[CIG-SHORT] No output

Tue Mar 13 18:14:21 PDT 2012

Birendra,

As I said in my last email, I think you will zero in on what is causing 
the displacements if you compare the computed stresses against your 
initial stresses as well as the solution. If you get upward 
displacements, then you have overcompensated for gravity in setting your 
initial stresses. If you get downward displacements, then your initial 
stresses are too small.

Brad

On 03/13/2012 06:08 PM, Birendra jha wrote:
> Hi Brad,
>
> You are right. Actually, at the command line I was just typing
> "pylith test". I don't know when my mind got convinced that I don't
> need to specify file extension :). I didn't get any error, but of
> course, didn't get any output either, except output_info.vtk,
> output_t0000000.vtk. Sorry about that. It is interesting how this
> "bug" was reproduced by Charles through a typo.
>
> I am still trying to figure out why the displacements are not close
> to 0. I used the same density profile for material density in
> den_mono.spatialdb and for the initial stress in
> initial_stress_mono.spatialdb
>
> Is it related to the mismatch in resolution issue? I will try with a
> coarser density/stress profile.
>
> Thanks Birendra
>
> --- On Wed, 3/14/12, Brad Aagaard<baagaard at usgs.gov>  wrote:
>
>> From: Brad Aagaard<baagaard at usgs.gov> Subject: Re: [CIG-SHORT] No
>> output To: "Charles Williams"<C.Williams at gns.cri.nz> Cc: "Birendra
>> jha"<bjha7333 at yahoo.com>, cig-short at geodynamics.org Date:
>> Wednesday, March 14, 2012, 5:30 AM Birendra,
>>
>> Running your files worked with pylith-stable built from source as
>> well as the v1.6.3 x86_64 and i686 binaries on my linux machine. I
>> suspect that you had a typo in specifying test.cfg on the command
>> line like Charles did.
>>
>> Regards, Brad
>>
>>
>> On 03/13/2012 04:43 PM, Charles Williams wrote:
>>> Hi Birendra,
>>>
>>> Initially, I thought I had the same problem you did
>> when running on my Linux machine.  I think discovered that I had
>> misspelled the name of the .cfg file (I typed test.cft rather than
>> test.cfg).  Once I did it correctly, everything seemed to work
>> correctly on the Linux machine as well.  I'm not sure if this could
>> be your problem or not.
>>>
>>> Cheers, Charles
>>>
>>>
>>>
>>> On 14/03/2012, at 12:07 PM, Brad Aagaard wrote:
>>>
>>>> Birendra,
>>>>
>>>> I get vtk files in my output directory with the
>> solution and state
>>>> variables. The displacements are nonzero (~200m of
>> vertical motion). You
>>>> can check your solution and initial stress state
>> using the solution. If
>>>> the stresses match, then the displacements should
>> be zero.
>>>>
>>>> Regards, Brad
>>>>
>>>>
>>>> On 03/13/2012 04:01 PM, Birendra jha wrote:
>>>>> Dear Brad, Charles,
>>>>>
>>>>> I clearly forgot to include the output
>> directory in the zip file. But I have that on my computer so that's
>> not the issue.
>>>>> I will try to reduce the length of the
>> spatialdb files. Yes, it doesn't make sense to have 1 m resolution
>> in spatialdb files when the grid cels are 100s of m. I may not be
>> able to change to a tetrahedral grid but I'll try.
>>>>>
>>>>> Granted all that, I am sure it runs in few
>> minutes. Are you able to see the vtk files in output folder? Are
>> the displacements zero?
>>>>>
>>>>> I have run the same problem with constant
>> density (just 1 line in density spatialdb and 2 lines in initial
>> stress spatialdb) and get the desired result of zero displacement.
>> Now I just want to see it with depth varying density. As you can
>> probably imagine, I have density file for each of the 33 regions. I
>> have their respective initial stress files too.
>>>>>
>>>>> regards Birendra
>>>>>
>>>>> --- On Wed, 3/14/12, Brad Aagaard<baagaard at usgs.gov>
>>>>> wrote:
>>>>>
>>>>>> From: Brad Aagaard<baagaard at usgs.gov> Subject: Re: No
>>>>>> output To: "Birendra jha"<bjha7333 at yahoo.com> Cc:
>>>>>> cig-short at geodynamics.org Date: Wednesday, March 14, 2012,
>>>>>> 4:17 AM Birenda,
>>>>>>
>>>>>> I was able to run your test.cfg file after
>> creating the
>>>>>> output directory (as Charles noted), but
>> initializing the
>>>>>> integrators is taking a rather long time.
>> Your spatial
>>>>>> databases that describe the variation in
>> physical properties
>>>>>> are quite large. For every quadrature point
>> in every cell
>>>>>> PyLith has to find which points to use for
>> the
>>>>>> interpolation. Using a cloud of points with
>> a much finer
>>>>>> spacing than the size of the cell will
>> cause PyLith to run
>>>>>> much slower than necessary.
>>>>>>
>>>>>> The solution did converge, but when I
>> looked at your mesh I
>>>>>> see some cells that have very poor aspect
>> ratios so the
>>>>>> solution may not be very accurate. See the
>> PyLith Wiki
>> (http://www.geodynamics.org/cig/community/workinggroups/short/workarea/pylith-wiki/index_html)
>>>>>>
>>
for hints on checking the aspect ratio.
>> Using tetehedral
>>>>>> cells will give you much greater
>> flexibility in varying the
>>>>>> discretization size while keeping the
>> aspect ratio of cells
>>>>>> close to 1. See examples/2d/subduction for
>> an example of
>>>>>> using the bias scheme.
>>>>>>
>>>>>> Regards, Brad
>>>>>>
>>>>>>
>>>>>>
>>>>>> On 03/13/2012 03:08 PM, Birendra jha
>> wrote:
>>>>>>> Dear developers,
>>>>>>>
>>>>>>> I am trying to run a simple case like
>> Step16 in the
>>>>>> tutorial except
>>>>>>> that density now varies with depth. I
>> set the initial
>>>>>> geostatic
>>>>>>> stress from this density profile. I
>> expect to see a
>>>>>> zero
>>>>>>> displacement. I have attached the files
>> as zipped.
>>>>>>>
>>>>>>> Why I am not getting any output (It
>> must be some small
>>>>>> mistake on my
>>>>>>> part. Just can't figure it out)?
>>>>>>>
>>>>>>> Thanks and regards Birendra
>>>>>>>
>>>>>>
>>>>>>
>>>>>
>>>>
>>>> _______________________________________________ CIG-SHORT
>>>> mailing list CIG-SHORT at geodynamics.org
>>>> http://geodynamics.org/cgi-bin/mailman/listinfo/cig-short
>>>
>>> Charles A. Williams Scientist GNS Science 1 Fairway Drive,
>>> Avalon PO Box 30368 Lower Hutt  5040 New Zealand ph (office):
>>> 0064-4570-4566 fax (office): 0064-4570-4600
>>> C.Williams at gns.cri.nz
>>>
>>>
>>> Notice: This email and any attachments are
>> confidential. If received in error please destroy and immediately
>> notify us. Do not copy or disclose the contents.
>>>
>>>
>>
>>
>