How does a bot work?

[SayianSentry]

I'd be interested in seeing your syllabus or something showing this as a class. Study project. Whatever it may be.

I think you dont quite understand xD where i live it is normal to have a study project in 3rd semester computer science. There are multiple projects and they are different every year. It is something the professor decides. It is more about learning to work in a group and getting project management right

[Jaladhjin]

I think you dont quite understand xD where i live it is normal to have a study project in 3rd semester computer science. There are multiple projects and they are different every year. It is something the professor decides. It is more about learning to work in a group and getting project management right

Well it's pretty cool that's something offered in a school anywhere.

Just seems kinda far fetched like a history class doing the evolution of beer pong

[Filint]

In my degree course we have to do a self guided project in the 3rd year - we can choose what we want within reason and have to research, design and develop a piece of software. We also need to document the process. It's pretty fun and very rewarding. Then have to do the same in the 4th year if doing a masters :P I thought it was a standard computer science thing

[SayianSentry]

Yeah i almost have the same thing. just that the all the computer science lecturers decide what projects exists.

[Sklug]

In my degree course we have to do a self guided project in the 3rd year - we can choose what we want within reason and have to research, design and develop a piece of software. We also need to document the process. It's pretty fun and very rewarding. Then have to do the same in the 4th year if doing a masters :P I thought it was a standard computer science thing

Self-guided projects would depend on the course and the way the student organized their schedule. Typically some CS 300+ "CORE" class would have some self-guided project in your 3rd year, as well as other upper division elective courses may have your self-guided, but with limits set by the professor, as in (focus on a self-guided project interacting with a database).

As for the "Have to do the same 4th year if doing a Masters," I'd say that's not really true, at least in regards to the US market. I've noticed in a lot of places to get a Masters just means 2 years of extra classes. However, typically in the US, Master's programs are quite competitive and applying to them is like applying to a highly competitive job position. You can't just choose to keep taking classes, you have to be accepted. Thus, your senior level classes and core classes would be wholly independent of any requirements to get into Graduate school for a masters. I'd say what would play a much higher predictor of getting into grad school is GPA and performance at previous internships that you had taken, or maybe you already had a lot of experience working in the industry and you were now coming back to finish a Masters.

Just my experience.

[Filint]

Self-guided projects would depend on the course and the way the student organized their schedule. Typically some CS 300+ "CORE" class would have some self-guided project in your 3rd year, as well as other upper division elective courses may have your self-guided, but with limits set by the professor, as in (focus on a self-guided project interacting with a database).

As for the "Have to do the same 4th year if doing a Masters," I'd say that's not really true, at least in regards to the US market. I've noticed in a lot of places to get a Masters just means 2 years of extra classes. However, typically in the US, Master's programs are quite competitive and applying to them is like applying to a highly competitive job position. You can't just choose to keep taking classes, you have to be accepted. Thus, your senior level classes and core classes would be wholly independent of any requirements to get into Graduate school for a masters. I'd say what would play a much higher predictor of getting into grad school is GPA and performance at previous internships that you had taken, or maybe you already had a lot of experience working in the industry and you were now coming back to finish a Masters.

Just my experience.

Yeah, the self guided project only makes up one eighth of that year's marks. There are 7 other modules over the year - of course all include small assignments like you described.

Masters are really competitive here too. I was lucky enough to get on to the masters course based on my academic performance - in this case it's an extra year tacked on to the end of the BSc degree. Approximately 1/5 of masters students are then offered to continue with a PhD - my hope is to be able to do that. Like you say internships play a massive role here too - luckily our careers department has some fantastic connections which is pretty helpful.

[ioctl]

Heya Guys,

in my university i signed up for the study project: Automisation of MMORPG's(especially WoW).
But the thing is....there is no literature for that. So i kinda need your help
I just have a few questions it would be awesome if you helped me out
Also...im sorry if i missed out some threads which tell me exact things im going to ask you now:

1. What has DirectX to do with bots?
2. There are index databases but i dont have a clue what they do inside of bots? (maybe getting informations faster out of the memory)
3. How does the bot avoid Warden?
4. How does a bot work at all? (i have an idea but im not sure if that is the real deal)
5. How can programmers update their bots that fast?

Im sorry if im a real noob :/ I started this project a few months ago and i thought this is the best way on getting real informations

By the way i dont need to write a bot myself...i just have to describe how a bot works

Thanks =)

1. Some people inject code into the DirectX screen flush handler to render interesting stuff on-screen as part of the game world, but it is mostly a convenient place to avoid synchronization problems with the main game -- the game does not update internal state while executing DirectX, so you can read in-memory structures without worrying about stuff changing out from under you.

4. In general the inputs for your bot are some combination of:
a) Image analysis of the game screen.
b) Reading game memory and picking out interesting structures.
c) Collecting data using the game's built-in modding API
d) Analyzing data files provided with the game (various data tables, terrain/model data, etc.)
e) Analysis of the network traffic (this is pretty hard, as traffic is encrypted).
Choices for your bot's output are:
a) Synthetic keyboard/mouse events sent to the game.
b) Modification of game memory to trigger interesting events
c) Tricking the game into executing your code for you.
d) Spoofing network traffic (again, pretty hard)

What you do in-between input and output depends on what you're trying to do, and how much effort you want to expend. For something simple like a fishing bot, you can get away with continuously scraping the screen for blobs of the correct color, and blindly clicking on them. A more sophisticated approach could involve reading the list of nearby game entities out of memory, and calculating their positions on the screen. More complex actions (moving around, collecting engaging in combat) could require something you might run into in an AI class, like a hierarchical finite state machine or a planner.

5. Most of the time, very little changes between releases. Often times, just updating a couple of well-known memory locations will be sufficient to get a bot up and running. It's also possible to create tools to automatically extract required information by static analysis of the binary, or inspection of the running game.