Reddit reviews Artificial Intelligence for Games
We found 19 Reddit comments about Artificial Intelligence for Games. Here are the top ones, ranked by their Reddit score.
NewMint ConditionDispatch same day for order received before 12 noonGuaranteed packagingNo quibbles returns
You may want to narrow that down a bit, but okay, here are some highlights, with amazon links to help disambiguate.
http://www.amazon.com/Game-Programming-Gems-Adam-Lake/dp/1584507020/
www.amazon.com/Game-Engine-Architecture-Jason-Gregory/dp/1568814135/
http://www.amazon.com/Game-Engine-Design-Second-Edition/dp/0122290631/
http://www.amazon.com/Game-Engine-Architecture-Engineering-Applications/dp/012229064X/
http://www.amazon.com/Artificial-Intelligence-Games-Second-Edition/dp/0123747317/
http://www.amazon.com/AI-Game-Programming-Wisdom-CD-ROM/dp/1584500778/
http://www.amazon.com/Real-Time-Rendering-Third-Edition-Akenine-Moller/
http://www.amazon.com/Real-Time-Cameras-Mark-Haigh-Hutchinson/dp/0123116341/
http://www.amazon.com/Code-Complete-Practical-Handbook-Construction/dp/0735619670/
There are many others, but looking at my shelf these are the big ones that stick out for recommendations.
My first reaction was to run away from this question, but it's really not that tough in the end.
The classic textbook is "Artificial Intelligence A Modern Approach" by Russel and Norvig. That will give you the agent based philosophy of building AI agents.
There are lots of game-based tutorials for making AI agents. It gives the development a nice goal and they're fun to work with. A few random places to look:
http://www.raywenderlich.com/24824/introduction-to-ai-programming-for-games
http://aigamedev.com/
http://shop.oreilly.com/product/9780596005559.do
http://www.amazon.com/Artificial-Intelligence-Games-Ian-Millington/dp/0123747317
Once you have the basics of search, pattern detection, flocking, reinforcement learning, supervised vs unsupervised learning. Then I'd suggest you start looking into the more interesting bits. Right now the really cool terms you should search for are: deep learning, deep belief networks, convolutional neural networks, random forests, biologically inspired algorithms. There's more, but those will keep you occupied for plenty of time.
There's lots of pre-built AI packages you can dig through and find tutorials for:
http://en.wikipedia.org/wiki/List_of_artificial_intelligence_projects
Weka
Orange
PyBrain
OpenCV
You can also find AI based programming competitions which give you that "I'm better than thou" feeling as you hack together algorithms.
http://en.wikipedia.org/wiki/Competitions_and_prizes_in_artificial_intelligence
This is really cool stuff, but there's a lot to learn about. Dive in and try using tools. Sometimes you'll want to jump off of a bridge in frustration, but then it'll work and it'll be almost life changing at times (not the jumping off the bridge bit, but that would be too I guess).
Good luck and have fun out there.
Check out Programming Game AI by Example and Artificial Intelligence for Games. They are both decent game AI books, the former actually develops an AI for a soccer simulation in chapter four. There's also [Behavioral Mathematics for Game AI] (http://www.amazon.com/Behavioral-Mathematics-Game-Dave-Mark/dp/1584506849/), it expands on the concepts of utility and decision making which are only touched upon in the first two books. It's purely theoretical but very interesting.
here, go read this. it's a very widely used book on game AI.
SubLight
Build your ship, hire crew, and voyage across the stars... only there's no fancy warp drives, wormholes, or faster than light travel. You're tasked with managing your ship on the generations long trip between each star system, keeping them happy, healthy, and under control!
Play it in your browser!
Or download the Windows, OSX, or Linux versions!
This week was spent on a lot of behind the scenes stuff, namely the crew AI. I did a full write up of the new AI on IndieDB, full of fun graphs, but here's a brief excerpt:
>I knew what kind of actions I wanted my crew to do, from various hijinks like joy-riding the turbo-lift, to full out cannibalism when starving! I figured the AI I was looking for was similar to "The Sims", "SimCity", or "Roller Coaster Tycoon". After glancing through the glossary of Artificial Intelligence for Games, I found mentions of "The Sims", and after some investigation, got exactly what I was looking for, Goal-Oriented Behavior.
>In brief, the crew have a limited number of Goals they are trying to satisfy, through Actions made available by modules on the ship. For my crew, or "Simunauts" as I've called them, they have these goals:
>* Eat
>Each of these goals have an associated value that increases over time and through interacting with the ship. A hungry "Simunaut" may hit up the cafeteria module to lower their Eat goal at the cost of increasing their Urinate goal.
>Here's a fun list of some basic actions available to the crew and you can see how important these goals are over time from this graph.
New:
Known Issues:
What's coming next week:
Thanks again so much for taking a look at this weeks alpha build of SubLight! Can't wait to hear your feedback and leave some for others!
As always, you can follow us at these fine locations:
SubLight on IndieDB | Twitter @Spacew00t | LunraGames.com
Artificial Intelligence for Games is a great book. Additionally, the various volumes of the AI Game Programming Wisdom series have some great examples.
ahh! 404!
But here's another link while this one is gone (I believe this is the correct one)
For the basics of mathematics I usually suggest "Mathematics for Computer Graphics" by John Vince.
https://www.amazon.com/dp/1447173341
On physics, there are a number of specialised books. Nevertheless, if with "physics" you mean "movement", the third chapter of "Artificial Intelligence for Games" by Ian Millington is for you (beware, source code is provided in C++, not C#).
https://www.amazon.com/dp/0123747317
About shaders, there are a number of sources out there. From a didactical standpoint I like the following two blogs:
https://www.alanzucconi.com/
http://catlikecoding.com/
And another interesting source is also http://www.shaderslab.com/shaders.html
If you prefer a book, a good primer can be "2D Shader Development: Foundations" by Francisco Tufro. Not the absolute best, but may give you a nice jumpstart in a limited number of pages
https://www.amazon.com/dp/B07BGYF7F8 (also available from the author's website).
Happy hacking.
The Art of Game Design - Jesse Schell is very very good.
Game AI (Millington and Funge new edition iirc) is very very good.
Some non-game-design books that are very useful for those doing game design:
Scott McLoud: Making Comics (the other two in the series are good but the section on plot, characterization, and development in this one is great)
Donald Norman: The Design of Everyday Things. (How design works and how people interact with technology and...)
Christopher Alexander et al A Pattern Language: Towns, Buildings, and Construction (Thinking about scale and design elements and modularity and...)
Kevin Lynch: The Image of the City (How do urban spaces work - essential if your game is set in a city - how do people actually navigate)
Polti: The 36 Dramatic Situations (old, quirky, examines how there are really only a few human plots)
Matt Frederick: 101 Things I Learned in Architecture School (how to think about and execute simple art, improve your design sense, ...)
Procedural Generation in Game Design is really good. I would read chapter 26 after reading Chapter 1 through. The order of the chapter should have really been changed.
Artificial Intelligence for Games
https://www.amazon.co.uk/Artificial-Intelligence-Games-Ian-Millington/dp/0123747317
Easy python, easy to understand.
One other book that is really nice is Algorithms 4th edition if you know Java.
Not really programming books, however fighting fantasy books are really nice. Also, RPG books like D&D, Rolemaster and GURPs are good to study.
This book is also pretty good for all things game AI related.
Also: Artificial Intelligence for Games
I can see how AI is a really neat goal!
However, I would do small steps. AI is more of a Computer Science problem than a programming problem, so you'll need a different set of skills to make useful AI.
I am normally a big advocate of the trial-and-error approach, but I don't think it's quite as feasible for learning AI/Machine Learning.
My advice:
This might seem daunting to you, but I don't mean to demoralize you. It's quite a nice journey of the mind, so to say.
So yeah, best of luck with your plans!
This book isn't bad:
http://www.amazon.com/Artificial-Intelligence-Games-Second-Edition/dp/0123747317
And here is another, I've never read it though:
http://shop.oreilly.com/product/9780596005559.do
You'll find heaps of Game AI info here:
http://aigamedev.com/
I really enjoyed that book too, but I would say it's basically only about two subjects: state machines and path-finding. It covers them really well but you might want to get another broader book too. Maybe after reading this one.
If you want a broader view - I found this book to be pretty great: http://www.amazon.com/Artificial-Intelligence-Games-Ian-Millington/dp/0123747317/ref=pd_sim_b_4?ie=UTF8&refRID=09B2HPTBM3NA2CNG477Y
> However, I would do small steps. AI is more of a Computer Science problem than a programming problem, so you'll need a different set of skills to make useful AI.
That is probably true and I think I will definitely heed your advice.
>Go on with basic Computer Science. Trying to accomplish something in AI without the basics ready to hand is like digging a hole with a spoon.
Is...... computer science an actually course or do you mean beginner friendly programming languages like python?
> https://www.udacity.com/course/cs101 [...] https://www.udacity.com/course/ud120 [...] http://www.amazon.com/Artificial-Intelligence-Games-Ian-Millington/dp/0123747317/
Thank you so much!
When I was still working in C++ (also my beginner language for some years), before I switched to Java, I spent some time working through this book.
Granted, you're going down a difficult route by starting with C++, but it's doable after a couple of hard months of practice and coding projects. Work your way up to implementing algorithms by reading the requirements (and none of the code) and then you'll be good to try this book out.
As a former hobbyst, honestly, it's a lot of fun and very rewarding to work towards this goal :)
They're not free but you might check out these two books.
Artificial Intelligence for Games