• Content count

  • Joined

  • Last visited

Community Reputation

141 Neutral

About zgzg2020

  • Rank
  1. Hi, I am doing some research using the Game engine ioquake3. We were able to reduce the processing time to a certain amount while maintaining the Game's sanity. As a next step, I am looking for ideas to benefit from this additional CPU space. I want implement ideas that would prove that this additional power is beneficial. Currently, I am reading in these forums to find some ideas for it. So, I thought why shouldn't I just ask and see what answers I may get. For example(this is just off of the top of my head, so please don't focus on this example), with more CPU power, you would have the possibility for bigger maps with more players. This should open doors for creating more diverse style of play-modes, such as King Of The Hill...etc Any feedback is welcomed!
  2. Thanks SimonForsman. You cannot imagine how encouraging this is!! I have downloaded the SDK for HL2, because I read on their Wiki that many functions that I am looking for are already implemented there. As for the Quake I or II, my professors weren't see eager when I spoke about them, because they are a bit outdated, at least to their perspective. And I want to follow what they think. Thank you again very much for your quick response. However, I can't understand why there aren't any other replies to post.
  3. Hello everyone, Me: I am a fully capable of programer. However, I have never developed any Video Games yet. So, my experience and knowledge in Video Gaming development is NONE. And, my AI experience is also NONE. Target: I want to experiment with AI Parallelization in Video Games found in games such as Metal Gear Solid, where the AI Agent's main behaviors are behaviors such as Hunt, Attack, Flee, Guard, Patrol...etc. Therefore, I WANT to DEVELOP a fully functioning environment with fully functioning AI Agents. Details such as Weapon, complex maps...etc are unimportant. Done so far: I was naive enough to begin reading about OpenGL, in hopes of developing the program at least with minimal requirements. I am not conidered with graphics. My main concern is VERY SMART and VERY FELXIBLE AI! That is why began working building my miniture world on my own. However, what I have "Done so far" is not worth mentioning at all. What I have found: For some reason, I can't seem to remember, I searched the keywords "metal gear" in GameDev.net. And, this link came up http://www.gamedev.net/community/forums/topic.asp?topic_id=397196 I could not believe my eyes that something like what I wanted do, can be done more professionally, moreover, more effeciently! PLUS, more easily!!! Therefore, I began researching MODS and this link came up http://www.gamedev.net/community/forums/topic.asp?topic_id=154144 which is exactly what I want to do! I want to focus on AI ONLY! In conclusion, My questions are: 1- Is it true? Can I develop a Metal Gear-ish game (Camera, cut-scenes unimportant) using a game MOD while maintaining full control, or better yet, modify the existing AI algorithms to better suit my needs? And, eventually, truly focus on AI alone. 2- If the answer to question 1 is yes, then: Since, a fair amount of time in Gaming Years have passed since those posts above have made, which is the best choice as MOD for me? Is it Half-Life 2 MOD SDK. 3- If the answer is no, then: What type of AI algorithms are best for me? And which resources and keywords can be useful for me? 3- I already have the Russell and Norvig AI book, but haven't read it yet, because I was focused on OpenGL. Is the Russel Norvig book enough for me to begin developing the AI algorithms, or refining the existing ones at least? If not, which books are recommended for such tasks as mine? Final Note(s): 1- Money is no issue when it comes to buying books or Games, for liscening and development purposes. So please be more eager to recommend something you might think is best, even if it costs a little bit more. 2- I am squeezed on time...! I need to finish the whole project within less than TWO MONTHS! I already know it is a very short period of time, so please be kind by not reminding me : ) So I am willing to do everything needed, and I am very eager to learn!!! 3- Keywords are the most critical when doing research. So, please let me know of any useful, and accurate, keywords I might need to reach my goal. 4- Any and all pieces of information are appreciated!!! Thank you in advanced!!!
  4. Thank you for your kind reply. I could not benefit well from it, though... It is because of my rather primitive skill at both 3D graphics and vector mathematics, nor can I imagine what you say ( or others say ) well in the aspect of the program I am now working on. My 3D graphics teacher gave us an assignment. Like always, he wrote us a program and removed parts of the code, and asked us to write the missing parts so it will operate in the manner he wants. I personally dislike this method of teaching when the prerequisites of the course are very far from the course content. But here is not the place for such woes.
  5. Hi, I am very new at computer graphics. I have a practice, but I am really stuck, I read everywhere on the net, and in a book I have with me (real-time rendering by Moller) the past few days. But couldn't implement it properly : I have a moving sphere that I choose it's direction and speed. It bounces inside a box. And in this small application I can insert static balls ( balls that don't move when they get hit by that moving ball ). As I understand Collision detection is done in three steps : 1- Detecting whether or not a collision has happened ( will happen ). ( already been done ) 2- Finding the collision intersection between the colliding objects. ( here is where I am stuck now ) 3- Deciding the proper response to that collision. ( haven't reached this part yet ) In the application only one ball is moving, all the other ball(s) ( if any ) do not move. So, finding the intersection I think should be straight forward. Subtracting the center coordinates of the first ball from the second ball and dividing it by two, then adding the result onto the first ball. It doesn't matter which whether the moving ball is the first or the second, no? Doing that gives us the collision point. Then we must find the normal of that point, I know it is VERY simple, but I must admit that I can't figure it out... --- As for the third part, I will leave it later, when I understand the second, and implement it correctly. Your kind help is mostly appreciated! zgzg2020
  6. Hello, Does "SDL_KEYDOWN" mean the transition from unpushed -- to --> pushed? Or does it mean the state of the key being DOWN "de-pressed"? If it would be said in computer logic, then is it the transition from 1 to 0? Or is it the steady state of it being at 1? I believe it is the first. Because if it was the second possibility then SDL_KEYUP would invoke all unpressed keys to be true all the time, no? --------------- I have another question that is relating to this one. I am currently writing a Tic Tac Toe game. And I implementing the controls. I wrote it in two different ways. currentPosition[ 0 ] is the x coordinate on the Tic Tac Toe board. --- Type one --- if( event.type == SDL_KEYDOWN ) { switch( event.key.keysym.sym ) { case SDLK_LEFT: if( currentPosition[ 0 ] > 0 ) { currentPosition[ 0 ]--; } break; . . . --- End --- --- Type two --- if( event.type == SDL_KEYDOWN ) { switch( event.key.keysym.sym ) { case SDLK_LEFT: xDisposition -= 1; break; . . . if( event.type == SDL_KEYUP ) { switch( event.key.keysym.sym ) { case SDLK_LEFT: xDisposition += 1; break; . . . move_courser( xDisposition, yDisposition ); void move_courser( int xShift, int yShift ) { currentPosition[ 0 ] += xShift; currentPosition[ 1 ] += yShift; if( ( currentPosition[ 0 ] < 0 ) | ( currentPosition[ 0 ] > 2 ) ) { currentPosition[ 0 ] -= xShift; } } --- End --- I left out the y movement for simplicity The first type yields the better result. The result there is when a key is pressed (even kept de-pressed) the shift is always one cell at a time. The second type yields an unusable result for a Tic Tac Toe control. When a key is pressed (de-pressed) it goes VERY fast over the board, and only stops when the key becomes unpressed. I read in lazy foo's tutorials that event handling and logic must be separated. The first type I wrote by my own way. But then I felt that I didn't separate the logic from the event handling so I had to change it. I went reading through his tutorials again, and I rewrote it in this way trying to imitate his method. Could someone kindly explain why are the results differ in these two types of code? Regards
  7. Movement in SDL

    Besides your current problem, I really recommend visiting Lazy Foo's SDL tutorials<\a>. There SDL general game concepts are explained in a fairly practical way. I am a beginner, it helped me understand general game ideas via SDL. Good luck!