Advertisement Jump to content
  • Advertisement
Sign in to follow this  
  • entries
  • comments
  • views

About this blog

Eventually I will start somthing... I Promise.

Entries in this blog


Graphics Renderer woes

Currently I am toes deep into my 3rd year project/game/game engine. This is my attempt to mull together what i know and learn what i dont. But the first and formost I am hit with the dreaded graphics engine.

The dilemma is that for my finale year project at uni the stipulation is you have to build your own renderer. I dont like graphics, i've never played a game for graphics i comprehend the basic and thus what to build a game engine with a simple renderer. So i have been trying to learn opengl after a years worth of unversity being taught directx. I am also trying to develope genericlly so i can port my work to the mac os x once i get a new one having sold my bottom of the line underpowered ibook.

Not being a fan or impressed by graphics in game i plan to make a cel shaded renderer simply because at the heart... or rather at the sight its basic and your gameplay has to shine through other wise you just have months and months of work on a gimmick.

I hopefully building a very object oriented project right now I only have the renderer which only creates a window. It tends to complain when i exit out and try to return to my native resolution. I dont know if I should be concerned about that at this stage seeing as i dont even have a cube to the screen.

In making the renderer of my game engine I mulling over the archetiture and things I can cut or dont need but a simple renderer. In my head I just want a camera and models loaded. Questions like is there a point to creating a primitive draw class to dynamically draw 3d shapes? Should i load more than one kind of image file or model file etc. Basically how much is need for the bare minimum for a renderer but not to minimum that i hurt myself.




The after tastes of kiwis

Life Update: Currently I am heading into this semester end of term exams and revision as well as a evaluation write up for HCI (human computing Interaction).
But not without completing a game in between breaks. Castlevania portrait of ruins was GREAT and difficult even on normal. I've never played a Castlevania game so this game has now ensure i test all others to come.

Quiet a refreshing thought when you discover why people don't finish games. When I couldn't afford to buy my own games and live under the stipend of my parents I ensured I completed every game I own and more than once on all the difficulties i could beat. Now that i buy my own games that whole Need to complete games is over. I hate those review sites now days and yet there is not medium for you to test a game before you buy it pc,ps3,x360 has demos, sure but I don't own the later 2 and any demo that is on the pc rarely seems to be a game i am anticipating/want, The last was Company of heroes and I only just purchased that last month.

On Monday my course got a talk from rare about studio live and graphics, the game kameo just seems so much sweeter know the techy stuff behind it. Odd since I don't even own a 360, although was thinking about it solely based on the whole xna aspects and the very few uber titles that wont grace the pc.

I will no doubt be posting my 3rd year project on my journal, but I also need to do side pieces from CV purposes.

My finale year project I am going to focus on game play and building a game engine/mechanics of a game as well as story and general game design. My lectures want us to do one specific topic in an area we would must likely like to do in industry. Thus I am focusing on game play as I want to do that stuff and not AI or Graphics.

So I am going to do a game and build a render with opengl 2.0+. I also had the whole "It be cool to do it on 64bit" but i don't know where to start on that. Thus is the whole premise of the 3rd year project "self-learning".

I have plan to API alot of stuff as it wont be a focus on my prodject, well not alot of stuff but physics and is there a AI api? I was going to do sound myself but depending on timing may or maynot be an API.




Intro April 2007

Currently I am sitting at my pc stomps as to what to code next or should i go to bed? Its currently 2 in the morning.
The command prompt pops up, thus informing me mechcommander 2 source code just finished installing.

It was free from microsoft.

Current project:
I am working on a second year collaborative project where the two game course at UCLAN come together to make a game.

WHOO frigging HOO.

I have not enjoyed this project from the start, simply and purely because it is boring and not fun.

Shouldn't a game be fun?

But it is for a grade and soon will it be for a pay check?

About a month ago team17( makers of worms) came and talked to us, it was cool. They just gave us an industry point of view from there dev studio.

Some questions were asked like what are you working on now, how decides what gets made.

Team17 pretty much confirmed the publisher decides the game. This bring back to the fright that you could or will be working on a game that is not fun or just horrible.

This group project is "an open battle field tank game", I am responsible for the enemy AI.

This in itself is a problem as I like AI but I don't want to specialize in it. To clarify "like"; In games I buy I will always be amazed at what the AI can do over how pretty the game is.

Currently in this project I have sorted out a FSM, but that is about it. I never been motivated to do much work on it, opting to work on all other projects and neglecting this till the last 3 weeks.

Things I have to do.
-Targeting: I want to get the tank to turn its turret and fire on the player, currently There is a FOV, when the player is "seen" then turret is lock at the point it saw the player.

-Timing for shooting: First toooooooo slow now tooooooo fast. I need to get the tanks to wait, setting up a reload time before they can fire. SO I need to figure out converting the frames per second to real life seconds.

-Steering: Originally I was simply going to have a hillclimb algorithm, The one on this site in game. But we have far less building in are terrain. Now I will sort out a steering avoidance. where the AI will get a target point and go to the point. If there is a building "arch" around the building and continue the straight line.

-Other: Stuff I am missing that will show up later.



Sign in to follow this  
  • Advertisement

Important Information

By using, you agree to our community Guidelines, Terms of Use, and Privacy Policy. is your game development community. Create an account for your GameDev Portfolio and participate in the largest developer community in the games industry.

Sign me up!