[prev in list] [next in list] [prev in thread] [next in thread]
List: kde-edu
Subject: Re: Review Request 117816: Use marching cubes algorithm from Aqsis
From: Percy_Camilo_Triveño_Aucahuasi <percy.camilo.ta () gmail ! com>
Date: 2014-04-29 0:14:40
Message-ID: 20140429001440.12926.25276 () probe ! kde ! org
[Download RAW message or body]
[Attachment #2 (multipart/alternative)]
> On April 28, 2014, 11:15 p.m., Commit Hook wrote:
> > This review has been submitted with commit \
> > 6a896c373e0940c09a5efae6ee067bb9f59be7e0 by Percy Camilo Triveño Aucahuasi to \
> > branch master.
>
> Aleix Pol Gonzalez wrote:
> Oh come on. You're a mathematician, 4 or 6, what difference does it make? ;)
hehehe ... :D :D ok If we use 8 as initial cube edge, then some examples (surfaces) \
in plots3ddemo will not be plotted totally. I guess, this was because our old \
implementation covered more space at the initial iteration (partition) see:
octree.cpp:25 raiz->cubo.setCenter(initial_edge/2,initial_edge/2,initial_edge/2);
With 12 as new size the issue is solved and the surfaces are plotted ok.
If you want a more deeply explanation of this topic, then the answer is in the \
Octree. Since we were using an octree based space partition method, instead Aqsis \
approach is using a fixed interval to traverse through x,y and z. The reference to \
create a new space in the octree method was the center of the father/old cube.
octree.cpp:47 father->nodes[i]->cube.setHalfEdge(father->cube.halfEdge()/2);
Instead, in Aqsis, the reference to generate a new cube is the previous one, since \
you just need to jump to the next cube using the interval length. To jump intervals \
we have this variables, in new version of marchingcubes.cpp:98-99, see:
hx = (xmax-xmin)/i_size_x;
hy = (ymax-ymin)/i_size_y;
hz = (zmax-zmin)/i_size_z;
In any case, and for sure, I trust more in Aqsis implementation than our old one.
If you need more explanation, I can study this to confirm and point out wich of those \
quotients in octree (that are edge/2) are causing the difference ;)
> p
- Percy Camilo
-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/117816/#review56811
-----------------------------------------------------------
On April 28, 2014, 11:15 p.m., Percy Camilo Triveño Aucahuasi wrote:
>
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://git.reviewboard.kde.org/r/117816/
> -----------------------------------------------------------
>
> (Updated April 28, 2014, 11:15 p.m.)
>
>
> Review request for KDE Edu.
>
>
> Repository: analitza
>
>
> Description
> -------
>
> Replace our marching cubes algorithm implementation for the one of Aqsis project. \
> The Aqsis implementation is fastest and offers topological guarantees, in other \
> words we'll have better surfaces as result (without missing faces, etc.)
> Finally, we'll have less code to maintain :)
>
>
> Diffs
> -----
>
> analitzaplot/private/utils/octree.cpp 6106120
> analitzaplot/private/utils/octree.h 1cf4b24
> analitzaplot/private/utils/marchingcubes.cpp 49331f1
> analitzaplot/private/utils/marchingcubes.h ef0b0c4
> analitzaplot/private/utils/lookuptable.h PRE-CREATION
> analitzaplot/private/backends/implicitsurface.cpp 68c7918
> analitzaplot/CMakeLists.txt 9890449
>
> Diff: https://git.reviewboard.kde.org/r/117816/diff/
>
>
> Testing
> -------
>
> Builds ok.
> All analitzaplot tests pass.
> Also, there is no licenses issues since the code is GPL >= 2 (same of Analitza)
>
>
> Thanks,
>
> Percy Camilo Triveño Aucahuasi
>
>
[Attachment #5 (text/html)]
<html>
<body>
<div style="font-family: Verdana, Arial, Helvetica, Sans-Serif;">
<table bgcolor="#f9f3c9" width="100%" cellpadding="8" style="border: 1px #c9c399 \
solid;"> <tr>
<td>
This is an automatically generated e-mail. To reply, visit:
<a href="https://git.reviewboard.kde.org/r/117816/">https://git.reviewboard.kde.org/r/117816/</a>
</td>
</tr>
</table>
<br />
<blockquote style="margin-left: 1em; border-left: 2px solid #d0d0d0; padding-left: \
10px;"> <p style="margin-top: 0;">On April 28th, 2014, 11:15 p.m. UTC, <b>Commit \
Hook</b> wrote:</p> <blockquote style="margin-left: 1em; border-left: 2px solid \
#d0d0d0; padding-left: 10px;"> <pre style="white-space: pre-wrap; white-space: \
-moz-pre-wrap; white-space: -pre-wrap; white-space: -o-pre-wrap; word-wrap: \
break-word;">This review has been submitted with commit \
6a896c373e0940c09a5efae6ee067bb9f59be7e0 by Percy Camilo Triveño Aucahuasi to branch \
master.</pre> </blockquote>
<p>On April 28th, 2014, 11:22 p.m. UTC, <b>Aleix Pol Gonzalez</b> wrote:</p>
<blockquote style="margin-left: 1em; border-left: 2px solid #d0d0d0; padding-left: \
10px;"> <pre style="white-space: pre-wrap; white-space: -moz-pre-wrap; white-space: \
-pre-wrap; white-space: -o-pre-wrap; word-wrap: break-word;">Oh come on. You're a \
mathematician, 4 or 6, what difference does it make? ;)</pre> </blockquote>
</blockquote>
<pre style="white-space: pre-wrap; white-space: -moz-pre-wrap; white-space: \
-pre-wrap; white-space: -o-pre-wrap; word-wrap: break-word;">hehehe ... :D :D ok If \
we use 8 as initial cube edge, then some examples (surfaces) in plots3ddemo will not \
be plotted totally. I guess, this was because our old implementation covered more \
space at the initial iteration (partition) see:
octree.cpp:25 raiz->cubo.setCenter(initial_edge/2,initial_edge/2,initial_edge/2);
With 12 as new size the issue is solved and the surfaces are plotted ok.
If you want a more deeply explanation of this topic, then the answer is in the \
Octree. Since we were using an octree based space partition method, instead Aqsis \
approach is using a fixed interval to traverse through x,y and z. The reference to \
create a new space in the octree method was the center of the father/old cube.
octree.cpp:47 father->nodes[i]->cube.setHalfEdge(father->cube.halfEdge()/2);
Instead, in Aqsis, the reference to generate a new cube is the previous one, since \
you just need to jump to the next cube using the interval length. To jump intervals \
we have this variables, in new version of marchingcubes.cpp:98-99, see:
hx = (xmax-xmin)/i_size_x;
hy = (ymax-ymin)/i_size_y;
hz = (zmax-zmin)/i_size_z;
In any case, and for sure, I trust more in Aqsis implementation than our old one.
If you need more explanation, I can study this to confirm and point out wich of those \
quotients in octree (that are edge/2) are causing the difference ;)
> p</pre>
<br />
<p>- Percy Camilo</p>
<br />
<p>On April 28th, 2014, 11:15 p.m. UTC, Percy Camilo Triveño Aucahuasi wrote:</p>
<table bgcolor="#fefadf" width="100%" cellspacing="0" cellpadding="8" \
style="background-image: \
url('https://git.reviewboard.kde.org/static/rb/images/review_request_box_top_bg.ab6f3b1072c9.png'); \
background-position: left top; background-repeat: repeat-x; border: 1px black \
solid;"> <tr>
<td>
<div>Review request for KDE Edu.</div>
<div>By Percy Camilo Triveño Aucahuasi.</div>
<p style="color: grey;"><i>Updated April 28, 2014, 11:15 p.m.</i></p>
<div style="margin-top: 1.5em;">
<b style="color: #575012; font-size: 10pt;">Repository: </b>
analitza
</div>
<h1 style="color: #575012; font-size: 10pt; margin-top: 1.5em;">Description </h1>
<table width="100%" bgcolor="#ffffff" cellspacing="0" cellpadding="10" \
style="border: 1px solid #b8b5a0"> <tr>
<td>
<pre style="margin: 0; padding: 0; white-space: pre-wrap; white-space: \
-moz-pre-wrap; white-space: -pre-wrap; white-space: -o-pre-wrap; word-wrap: \
break-word;">Replace our marching cubes algorithm implementation for the one of Aqsis \
project. The Aqsis implementation is fastest and offers topological guarantees, in \
other words we'll have better surfaces as result (without missing faces, etc.)
Finally, we'll have less code to maintain :)</pre>
</td>
</tr>
</table>
<h1 style="color: #575012; font-size: 10pt; margin-top: 1.5em;">Testing </h1>
<table width="100%" bgcolor="#ffffff" cellspacing="0" cellpadding="10" style="border: \
1px solid #b8b5a0"> <tr>
<td>
<pre style="margin: 0; padding: 0; white-space: pre-wrap; white-space: \
-moz-pre-wrap; white-space: -pre-wrap; white-space: -o-pre-wrap; word-wrap: \
break-word;">Builds ok. All analitzaplot tests pass.
Also, there is no licenses issues since the code is GPL >= 2 (same of \
Analitza)</pre> </td>
</tr>
</table>
<h1 style="color: #575012; font-size: 10pt; margin-top: 1.5em;">Diffs</b> </h1>
<ul style="margin-left: 3em; padding-left: 0;">
<li>analitzaplot/private/utils/octree.cpp <span style="color: \
grey">(6106120)</span></li>
<li>analitzaplot/private/utils/octree.h <span style="color: \
grey">(1cf4b24)</span></li>
<li>analitzaplot/private/utils/marchingcubes.cpp <span style="color: \
grey">(49331f1)</span></li>
<li>analitzaplot/private/utils/marchingcubes.h <span style="color: \
grey">(ef0b0c4)</span></li>
<li>analitzaplot/private/utils/lookuptable.h <span style="color: \
grey">(PRE-CREATION)</span></li>
<li>analitzaplot/private/backends/implicitsurface.cpp <span style="color: \
grey">(68c7918)</span></li>
<li>analitzaplot/CMakeLists.txt <span style="color: grey">(9890449)</span></li>
</ul>
<p><a href="https://git.reviewboard.kde.org/r/117816/diff/" style="margin-left: \
3em;">View Diff</a></p>
</td>
</tr>
</table>
</div>
</body>
</html>
_______________________________________________
kde-edu mailing list
kde-edu@mail.kde.org
https://mail.kde.org/mailman/listinfo/kde-edu
[prev in list] [next in list] [prev in thread] [next in thread]
Configure |
About |
News |
Add a list |
Sponsored by KoreLogic