دانلود کتاب AI Techniques for Game Programming

I know what you're thinking: Heck, a book published in the early 2000s with reviews in 2005 has to be pretty dated as we look at 2010 and beyond. Wrong! Sure, most AI programmers have moved past Win 32 with numerous new techniques in C++ and Java, and the author's subsequent book (Programming Game AI by Example Programming Game AI by Example) is outstanding in filling in details left out for beginning and intermediate programmers here. However, TGP has all the makings of a classic, and if you miss it, your AI library will have a glaring and lonely hole. Our aeronautic simulation group at xtmh dot com hires numerous fresh grads from quality practical schools like Full Sail as well as quality abstract schools like MIT. What we find is that students who have grounded themselves in Buckland's two books, then evolved by staying up to date with his refresing openness to communication and support, have an outstanding balance of higher math skills (like quadratic programming, tensors and vectors) and practical "biological" agent motivation wisdom. Let's take another example: assume you love game design, but your skill set is in writing, dialog and character development, not MIT-level tensor mapping. Let's also say you've read the wonderful ultimate guide to writing and design (The Ultimate Guide to Video Game Writing and Design [ULTIMATE GT VIDEO GAME WRI -OS]. You are a skillful character, situation and dialog writer and designer, and will likely find a welcome place in supercomputing, the game industry, entertainment, or even the rapidly integrating fields (thank you pixar) of film/TV/gaming. To be successful and at the top of your game, you'll need to communicate your characters' motives, flaws, quests, wins and losses (a lot like real life) to the AI geniuses who will execute your vision in code. BOTH of Dille's books are written at a basic enough level to help you start translating your characterizations into agent behaviors. The opposite also is true-- if you're more of a codie, you'll love opening your mind to thinking at a more agent motivational level, with both time tested and new models of behavior in varying situations. Let's be honest-- the games of 2020 and beyond are all about intelligent agent interaction. The oldest problem in gaming-- how to dumb down a smart hero character enough to make them need a quest, but at the same time not make them look like idiots (stop, I don't want to hear about amnesia), has a corrolary in real life: sure, we could have been made smart enough to know all of life by just hacking our own brain, but God and our own Higher Self User put this odd dichotomy of an unconscious brain able to do array processor and direct geometric tensor mapping-- basically matrix calculus of partial derivatives-- something no supercomputer can do yet (we still have to convert geometric matrices and tensors to numeric models for processing even in supercomputers)-- when doing as simple an act as crossing a street in traffic; with a "conscious" brain that sometimes has trouble with four function math! If you need to figure out the paradox of dumbing down a genius in your character development, and the dicotomy seems too far fetched-- simply check out your average friend, or look in the mirror! Is this a digression? Nope-- we're just hinting at how cool a combination you will be able to imagine if you combine your study of data structures and algorithms with the now well established multi agent AI techniques. A hint at the future: we're finding clients as sophisticated as NASA and Los Alamos looking for new models of AFFECTIVE programming right now. Meaning, characters that not only learn and think as they move through your sims, but also are motivated by that other squishy reality-- heart. So the critics who think Buckland is out of date or too simple-- map your tensor into another frame of reference-- that of bridging code with biological and emotional motivation-- and you'll see why the most interesting games, sims and even real life "models" of cellular and synaptic structure and function need BOTH high level math and fun and interesting situational behaviors for the best games, and lives.