Hello everyone,
Thanks for your replies! I know it's been four months since I posted on
this thread, but I'm still interested in doing the project. I've been
caught up with university course work and other parts of my project. I
apologize for the lack of communication.
So, my project timeline started in the first week of December. The total
time required for this project is 4 months. That's a month in December and
then February to May (1st week of May is the submission).
I've titled my project:
*"Analysis, Effective implementation and Applications of NURBS in CAD"*
The month of January was dedicated to completing my university course work
for the final semester, leaving the next three months (Starting from the
1st week of Feb), for project development. Instead of doing just a coding
project, I thought I'll expand the scope of my project to also front-end
analysis using CAD softwares. I used the entire of December to do the
analysis part of the project. In this time period, I studied the
applications and features of NURBS in different CAD softwares (Open source
and Proprietary). Analysis would mean how effectively an algorithm is
implemented in different softwares. Though I've made some progress in that
part of my project, I'm still not very convinced as to how I can compare
different implementations of the concept of NURBS. I have a few questions
regarding this:
1. Which are the CAD softwares that I should compare for the best
implemented models?
2. On what basis do i compare? ( I was thinking the visual output and the
algorithmic output)
3. When comparing algorithms, what factors do I use to compare? ( Like
runtime, effectiveness etc.)
After doing this sort of analysis, It would enable me to understand the
problems and advantages of different NURBS implementations across different
CAD softwares. I am then planning to effectively implement NURBS in CAD
(Choosing one of the NURBS projects in the project ideas page. "Effective
implementation" would mean to atleast "try" and make a more effective model
compared to other softwares. I really don't know how I can achieve that.
Would need guidance on that, Is it possible to build a better algorithm
than the ones that are out there?). After implementation, I want to study
the outputs using BRL-CAD(Applications).
As far as the project idea is concerned, I thought I could pursue the
project of converting basic implicit models to NURBS format. I found this
project on the following link:
http://brlcad.org/wiki/Implicit_to_NURBS_conversion
First off, I want to know if this project is the same as the one found on
this link titled "Implicit-to-NURBS"?:
http://brlcad.org/wiki/NURBS_TODO#Implicit-to-NURBS
And I had a doubt about the CSG/BREP functionality in BRL-CAD. So is NURBS
implemented only using BREP or also CSG? (I could understand that since
NURBS is a surface representation format, that BREP would be more apt).
But, is it possible to implement NURBS using CSG?
My aim is to try and work on some smaller NURBS projects on the TODO page
before getting into converting implicit primitives. This could help me in
understanding NURBS better and would also give me more experience with it.
Some suggestions for smaller projects would be of great help!
So, by conversion of implicit models, does it mean that I need to convert
all the primitives listed on this page:
http://brlcad.org/wiki/BRL-CAD_Primitives
to NURBS representations? How many of those are already converted to NURBS?
As far as my understanding of NURBS is concerned, I've been studying the
basic equation and its derivation. I have learnt about Control points,
weights, knot vectors and degrees of curvatures and how these parameters
are necessary to draw a NURBS curve or surface.I had a few questions
regarding it:
1. Can we modify a NURBS curve only by changing the knot values and control
point values?
2. How do we know what set of knot values and control points would give
what curve/surface?
3. So is every primitive shape that I need to code, differing only in knot
values, no. of control points and the BREP structure?
4. Would this entire project be just to modify the knots, control points
and BREP structure of the different primitives I need to code?
I also want to know if I have enough time to complete this project? My
mathematical fundamentals are pretty strong and I understand the concept
behind NURBS. I've been reading up from different links about NURBS
equations and properties of NURBS curves. Is there anything more than the
fundamental NURBS equation that I need to learn for this project?
Finally, Is this a time intensive project? Would I be able to complete it
before the first week of May?
I am very excited to work on this project and would love support from you
guys. Again, I apologize for the lack of communication and the overload of
questions on one single email. A continuous and sustained relationship is
important for productivity, I promise to be more active on the mailing
lists and IRC.
Thank you,
Kaushik
On Sat, Oct 15, 2011 at 10:33 PM, Abhijit Nandy <[email protected]>wrote:
> Hello Kaushik,
>
>
> > Tons. NURBS is definitely the hot topic, but is also one of the
> hardest. If you're up for it, tessellation of NURBS geometry (i.e.,
> converting a set of trimmed NURBS surfaces into a polygonal mesh) would be
> a great self-contained project that isn't "too hard".
>
> Ah, now thats vetting my interest in NURBs :).
>
> So I see that Sean already replied. To add on to it, perhaps you can try
> to play around with some of the NURBS geometry code in BRL-CAD first before
> deciding. Simulation is also a great idea and I can help you get started in
> the related code. Solid modeling is a bit different from how one would
> proceed with 3D modeling tools like 3DS Max, so you can try some of the
> tutorials and example programs first.
>
> Also if you hang around in IRC then you can get a lot of info from
> developers who have known and are familiar with the software for a long
> time.
>
> Regards,
> Abhi
>
>
> ------------------------------------------------------------------------------
> All the data continuously generated in your IT infrastructure contains a
> definitive record of customers, application performance, security
> threats, fraudulent activity and more. Splunk takes this data and makes
> sense of it. Business sense. IT sense. Common sense.
> http://p.sf.net/sfu/splunk-d2d-oct
> _______________________________________________
> BRL-CAD Developer mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/brlcad-devel
>
>
------------------------------------------------------------------------------
Try before you buy = See our experts in action!
The most comprehensive online learning library for Microsoft developers
is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
Metro Style Apps, more. Free future releases when you subscribe now!
http://p.sf.net/sfu/learndevnow-dev2
_______________________________________________
BRL-CAD Developer mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/brlcad-devel
