Jump to content

  • Log In with Google      Sign In   
  • Create Account


Member Since 27 Feb 2013
Offline Last Active Jul 19 2014 10:58 PM

Posts I've Made

In Topic: Questions About Game Development

10 March 2014 - 02:54 AM

Hello Mark, and welcome to GameDev! :)


1) I'm a college student myself, so I'd like to address a few things that you might be a little misled about. I'm currently in my junior year at university, and I've only now made a solid decision on what majors/minors/etc. I'm going to graduate with. You've got time to try some things out, so don't stress too much about making the perfect decision now! But having passions and a rough idea of what you might want is a great, great idea. College needs more students like that.


On top of that, don't be worried about free time in college. There's tons of it, lying everywhere around you. I take 2 students worth of classes (the usual is 3-4 classes, and I take at least 6), work over 20 hours a week to pay for college (outside of loans), I'm a part of a social club, and engineering club, and a game programming club. The social club takes no real time; the engineering club I'm making control systems and electronics for a UAV, and the game programming club I'm making a game. This game -- alone yes (at least while I make the engine), but many people to help me out along the way.

This sounds like a lot, but seriously -- I still have time to hang out, sleep, have a social life, and relax. I'm not saying do what I do but, whatever workload you have, you can make it work. Most places have gamedev clubs and such -- use those!! You can find the time to do the things you want -- like make games and a portfolio, if this is what you want to do -- if it's what you want and you have the passion for it (and a little bit of skill, and luck!).


When applying to internships, current projects are HUGE!!!!!! I've nailed so many internships talking about the project's I've done and showing off physical results and companies love that. If you have experience, that's worth a ton to a company -- you are more versed in how things go, and you've seen with your senses, exactly how things work and how things go wrong. I've been accepted to an internship because of a project I failed -- and failed as in a UAV turned into a flying flame-ball of death -- but the company saw the experience I gained, especially from the mishap, and how careful and how meticulous my planning is when working on future projects. All in all: experience is key.


2) I don't know much about this because I've only had internships and jobs as an engineer, but I've seen that people like seeing that you have knowledge of a lot, and can handle that, but really are phenomenal at one or two specific topics. It shows that you can learn quickly and retain knowledge, while you can also become an expert in things that need to be done by you, too.


3) I don't know :P


4)  I'd say the internship programs are best. I've had friends go through the program at Blizzard, and they loved it and learned a crazy amount of amazing things during their time there.


5) Got no experience here either, sorry!


I hope this helped. Sorry for the wall of text! I hope the college perspective can help you understand things and show you'll have plenty of free time :D

In Topic: An Ever Changing/Learing AI

24 February 2014 - 06:19 PM


This notion that in order to have learning AI you need programs that modify their own code is preposterous, but also very common. The divide between code and parameters is an arbitrary one, and it's much more productive to think of learning as changing parameters.


The other thing that should be mentioned is that adaptive AI is something you probably don't want in a video game. It's hard enough to debug a non-adaptive AI to make sure that the bad guys don't get stuck in corners or do other stupid things that break the illusion. If the AI can change once it's in the hands of the player, this can become a much harder problem, unless the adaptability is very limited and controlled.

Could you point me in the direction of a good starting place to learn the basics of AI. I don't intend to dive into it much as I stated I am mainly graphics oriented however I would like to have a better understanding of what I am talking about so that in the future I don't make myself look foolish like I did in this post. 



I'd check out the first post on the AI Forum...it'll point you to many great books and sites to get your AI blood flowing!

In Topic: A book to match with “Mathematics for 3D Game Programming and Computer Graphics”

13 February 2014 - 06:32 PM

Hell I never knew about LWJGL debug mode until very recently when I saw it in the sources and googled it, and I started using LWJGL one year ago.


Thank you, good sir :D

In Topic: Working with A* in a 3D voxel world

12 February 2014 - 01:10 AM

I think you have the right idea that it's best to start simple,and optimize only if you need it, as it may turn out your satisfied with the simple solution.


I think you're right. I'm just going to use A* with JPS and calculate per-frame for NPC's. They'll also have a limit to their search so their path list doesn't get too long :P But 32-64 path-points per NPC should be good for now.

In Topic: Working with A* in a 3D voxel world

30 January 2014 - 01:45 PM

Ok I'm going to say a little more on my implementation of my game so people know how things are running under the hood:

  • The game is voxel-based, like Vox or Minecraft.
  • The world is made of an 32x32 grid of Zones. Each Zone is made up of 1x16x1 Chunks, and each chunk is made of 16x16x16 voxels, arranged in a rolled out, 1-D array of shorts (Voxel ID's). The world is then 496x256x496 voxels. This might extend as far as 1024x1024x256 voxels if I want to push my graphics card and optimize/multithread my single-threaded game engine, pushing terrain data as far as a little over half a gigabyte of terrain data.
  • For path-finding, I query these voxel's faces to see what areas the NPC can move from. This way, no NavMesh or anything like that needs to be generated, saving on a ton of memory. Querying for 10-100 voxels a frame is no problem.

In THIS case, I'm looking at how this scales, and if I need to generate any type of navigational mesh. Obviously the path points I set for the NPC is every block; maybe I can save some memory with spreading out points (JPS may help with this) using an optimized navigation mesh? Terrain data takes up almost 256Mb data in memory, and re-meshing chunks takes a bit of time too. I need to balance time and memory for AI navigation.


What I've seen so far is that I can limit the number of queries an NPC makes based on two things: frame length and number of NPC's. Maybe only navigate 1 zone at a time (or possibly something a little larger, since NPC's are 1x1x2 voxels in size and a zone is 16x16 on it's top face so...maybe 2x2 or 4x4 zones at a time...?) to save processing power. How would this scale, though, to large castle, city walls, etc...? Would the NPC be forced then to follow the side of the wall all the way around...?

Maybe I'll have to play with this some more, and I'm just over-thinking this right now.