# How the boundary condition is calculated

**URL:** <https://community.freefem.org/t/how-the-boundary-condition-is-calculated/1113>\
**Category:** General Discussion\
**Created:** [July 22, 2021, 9:36am UTC](https://community.freefem.org/t/how-the-boundary-condition-is-calculated/1113 "2021-07-22T09:36:14Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![yongxing](https://yyz2.discourse-cdn.com/flex030/user_avatar/community.freefem.org/yongxing/32/3632_2.png) [@yongxing](https://community.freefem.org/u/yongxing)\
**Post date:** [July 22, 2021, 9:36am UTC](https://community.freefem.org/t/how-the-boundary-condition-is-calculated/1113/1 "2021-07-22T09:36:14Z")

</div>

Dear all,  
I am wondering how FreeFem imposes the Dirichlet boundary condition when using **tgv= -1**?  
I have tried both the direct solver LU and iterative solver hypre, **why it’s not exactly zero at the wall boundary** as shown below:

 ![image](https://canada1.discourse-cdn.com/flex030/uploads/freefem/original/1X/77385bd66a386965876e3861be440be615250b6e.jpeg)

or this might be a visualisation issue in paraview?

Best,  
Yongxing.