Chess has very well defined rules, and there is always a limit on things you can do. It would also be fairly straightforward to map out every single possible game of chess.
Interactions between humans, such as asking a question and getting an answer, however, do not have as clearly defined rules, and the number of possible “moves” is much, much, greater.
Computer games are created with the art of programming and it takes just as long to learn and master painting as it does programming.
After a life time obsession with programming I can honestly say that being able to program a game with a language such as C, C++ or Java blows “short cuts” such as Games Factory or DarkBasic out of the water. Not to say they weren’t fun to use because they were, but the more time and effort you put into learning a proper computer language and the maths involved in games programming, the more you will see the beauty behind the art of programming and less need for such shortcuts.
The languages themselves allow you to kind of write a “book” which coupled with Maths allows that book to come to amazing life. APIs and game creator packages come and go so quickly, but knowledge of a language such as C++ or Java lasts for so much longer, as does your knowledge of how to put them to good use.
So if one wants to make games I strongly recommend they focus on learning programming.
Playing chess according to a set of rules is not at all the same process as understanding what another person is describing and then creating it for them. I don’t know how you could even think to compare the two.
Get as close as you can with coding, and we will all tell you how to fix it,
this is better than a computer doing it for you, as learning is half the fun.
Lateral thinking is involved in this approached. Sooner or later reinstatement solutions will come to the rescue. Master the basics, the option of shared resources between man and computer will be the magical mix.
I think a search engine that can understand what you draw could be used to do this.It would be able to match shapes to ideas.I would have people using the search engine draw pictures and then have other people give their interpretation.Then the person that made the drawing would type in what he was trying to convey with the picture.And then it would be stored in a data base, so that it can be retrieved when those pattern of shapes appear again.They have already made ai systems that can understand things by shapes.
To my knowledge there are no computer programs that can understand anything currently, unless your working definition of “understand” is so broad as to be useless. A search engine doesn’t understand what you type into it, it just looks for close matches to what you typed. So unless you’re planning on making a database of every possible thing a human could creatively make up, I don’t see how this could work without real artificial intelligence.
I mean pattern recognition.
A pattern is no shape and a shape no image.
And neither of those has a formalized context.
Example:
A user draws a triangle and flags it “conifer”
Next user draws a triangle and flags it “Pyramid”
Next user draws a triangle and flags it “Triangular door”
Next user draws a triangle and flags it “Triforce”
All a machine is able to find out is that there is a group of similar integral images, maybe even recognize the shape, with different meanings.
So, like I said, you need a predefined formalized language with grammar, syntax and semantics.
And that’s what modern programming languages are.
You try to invent something new here, without knowing the current state of the art, tech and research, nor any theory behind it.
People already thought and philosophized about this almost 150 years ago without even having computers and it really took off in the 1960s.
What if i made it recognise patterns,textures and also patterns within patterns.Because a pyrimid would be a triagular pattern and it would also have a pattern within that.Plus a conifer would also be like that.You understand.You would have a subcetagory within a category.Like triangular objects is the main category and the different names of the objects is the subcategory.
You can join the other thousands of computers scientists who have been trying to make that dream a reality for the last 60 years or so. Good luck. Let us know when you’ve invented a super smart computer brain that can spit out a video game after a human just describes their idea to it. Until then, creating things takes effort. Deal with it.
for me, that’s because game-making is more complex. Give the best chance to make a game, so the player won’t be boring to play it
Computers are stupid. Period.
Computers are dwarfed by the human brain in computational power (human brain = hundreds of billions of parallel processors).
Even if computers had equal power, they still wouldn’t understand humans, because computers aren’t humans. Computers with such brain power would be an alien species.
Chess, by the way, is an extraordinary stupid game. Games like soccer or basketball are much more complex.
You are mixing:
formal languages
pattern recognition
pattern matching
shape grammar
shape recognition
behavioral logic
learning software
and many more.
Those are individual fields of study, were individual people wrote and write master thesis about it and whole departments at universities research in.
One of the biggest issues is that a computer can see, match and differ very well, but not comprehend.
And when they’re able to combine it to something, it’ll surely not be someone funding a game engine.
Sadly enough it’ll most likely be of militaristic nature for autonomous fighting machines.
But that’s ethics in computer sciences and a whole different topic to discuss.
A bad comparison.
Chess isn’t stupid but a finite system. You can play all possible games with all possible outcomes.
That’s exactly what I meant with “stupid”. Not only is chess a finite system, the number of choices at every move (ply) is very limited (a few hundred, at most). Therefore, it can be faked with brute-force searches (and a good opening book).
Name one game that isn’t stupid by that definition?
I believe I mentioned soccer and basketball. Let’s consider computer-controlled robots with more or less human specifications (in terms of strength, speed, eyesight, and means of communication). I don’t think we will live to see a team of such robots defeat a human team.
Getting more on-topic with the original post, I think something a little like the Playstation 3 game ‘Little Big Planet’ might be of interest to you.
It has a very, very robust and expansive “create mode” that lets you create levels in an intuitive manner. There could be something similar for the PC, although I won’t profess to be all that much of a gamer, so I wouldn’t know.
Making a game is an art form, though. Like many forms of art, it take time to master and to carry out. A game engine where users can just ‘type in’ what they want, and magically a game is made for them wouldn’t be all that useful to anybody. The games made with it would be more-or-less the same, and you would gain next to no knowledge from making them.
If you want to play a game with a particular aspect, just search Google. Chances are someone out there has already made it.
Some people have had people develop games for them.So it would be useful.It is just like fast food except
you don’t have to pay for it.And they probably said that there was no market for fast food.Do you think 3d fax machines are useful?
I suggest you research computational theory and then run the math, the complexity of the issue you want solved is beyond our current scope of technology
Anyhow, we have computers that can generate content based on images from real life, perfect replicas of environments yet I don’t see people preferring that over the same environment drawn by an actual person, or to put it in simpler terms:
People like content generated by other people, not computer generated content this is because the content created by humans has slight imperfections giving it personality, something the computer generated content does
Besides, I don’t think you understand game development for your idea to work someone would a) have to write a game engine that can handle every theoretically possible scenario and all the theoretically impossible ones as well since theory only takes it that far, already there we have an infinitely long workload (because there is an infinite possible scenarios and variations of said scenarios and the engine would have to handle them all)
But let’s invoke the Infinite Monkey Theorem, now we have the engine, that’s cool now we have to write an AI that can interpret the text the user inputs, no matter the language (we can’t be discriminating non-English speakers here) and we must also make sure it can decipher the text no matter what is inputted, this means it must handle text like l0l, 1//\46!||3 @ 6@3|/| //3RZ J00 R P07@70 &&& đ1!3Z P1||Z!11!one! or I wnt gm wr shtng nd xplsns wif prty bebs nd gnt rbts., and then we must also make sure said AI can read their minds as to make sure that they do exactly what was originally intended in the first place, otherwise the customer will call it crap because it didn’t make the intricate game they imagined from their pathetically vague one sentence description
Oh and that is still not taking into account that it would have to be able to generate all dialog, all art and all sound on the fly as it creates the game, or perhaps we could invoke the infinite monkey theorem again and thus generate the entire infinite art, audio and text library required for such a game.
Now a more realistic scenario: Learn math, Learn programming, Grow the fuck up and stop whining, creating things is hard, that’s what makes it worth doing.
P.S You are one of the groups of people I can’t stand, mainly the group that thinks game development is oh so easy. You disgust me and I want to punch you, in the face, really really hard.