Part of the Khronos Group

The Industry's Foundation for High Performance Graphics

from games to virtual reality, mobile phones to supercomputers

Results 1 to 2 of 2

Thread: Bachelor Thesis

  1. #1
    Newbie Newbie
    Join Date
    May 2017

    Bachelor Thesis

    Hello, I need an advice for my Bachelor thesis.
    I'm studying Computer Engineering, but I've understood that I really don't like it.
    I plan to continue my studies in the graphic field, but now to complete this bachelor I have to prepare my bachelor thesis (and it has to be in the computer science field ;D)
    Knowing the fact that I like the graphic field one of my professors has proposed to me to do a thesis in OpenGl,
    probably something with bezier splines and mesh generation through them (and he even mentioned OpenCl).
    I know that programming in OpenGl is not like doing graphics in an artistic way, but it's the field closest to my interests now.
    Now, I know very little about openGl, but I'm studying the "superbible" right now and, from the time I'll have finished it, I'll have 3 months full-time to work on a thesis.
    My question is: with beginner knowledge of OpenGl (and a good background in Math and Linear Algebra, two things that I like enough) is it possible to do
    a good thesis in OpenGl in 3 months of total dedication? And which could be an interesting argument?
    I really like something like procedural generation of terrain, trees and so on, but knowing nothing about the argument I don't know if I can really do it in the time I have.
    I have to decide if trying it or give up and do something that I don't like but that is easier to achieve (probably something about databases).
    And I can't ask for any advice in my university because all the thesis here are about big data, database, cyber security ecc.
    I really hope a thesis like this could be well seen from the commission.
    I'll ask my professor for further information next week, but I'd really appreciate advices from people who know OpenGl
    thank you very much

  2. #2
    Intern Contributor Brokenmind's Avatar
    Join Date
    Feb 2014
    Aachen / Germany

    some thoughts, from my understanding (might be wrong):

    I think I know how you feel (ok let's face it, people who frequent here are likely to outright LOVE openGL ). I'd have tried the same at the time, but it's not that easy... Most thesis topics from CS are simply mathematics used for some CS topic. OpenGL for that matter is simply huge amounts of math (be it for display or underlying structure) with a screen to see the final results. So finding a purely OpenGL topic to write your thesis about might end up very difficult...

    I know that one of my bachelor colleagues wrote a thesis about something like the cyclic occlusion of triangles or how triangles cover each other in a loop (can't access it now because login credentials invalid), so it is in definitely possible. He took a bit longer than the others, but he now has his BSc
    My thesis was about the intersection of geometric bodies consisting of triangles. Not really mathematical as the most math I used was some cross products and matrix determinants, but it made for many nice pictures and I'm completely happy with it.

    What your professor mentioned about Bezier splines (they came up in my BSc defense and had nothing to do with my thesis, wow) and mesh generation... Not sure how they relate to to OpenGL as again, this is just pure maths. If you go to OpenCL, as far as I know that's just numerics done using GPU so you might even lose the visual aspect.

    That said, you say you like graphics but know very little about OpenGL. If you have never used it before and just want to learn it for the thesis, I think it's a bit over-ambitious because the field is just far too broad. If you had a promising topic at hand and needed to specialise for that, 3 months might be enough. And if you don't exactly know that you will be wanting to heavily do OpenGL in your later job, you don't need to prefer the attractive and fun-to-do OpenGL thesis in favor of the easy-to-do database thesis. I know how shallow those can be, and colleagues passed with their theses where I had never believed they would.

    If you really want to read into procedural landscaping, there might be some nice places to look for inspiration (let's see how many of these you already know):

    Other gimmicks: What the God of GLSL can do with mathematics (and his youtube channel)

    I hope I didn't scare you off with all this because it's pretty advanced stuff... But if you really want to do it, you can surely find some good topic in there.

    What might be an interesting topic: Compare different types of pseudo-random noise functions and determine whether they are "really" random, where their limitations are, what you can do with them and so on...

    If you need more input on this, I'm all ears


Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts