View Full Version : Thesis in Graphics!

09 September 2003, 01:56 AM
Hi, I'm quite new here <n00b> I'm just starting to read the posts, anyway I'm a university student in search of a thesis subject, in of course the most exciting topic in the computer universe, 3Dgraphics! :wavey:

Anyway can you guys help me out on whatz the hottest craze/challenge/technical hurdles out there that can make a good thesis topic? Something like raytracing on Programmble graphics hardware...Shadows/Lighting Algorithms... newEngineEnhancements that may make a more realistic rendering ala FinalFantasy!

Like anything that may interest you if given the chance to have a dissertation/Research on a specific field in the 3D graphics arena. May it be hardware/architecture or in the software rendering techniques/mathematical optimization algorithms...

Appreciate your help, Thanks! =)

09 September 2003, 02:56 AM
Also some ideas that can be posted in a scientific 3dGraphics journal, with some very technical aspects that Tim Sweeney and John Carmack may even regard as interesting stuffs will do... something that may paved way for some other students out there interested in doing some Techie Graphix stuffs!

Just any related idea will do but I challenged everyone with their creative talents to really come up with some great and fascinating ideas!

-Thanks! =)


09 September 2003, 10:43 AM
Well you should learn things in the following order:

- Basic 3D mathmetics (rotation, translation
- wireframe drawing
- poly filling
- texturing
- viewport/camera programming

that was the basic stuff :P

now the more complicated stuff:

- subdivision modelling
- advanced shaders ( A must for photorealistic rendering!)
- Physics engines (gravity and other dynamics)
- (Advanced) particle systems (For the effects part of it)
- (Advanced) Ray tracing (For sharp realistic shadows)
- Shadow maps (For soft shadows)
- Global Illumination (adds very much realism to a scene)
- HDRI (this one too)
- and many more, that I can't come up with at the moment...

And you could alway develop your own new "technologies" ;). However I think the ones mentioned above will keep you busy for the moment :p :D

09 September 2003, 10:07 PM
Thanks for the advice. Yep this could make me busy for a while.

Anyway some other new technologies are exciting by themselves. Maybe like cell-shading and the Carmack-reverse stuffs, how about those and the other "emerging" graphical technologies, I also heard some new exciting stuffs that may well be implemented in games..? huh any suggest other specific topics that you maybe thinking of [EvilBasterd]?

09 September 2003, 11:51 PM
I'm personally also trying to learn 3D programming.

Ofcourse in software. Because you learn the actual algorithms that way.

I almost forgot: Some other important programming feature you MUST implement in a renderer is a z and or w-buffer. For removing faces that are not used at the moment (because you ain't lookin at them. )

Cell shading is cool, but it's alreay part of (advanced) shading. With knowledge of shaders you can create this effect and many others :)

Another nice feature is motion blur, lens flares (if wanted), reflections and last but not least anti aliasing

09 September 2003, 07:53 PM
Ok thanks again,:thumbsup: I'll try to also look for your other suggestions, I'm also looking forward to writing a short article, [10-15 mins reading, 'bout 5-8 pages] patterned in a book I'm reading lately. The title is 'Problems in Graphics' (btw what I'm reading is 'Problems in Mathematics'), which like the book will cover topics open for further developments that can be addressed by the technically inclined public. Who knows maybe people may even get some nice ideas to delve upon... So I'm starting this thread as maybe to gather some topics for my thesis and for the graphical world in general (for the article) . It would only take a little of your time to think about it, so why not try. [Evilbasterd] has already given some ideas, maybe others could give more. :rolleyes:

I am already thinking of also a kind of new 'Shader' langauge for other game-centric application which integrates the best of what the current API's [OpenGL2.0&DirectX9.0b] can offer...


09 September 2003, 09:17 PM
Ughh!!!! No!!!! Please not another Shader language. Nothing personal here and if ts just for learning then great all the more power to you. But for me and many others there are WAY to many different Shading lang's out there. What really needs done is a standard set for which regardless of the Shader lang used it would work on all platforms, hardware and API. From there one could add on extensions for whatever one might want to torment the rest of us with. Sorry for my ranting I just hate the fact that I could use CG to do this wondeful thing, but oh wait it has to be on an Nvidia card. Or I could do that wonderfully thing with DirectX But it has to stay on Windows, blah, blah. Renderman's shading language had the right idea it just never flourished the way they and we had hoped it would have so many years ago.

CGTalk Moderation
01 January 2006, 02:00 AM
This thread has been automatically closed as it remained inactive for 12 months. If you wish to continue the discussion, please create a new thread in the appropriate forum.