6000 word essay. Need topic suggestions.
 

I have to write a 6000 word essay for something called 'The extended project', which is equivalent to an AS but I'm struggling for topic ideas.

It can be on anything. Anything. Ideally I want to do it on something related to Computer Science and it needs to be something that is open to discussion.

I thought about doing it on the increasing importance of GPUs but I think I might struggle to get 6000 words out of that.

So has anyone got any ideas on at least a vague topic I can write about?

Any suggestions will be greatly appreciated.
:beer:

Elephants :p

Re: 6000 word essay. Need topic suggestions.
 

i dont understand this part, it almost sounds like a key thing that will determine the type of essay (like how academic it is) before i can suggest anything

also what computer science thongs do you know about? would you rather do something you're familiar with or something thats totally new to you?

Re: Re: 6000 word essay. Need topic suggestions.
 

Haha.... do they even exist?? And no, don't really want to know if they do...

But on the topic, you could write about "Why RAID is not a backup" or "Mac vs PC" :p or "The considerations, pros and cons of render farms"

Re: Re: Re: 6000 word essay. Need topic suggestions.
 

BWAHAHA, didn't even know i did that typo!

6000 Words is not a lot really.

Thers a lot to go at with regards to parallel processing, the history and development would fill 6000 words easily, before going on to specilist applications within the GPU and the rise of the GPGPU with stream processing, development tools such as CUDA and the overall advantages and disadvantages for each.

6000 words would be easily acheivable on the subject for a well researched and referenced project.

I considered it, but it's not really relevant to the course I want to do... (I checked, there is no course on elephant :hug: )

We need to log 100 hours (I could write 6000 words in 10) so I think it's intended that we spend a lot of time on research, so something new would be good.
But I don't really know about anything that in depth anyway.

The project is intended for the best students (the only people that really do it are the ones trying to get in to one of the best universities). From the examples I've seen, many aren't that academic. But I think it is intended to be, especially since I'm doing a science (and most examples I've seen were for english type subjects). I would certainly prefer to do something more academic and it would probably make me look better if it was more academic.


lol. As far as Mac Vs PC is concerned, I think it would be a bit subjective and not academic enough (if my interviewer had strong feelings on the subject it might not go down well). Most of the people that apply for computer science seem to be gamers that only take the subject because they a) they can't think of anything else and b) seem to like anything to do with games so just do computer science because it is in some small way related and they can't think of anything else. I think these kinds of people would do something more trivial and I want to try and differentiate myself from them.

The other two are definitely good starting points, mostly because they are open for discussion. However, now that I think about, I think I want to do something more academic, that will require significant research.

So do you think this might be a good idea? Do you think it is a subject that's open to discussion rather than just stating facts?


Thanks for the help/suggestions everyone, it really helps. ;)

i would probably say that mac vs pc is a plausible topic, however, you can't take a fanboy approach, if it were me i'd take a whole hci (human computer interaction) approach
and by that i mean looking at each operating system from all the hci principles (things like how easy it is to use the interface).
they both break a lot of them and you REALLY can't be a fanboy to do it or you sound like someone who can't think and analyse
heck you could even pull other applications into the mix too.
maya vs mac vs pc
but if you're like any other normal person (one of those gamers that are doing compsci because they think they'll get into games not knowing that you actually have to know math and have decent problem solving and communication skills) then you'd find hci stupid and you will refuse to believe any of it or take it seriously because you'd rather not question anything or understand why someone says certain things

doing that though would require a lot of work since it's probably an entirely new concept/way of looking at stuff, but you could easily fill up 6000 words saying they all suck but none is better than the other and explaining why (you can provide graphs and stuff too... if you knew what i know about hci then you'd understand why thats possible)

you could look into inverse kinematics or various other animation techniques as well. and ways to improve it.
equally do the same with ray tracing.
you could do the history of computer graphics, like where blinn shaders came from and the math behind the magic matrices and texture mapping
but that'd be more of stating facts and you wouldn't be able to argue a point.

or you could go down an AI route, and explain various learning algorithms, when they work best, how to make them not work so well.


but yeah... i wouldn't say stwert's suggestions are completely off a plausible essay. you just have to take a realistic approach rather than a fanboy approach.

what an essay in response to the time article about hollywoods hi-tech sweatshops, and that guys open letter to james cameron about how 3D artists get a bum deal when it comes to film credit.

http://www.time.com/time/magazine/ar...990803,00.html

http://www.huffingtonpost.com/lee-st..._b_451922.html

do it on GPU and rendering

Yes, the intorduction to parallel processing sets the scene with history, discuss the process of "standard" parrallel processing, then go onto the GPU and its stream processing and limiting factors of uning this over other methods say genetic algorithms as they are limited in stream (if I remember correctly!).

The GPU is also limited in its instruction set over a CPU and therefore cannot "do" everyting a CPU can faster. There are lots of examples of advantages and disadvantages of the GPU for processing. In effect when it comes to a lot of stuff (encoding video, rendering etc) your taking advantage of its vector processing, thought this may not be to all applicaitions.

There are also power issues, not too sure of the difference (good to look up) is processing on the GPU watt/instruction more economical? Dont know the answer to this but worth a look.

Propose the idea and possibly package it as "Will the emergence of GPU processing mean the demise of the CPU" or something like that.

You may even get into giving CUDA a go, I did except I managed to make my screen black and white and could not get it back to colour untill I uninstalled it, even with a qualified quadro GFX card!

I would run the idea past the tutor, to get a feel for what they are after.

Cheers

Steve

You certainly make a convincing argument! I'd still be worried they wouldn't see it as that academic. It would be fine in an interview but most universities don't interview and would just read my personal statement, in which I'd be lucky to have enough room for the title and while I could give it a title that would it wasn't the typical mac Vs PC war flame bait, I still think they would see the topic and jump to conclusions (or at least not think it sounds that impressive).

To be honest, I think the tutor would just give me a blank look. He's not a computing/IT teacher (and my Computing teachers don't know/don't care about it).
I know it's marked by one of my teachers, but I'm not sure who. I better check to see if it's a computing teacher or not, because I might be in danger of getting too technical (although I do seem to remember there was at least the option to get it marked by a specific appropriate subject teacher).

The course is new (it's been done one year (but they haven't been marked and submitted yet)) so I'm not even sure the tutors know what the exam board expects.

But I think I'll probably do it about GPUs. It should be a good opener during interviews as well.

I would suggest Semiconductor .

Not really sure that that would be much of a discussion.

I didn't take the time to read everyones replies so my apologies if someone already mentioned this.

How about Real 3d. The technology is impressive and everything is going in that direction anyway. They already have 3d tv's. Soon, we will be playing video games in real 3d. Could be an interesting topic.

Just google "3d tv" and stuff like that. Should be plenty on the subject.