added 4 characters in body; edited tags
Jul 12, 2019 at 21:12 history asked Thangachelli Debopritama CC BY-SA 4.0