Hanging too long during pipeline running

[Saurabh]

Hii group members, has any of you ever come across a situation where the program takes too long to run the pipeline, in particular "Insert Precipitate Phases". Please check the issue in the inserted image below and a workaround for this will be highly appreciated..

[Michael Jackson]

This can happen if the input statistics approach certain edge cases. The only advice I can offer is to force quit DREAM.3D and then try again with slight different statistics. Usually the percent of precipitates is too large and the algorithm cannot match the requested statistics so one of the inner loops never exits.

--
Mike Jackson

On 4/20/20, 9:10 AM, "Saurabh" <dream3...@googlegroups.com on behalf of saurab...@gmail.com> wrote:

Hii group members, has any of you ever come across a situation where the program takes too long to run the pipeline, in particular "Insert Precipitate Phases". Please check the issue in the inserted image below and a workaround for this will be highly appreciated..

<about:invalid#zClosurez>






--
You received this message because you are subscribed to the Google Groups "dream3d-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to dream3d-user...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/dream3d-users/78158bc7-73f0-484d-b6b3-8bd2bbf0172b%40googlegroups.com <https://groups.google.com/d/msgid/dream3d-users/78158bc7-73f0-484d-b6b3-8bd2bbf0172b%40googlegroups.com?utm_medium=email&utm_source=footer>.

[Saurabh]

Thanks Mike! well pointed out. Great help.

On Monday, April 20, 2020 at 7:09:10 PM UTC+5:30, Michael Jackson wrote:

This can happen if the input statistics approach certain edge cases. The only advice I can offer is to force quit DREAM.3D and then try again with slight different statistics. Usually the percent of precipitates is too large and the algorithm cannot match the requested statistics so one of the inner loops never exits.

--
Mike Jackson

On 4/20/20, 9:10 AM, "Saurabh" <dream3...@googlegroups.com on behalf of saurab...@gmail.com> wrote:

    Hii group members, has any of you ever come across a situation where the program takes too long to run the pipeline, in particular "Insert Precipitate Phases". Please check the issue in the inserted image below and a workaround for this will be highly appreciated..

     <about:invalid#zClosurez>






    --
    You received this message because you are subscribed to the Google Groups "dream3d-users" group.

    To unsubscribe from this group and stop receiving emails from it, send an email to dream3...@googlegroups.com.

[Alex G]

Could  you elaborate on what the edge cases are? I'm having the same issue when trying to generate two phase structure. The pipeline almost always hangs on inserting precipitates when the size of the precipitates are approx 10 times smaller than average size of primary phase. But I have not found exact problem with statistics to avoid this.

понедельник, 20 апреля 2020 г. в 18:39:10 UTC+5, mike.j...@bluequartz.net:

[Michael Jackson]

The algorithm I think has an implicit assumption that the volume percent of precipitates should not exceed about 5~10%. There are also issues if you try to cram too many precipitates at the grain boundaries and the system cannot make that happen. These areas will produce infinite loops inside the algorithm.

 

--

Mike Jackson

To view this discussion on the web visit https://groups.google.com/d/msgid/dream3d-users/18d109d1-0499-4a09-b4f9-3d57eb3d60d5n%40googlegroups.com.