Looking for Bachelor Thesis Ideas CompScience

[TheBeardyOne]

Hey Guys,

In about 2 weeks I will start working on my Bachelor Thesis in CS. My educational background is Game Programming so I think they would like to see some relevence to game development for the thesis.
Anyways, I have some ideas for what framwework I would like to work in:

- NO user studies
- Should be performance related
- Something with DirectX 12

 

The scope is 15 ECTS credit points and the time span is about 3-4 months.

A teacher of mine suggested to maybe look into data traversel for big scenes in DX12 (Harddisk->Memory->Graphics card).
This would be relevant for something like open world games, which has become very popular amongst developers these days.

Anyways, I'm really terrible at coming up with ideas.. so I hope you guys could help me with ideas or just hint at something that you think I should look into.

Thanks in advance!

[Guest]

Real-time vs baked lighting? 

Raytracing on rtx cards vs raytracing on gtx cards? 

[Lord Mirdalan]

Try to make procedural maps better... 

[wasab]

I prefer to just hack and code than working in labs or write papers. 

[TheBeardyOne]

On 3/24/2019 at 12:54 PM, fpo said:

Real-time vs baked lighting? 

Raytracing on rtx cards vs raytracing on gtx cards? 

A comparison of raytracing RTX vs. GTX Compute shaders would imply that I need to buy a RTX card.. so probably won't do that ^^

[TheBeardyOne]

21 hours ago, Lord Mirdalan said:

Try to make procedural maps better... 

That's a pretty broad statement. In which way do you think they are bad?

[TheBeardyOne]

15 hours ago, James Evens said:

Does not work the other way around? they give you a topic related to there work

No we can present an idea and if they are okay with it you develop a plan. If the plan is good you just implement it and write down the thesis + opposition.

[Lord Mirdalan]

3 hours ago, TheBeardyOne said:

That's a pretty broad statement. In which way do you think they are bad?

I don't think they are "bad" per se, rather it's an interesting algorithmic area of study, one where you could delve into the minutia and have some fun.