On Jun 1, 2011, at 11:17 AM, Roy Stogner wrote:
> I get O(h^4)/O(h^3) L2/H1 convergence in ex14 after changing that to
> HERMITE/3...
Indeed... I just heard back from my post-doc and he had something wrong
with his MMS runs he fixed it and now he's getting the correct order ;-)
"Never a
On Jun 1, 2011, at 11:47 AM, Roy Stogner wrote:
>
> On Wed, 1 Jun 2011, Cody Permann wrote:
>
>> On Jun 1, 2011, at 11:34 AM, Roy Stogner wrote:
>>
>>> Struct padding in elem.h is such that adding 1 byte to
>>> sizeof(subdomain_id_type) won't necessarily add 1 byte to
>>> sizeof(Elem). At fir
On Wed, 1 Jun 2011, Cody Permann wrote:
> On Jun 1, 2011, at 11:34 AM, Roy Stogner wrote:
>
>> Struct padding in elem.h is such that adding 1 byte to
>> sizeof(subdomain_id_type) won't necessarily add 1 byte to
>> sizeof(Elem). At first glance it looks like it would add 4 bytes on
>> 32-bit syst
On Jun 1, 2011, at 11:34 AM, Roy Stogner wrote:
>
> On Wed, 1 Jun 2011, Cody Permann wrote:
>
>> We (the INL) have the need to grow the number of subdomains in our
>> simulations past the current storage capacity of subdomain_id_type.
>> This type is currently set as an unsigned char allowing o
> Should we make a new configure option to allow us to change the default size?
> Should we just make the change to the library? Are there other
> issues/concerns with changing this type?
I don't have a problem increasing this to 16 or even 32 bits. What is the
output of sizeof(DofObject) with t
On Wed, 1 Jun 2011, Cody Permann wrote:
> We (the INL) have the need to grow the number of subdomains in our
> simulations past the current storage capacity of subdomain_id_type.
> This type is currently set as an unsigned char allowing only 256
> subdomains. We are already starting to run and d
On Wed, 1 Jun 2011, Derek Gaston wrote:
> Not sure on that point but we're not getting the convergence
> rate we should for a simple scalar poisson problem so there is
> something going on there.
I get O(h^4)/O(h^3) L2/H1 convergence in ex14 after changing that to
HERMITE/3...
although
Developers,
We (the INL) have the need to grow the number of subdomains in our simulations
past the current storage capacity of subdomain_id_type. This type is currently
set as an unsigned char allowing only 256 subdomains. We are already starting
to run and debug libMesh using a locally patc
On Jun 1, 2011, at 9:12 AM, Dmitry Karpeev wrote:
> What is the asymptotic accuracy mixed discretization of a fourth-order
> PDE with a second order Lagrange basis?
> Not to say that there isn't a problem with Hermite, but it seems to me
> that this split lagrange space is richer than
> Hermite, s
What is the asymptotic accuracy mixed discretization of a fourth-order
PDE with a second order Lagrange basis?
Not to say that there isn't a problem with Hermite, but it seems to me
that this split lagrange space is richer than
Hermite, so it should have at least the same accuracy (or better).
Dmi
On Jun 1, 2011, at 8:28 AM, Jed Brown wrote:
> On Wed, Jun 1, 2011 at 15:38, Derek Gaston wrote:
> Has anyone done any MMS verification on cubic hermite convergence rates? I
> thought they were supposed to be O(h^4) but we're only seeing O(h^2) on a
> fairly simple 2D poisson problem with
On Wed, Jun 1, 2011 at 15:38, Derek Gaston wrote:
> Has anyone done any MMS verification on cubic hermite convergence rates? I
> thought they were supposed to be O(h^4) but we're only seeing O(h^2) on
> a fairly simple 2D poisson problem with an assumed solution of
> u=sin(8*pi*x).
Are you
On Wed, 1 Jun 2011, Derek Gaston wrote:
> Has anyone done any MMS verification on cubic hermite convergence rates?
Yes, but not in several years; it's entirely possible there's a
regression.
I'll give it a try to double-check this afternoon.
Of course, the long-term solution remains the same:
Has anyone done any MMS verification on cubic hermite convergence rates? I
thought they were supposed to be O(h^4) but we're only seeing O(h^2) on a
fairly simple 2D poisson problem with an assumed solution of u=sin(8*pi*x).
We started investigating this because we were doing some compariso
14 matches
Mail list logo
