The structure of the engine itself might vary considerably - consider the situation of the PS3 vs the PC.
With the PS3, you know every user has around 6 free cores, each of which is somewhat less powerful than one core of a C2D for certain kinds of data. So if you want a high end product that performs well, you parallelize. Massively.
Now consider the PC, where you can't even guarantee the user's got 2 cores, and even if you do, they might have 2-4, you've got a division of labor problem. You could spawn tons of threads, but cross-thread communication has its gotchas on every platform as well.
Consider also that any optimizations you do for data storage and speed of loading could suffer from endianness problems on various platforms, so you need to document very carefully every single time you use optimizations like that (or might be using one) or have some code which assumes endianness in information coming from some middleware.
You've got storage constraints - the XBox 360 is limited to a couple of DVDs whereas you've got the vast storage of Blu-ray on the PS3 and the unlimited patience for multi-disc installs for PC and Mac users. Also if you want multiplayer, you've got to worry about the differences in PSN, XBL, and whatever platform you decide to go for on PC and Mac.
Consider texture size restraints, shader characteristics, differences in hardware across PC which make it MUCH harder to target the PC or Mac as a platform (even Macs vary widely in the capabilities of their video hardware) and you're getting into nightmare territory
Adding Linux into the mix just complicates matters - binary distributions of closed-source software (due to licensing restrictions resulting from) relying on binaries for closed-source middleware that you don't even have the source for.... Eesh.
The software side of things in the games industry is very much not an afterthought, but the average multimillion dollar title is fast approaching the complexity of what entire operating systems were ten years ago - and these guys put these out on a 1-2 year development cycle. Your hearsay may be right about how shoddy things are, but it's not from lack of consideration but a simple mismatch between time constraints and the complexity of the product.
Don't forget Compuserve who was the granddaddy of AOL and modern ISPs and the first to bring nationwide dial-up home computer network access to American families.
I actually pulled the first open source program I ever used from a friend's dad's Compuserve after reading about it in a catalog listing from one of those generic BBS file collection CDs they used to sell.
Then you pair it with low draw systems and a large capacitor.
That's just their test rig anyway - the model. TFA States:
The recent tests examined technologies that would see a nuclear reactor coupled with a Stirling engine capable of producing 40 kilowatts of energy--enough to power a future lunar or Mars outpost.
If the crease did not destroy image detail (a creased Polaroid instant picture often gets nondestructive creases) this could remove warping and glare problems.
Contrast control, avoiding light pollution in the sensors when scanning an undersized object, lamp longevity, improved support for scanning coarse film grain photographs, glare reduction on shiny objects...
Honestly, we're looking at a serious clash of the titans here. Microsoft vs The State of Texas would be one hell of an interesting lawsuit.
A company with a profit margin greater than the GDP of most of the world's nations vs the state in the US with the second highest GSP would make the most interesting trial since OJ was trying on gloves.
The right question to ask is: Can YOU make money building your business on proprietary software that might not be there tomorrow?
Yes.
...And people have been doing it for 30 years and still are with 30-year-old software.
The burden of keeping the software running falls on the IT departments, but with the exception of hardware-bound eccentric UNIX flavored software, that's not much of an onus.
Our department uses (extremely domain-specific) MS-DOS accounting software from 1983 running under Dosbox. I recompiled Dosbox with a tweaked version of the Windows SDL library so that it didn't require that annoying stdout window which SDL apps seem to require by default and confuses the hell out of the secretary, set up a dosbox.conf for the program, and created a shortcut which launches the program by autoexec. Now this 26 year old software runs in a window that looks no different than the terminal window she's used to using to connect to our campus mainframes - which themselves are running an ancient version of IBM-made UNIX.
I don't see how that makes it difficult to make money, and the cost for this software was written off 20+ years ago - hardly a bad investment. We've talked to the company that made this software perhaps 6 times in those 20+ years, and each time was to resolve some minor irritation that could have been fixed with some amount of effort on our part, but we paid for support because it was cheaper. If the company had gone tits up a decade ago, we'd have lost perhaps a hundred and fifty dollars worth of time reentering some information to start over instead of paying seventy-five dollars for support. Not that significant a loss.
The certainty that the moment a company goes out of business, all of their closed-source software stops working is ludicrous. Even should the dongle for some antique software stop working, the DMCA (in the US) specifically provides for circumventing that copy protection should the company be unable to provide a working replacement. Other countries have even less draconian laws governing reverse engineering of software. That's not to say that it wouldn't be much easier (and cheaper) to use a piece of software that doesn't have copy protection and provides its source, but to delude yourself into believing that it's impossible to make money with outdated closed-source software is to ignore the fact that the vast majority of the business and governmental world is already doing it.
I agree, it should - but it has much less to do with the HTML5 canvas and much more to do with how the Javascript was written.
Don't throw the baby out with the bathwater - if the demo's pushing that kind of throughput, then imagine how a well-behaved sleep/setTimeOut based approach will perform on the same canvas element.
Considering Blizzard is trying to turn Battle.net into a platform, one which Starcraft II and Diablo III are built entirely upon, I can't imagine Diablo III is going to come out before they finish their Battle.net "upgrades."... They'd also be crazy to compete with themselves, so I think it's entirely likely we won't see Diablo III before 2011. Seems a shame to push back a game that's almost done by 2 years just so you don't have to write LAN support.
The whole thing reeks of a pay-to-play future for Battle.net.
I don't think Adobe's too worried about losing Flash. They've been working on browser-based Javascript alternatives for quite some time and overall, since their acquisition of Macromedia, they've been improving standards compliance in Dreamweaver to a significant enough degree that you can see vast improvements in the quality of the code it produces between versions.
Although their interfaces are overengineered and clunky and their DRM continues getting more and more draconian, the actual output of their software continues to be better for their consumers. They produce some solid XHTML by default with a lot less of the bloat than Macromedia's initial versions were drifting towards.
No doubt they'll be in a position come CS5 or CS6 to be producing these HTML5 canvas documents with Flash itself and focus on filling in gaps with the Flash plugin which aren't currently covered with standards - such as local file access, UDP sockets, etc.
Not to kiss the ass of Adobe, but they're on top of the game because they belong there - they know how to compete and stay relevant.
An additional though related to that - a really solid browser-based layout engine would relieve some of the pressure for nonprofits and small papers to continue clinging to products like Quark and InDesign or being too reliant on Apple and Microsoft products. If I was Google, I would already be offering to hire that gentleman for just such a venture.
Now that is impressive. Not only impressive, but exciting for those of us who have the unenviable task of trying to merge web CMS with print production.
Embedded EPS display of ad proofs would be just the beginning... Merge that with the H&J algorithms from *TeX and you've got the potential to actually do document layout in the browser - and the browser could print the resultant page to PDF and report the element properties via AJAX to a database backend.
When you're trying to demo a new technology in a flashy way, it helps to go with relevant. The zeitgeist is all in favor of microblogging and Twitter, as valueless as it seems, is attracting positive attention to the social aspects of the internet, which means the right kind of people (journalists) are looking in its direction. Hence, you produce a demo which uses Twitter and has shiny graphics and light lounge music and you'll have mainstream hipster journalists drooling and pointing. Which is good for everybody who has an interest in seeing HTML5 take off quickly.
Look, this is very simple. This uses one full core (more or less, 60% on a C2D T7800 in Chrome 3.0) because it's running in realtime. Just read the source (although it's a bit condensed.)
It begins rendering the next frame as soon as the previous one has ended then scales the movements based on the framerate. This means that even though what you're seeing may look like about 30fps, it may be running in the hundreds. The apparent framerate will only be as high as the resolution on your screen as the smallest unit of apparent movement is going to be the pixel without some sort of subpixel rendering.
If you find that raster graphics running at a couple hundred frames per second on a variably-sized canvas without use of a polled timer is ridiculous is "absolutely ridiculous" I invite you to learn something about graphics programming and try to do better.
Why would you advertise Internet Explorer anywhere? It's a broken piece of shit, and if you were able to run Internet Explorer, you'd already have it as it's pre-installed.
Actually, as per the language quoted in TFS, the concept of the notification page is infringing as well. You (the administrator) are transmitting a message to a user communication device (the web server) with grouping information (directories), user contact information (their login info) and response data (a return email address).
Having browsed their methodology section, they took the 150 most popular games across all systems, selected the highest selling titles for the nine major game systems (PC, Playstations, Xboxes, Gamecube and the portables) at the time (March 2005-February 2006) and then compared them against the US population... Not the respective populations of the countries which produced them.
Ignoring that the top ten is mostly american football and soccer titles, the rest were Star Wars titles, games with no discernable main character (GT4), Pokemon variants and GTA: San Andreas.
Each game was weighted
by the number of copies that it sold, meaning that a game selling 4 million
copies figured twice as heavily in the computations as a game selling 2 million
copies.
This methodology virtually ensures that they'll get figures with almost the exact relative composition of professional sports averaged against southern California, as the sports games and Grand Theft Auto: San Andreas outsold nearly everything else by a factor of 5-10.
Looking at the rest of the huge grossers, we get some more Japanese RPGs (with largely white-ish asian-ish compositions, it's hard to tell with JRPGs), World of Warcraft, and The Sims 2 and all of its variants, all of which skew the results severely as the characters are either created by the people playing them or they are of indeterminate or irrelevant race (Why are the Orcish people of the USA so underrepresented?)
In short, it's a decent idea for a study, but horribly executed. If they really want to do this for North America, they need to choose their sample better. Or generalize it to studying the markets that produced the titles, then stating how close the diversity of each market's games are to the population of that market.
AI cant make a conscious decision that is not preprogrammed.
Definitely. The job of the AI designer is to come up with a set of default behaviors and reactions which make the AI appear to be doing so.
You may not be able to make an AI figure out intent, but you can train them to recognize erratic motion - players in a pure deathmatch game don't often stop or double back quickly without an obvious reason, so something like that could trigger the bot to go into "cautious mode" and fire, say, a grenade to the entrance of that corridor then try to circle around. About 90% of the time it'd look like the bot was paranoid, but the few times it worked, the victim would be completely convinced.
I agree that we're a long way from being able to solve the problem of actual rational AI. I think we first have to figure out how logical frameworks and learning work before we can begin tackling making computers think and reason like people.
Fortunately, it's often a lot easier to make them LOOK like they're thinking and reasoning like people.
In fact, feeding bogus data to the AI is one of the realistic ways to limit, say, a racing game's agents - if they don't see the post in front of them because they aren't spending enough time per frame watching the road and are instead eyeballing their opponent, they're going to crash, just like any human. So you simulate that by using player proximity and the "erraticness" of the other opponents to model distraction and modulate the AI's awareness of dynamic obstacles and hazards.
Our illogical behavior is largely deterministic as well.
We tend to behave illogically only in response to specific stimuli (fear, anger, hunger, lust) or when our system is under strain (fatigue, extreme hunger or thirst, neurological stress), nearly all of which can be simulated effectively enough for a game simulation.
So now we examine the character of our illogical behavior - we prioritize actions inappropriately, mistake one input for another of a similar kind, suffer from reduced reflexes or recognition time, respond with an inappropriate reaction to a familiar stimulus, fail to suppress responses which we would ordinarily not allow in ourselves due to social strictures or personal beliefs, or simply fail to notice things in an appropriate timeframe.
What about that couldn't be simulated with an extremely simple system of defaults with a laundry list of pre-programmed failure behaviors? Illogic may be more complicated to simulate in a limited domain of actions than logic - the elevator can go up, down, or stop, but it's hard to make it change its mind when it's tired of going up - but illogic is really easy to simulate when the expected domain of AI activity includes nearly any action. This is the sort of condition found in most sandbox games - you expect the pedestrians and enemies to behave in an almost random fashion because you expect humans "in the wild" to be unpredictable. This means that anything short of obviously programmatical behavior or obvious illogical *group* behaviors will seem fairly realistic to the player - especially if the AI isn't just "instanced" appearing and disappearing with a short library of functions, but instead is programmed with agendas, no matter how simple (go from residential A to commercial B, append grocery bag model to arms, use carrying walk animation, return to residential A).
The AI in Ultima 7 was praised as being exceedingly lifelike because the AI had agendas, day/night schedules, and would respond to stimuli like violence, the appearance of a monster, etc in a variety of ways depending on their character role. This sort of realism (if not actually passing whatever Turing-test-like metric you employ when observing it) will serve to satisfy the requirements of suspension of disbelief on the part of the player.
One of the best things about video games is the potential to surprise the player with unexpected behaviors. The first Quake bots, even though following fairly simple nodegraphs, would continually surprise players by behaving in a fashion seen as "unpredictable" simply because the player themselves had not been taking the most efficient routes between "pickups."
The first "learning" Unreal bots would actually remember routes that bots saw players take and append nodes and traversal instructions so that it could follow or use that route in the future. As a result, you think you can evade a bot by leaping out a window they never go near then are alarmed to find that not only does it follow you out, it uses the same route to escape you in the next round.
The emergent behaviors of The Sims have been pleasing and surprising gamers for nearly 10 years now, all based on fairly simple wants/needs systems along with some basic stimulus response. The Sim is less intelligent than the average cockroach, and yet they are still capable of behaviors which seem satisfyingly realistic, at least in the short term. If a Sim is too tired to make a full meal, it might just grab a bag of snacks from the refrigerator. They might fall asleep on the couch instead of going to bed. These are all failure states in an ideal AI's daily routine, and yet they give the human touch - with very little computational cost.
The point here is that AI doesn't need to be perfectly human to be humanlike, and it's far from impossible to simulate illogical behavior - you just have to program some chaos into the system by which the AI selects actions.
Slashdot Mirror
The structure of the engine itself might vary considerably - consider the situation of the PS3 vs the PC.
With the PS3, you know every user has around 6 free cores, each of which is somewhat less powerful than one core of a C2D for certain kinds of data. So if you want a high end product that performs well, you parallelize. Massively.
Now consider the PC, where you can't even guarantee the user's got 2 cores, and even if you do, they might have 2-4, you've got a division of labor problem. You could spawn tons of threads, but cross-thread communication has its gotchas on every platform as well.
Consider also that any optimizations you do for data storage and speed of loading could suffer from endianness problems on various platforms, so you need to document very carefully every single time you use optimizations like that (or might be using one) or have some code which assumes endianness in information coming from some middleware.
You've got storage constraints - the XBox 360 is limited to a couple of DVDs whereas you've got the vast storage of Blu-ray on the PS3 and the unlimited patience for multi-disc installs for PC and Mac users. Also if you want multiplayer, you've got to worry about the differences in PSN, XBL, and whatever platform you decide to go for on PC and Mac.
Consider texture size restraints, shader characteristics, differences in hardware across PC which make it MUCH harder to target the PC or Mac as a platform (even Macs vary widely in the capabilities of their video hardware) and you're getting into nightmare territory
Adding Linux into the mix just complicates matters - binary distributions of closed-source software (due to licensing restrictions resulting from) relying on binaries for closed-source middleware that you don't even have the source for.... Eesh.
The software side of things in the games industry is very much not an afterthought, but the average multimillion dollar title is fast approaching the complexity of what entire operating systems were ten years ago - and these guys put these out on a 1-2 year development cycle. Your hearsay may be right about how shoddy things are, but it's not from lack of consideration but a simple mismatch between time constraints and the complexity of the product.
Don't forget Compuserve who was the granddaddy of AOL and modern ISPs and the first to bring nationwide dial-up home computer network access to American families.
I actually pulled the first open source program I ever used from a friend's dad's Compuserve after reading about it in a catalog listing from one of those generic BBS file collection CDs they used to sell.
Then you pair it with low draw systems and a large capacitor.
That's just their test rig anyway - the model. TFA States:
The recent tests examined technologies that would see a nuclear reactor coupled with a Stirling engine capable of producing 40 kilowatts of energy--enough to power a future lunar or Mars outpost.
That oughta be enough.
I'll pass that on to all twelve of them so that they can nod sagely and approve of you.
And I believe that the star of the next 10 years will come to be the MS Singularity.
Right, that's the point at which Microsoft becomes so attractive that ordinary laws no longer govern it?
If the crease did not destroy image detail (a creased Polaroid instant picture often gets nondestructive creases) this could remove warping and glare problems.
Contrast control, avoiding light pollution in the sensors when scanning an undersized object, lamp longevity, improved support for scanning coarse film grain photographs, glare reduction on shiny objects...
I could think of a few.
Well, anthrax is anaerobic...
Honestly, we're looking at a serious clash of the titans here. Microsoft vs The State of Texas would be one hell of an interesting lawsuit. A company with a profit margin greater than the GDP of most of the world's nations vs the state in the US with the second highest GSP would make the most interesting trial since OJ was trying on gloves.
The right question to ask is: Can YOU make money building your business on proprietary software that might not be there tomorrow?
Yes.
...And people have been doing it for 30 years and still are with 30-year-old software.
The burden of keeping the software running falls on the IT departments, but with the exception of hardware-bound eccentric UNIX flavored software, that's not much of an onus.
Our department uses (extremely domain-specific) MS-DOS accounting software from 1983 running under Dosbox. I recompiled Dosbox with a tweaked version of the Windows SDL library so that it didn't require that annoying stdout window which SDL apps seem to require by default and confuses the hell out of the secretary, set up a dosbox.conf for the program, and created a shortcut which launches the program by autoexec. Now this 26 year old software runs in a window that looks no different than the terminal window she's used to using to connect to our campus mainframes - which themselves are running an ancient version of IBM-made UNIX.
I don't see how that makes it difficult to make money, and the cost for this software was written off 20+ years ago - hardly a bad investment. We've talked to the company that made this software perhaps 6 times in those 20+ years, and each time was to resolve some minor irritation that could have been fixed with some amount of effort on our part, but we paid for support because it was cheaper. If the company had gone tits up a decade ago, we'd have lost perhaps a hundred and fifty dollars worth of time reentering some information to start over instead of paying seventy-five dollars for support. Not that significant a loss.
The certainty that the moment a company goes out of business, all of their closed-source software stops working is ludicrous. Even should the dongle for some antique software stop working, the DMCA (in the US) specifically provides for circumventing that copy protection should the company be unable to provide a working replacement. Other countries have even less draconian laws governing reverse engineering of software. That's not to say that it wouldn't be much easier (and cheaper) to use a piece of software that doesn't have copy protection and provides its source, but to delude yourself into believing that it's impossible to make money with outdated closed-source software is to ignore the fact that the vast majority of the business and governmental world is already doing it.
I agree, it should - but it has much less to do with the HTML5 canvas and much more to do with how the Javascript was written.
Don't throw the baby out with the bathwater - if the demo's pushing that kind of throughput, then imagine how a well-behaved sleep/setTimeOut based approach will perform on the same canvas element.
No doubt.
Considering Blizzard is trying to turn Battle.net into a platform, one which Starcraft II and Diablo III are built entirely upon, I can't imagine Diablo III is going to come out before they finish their Battle.net "upgrades." ... They'd also be crazy to compete with themselves, so I think it's entirely likely we won't see Diablo III before 2011. Seems a shame to push back a game that's almost done by 2 years just so you don't have to write LAN support.
The whole thing reeks of a pay-to-play future for Battle.net.
Hopefully they decide to restrict it to just their news websites...
Among Newscorp's other holdings is "Fox Interactive Media" which runs IGN, Gamespy, Myspace (no huge loss there), Rotten Tomatoes and Hulu.
I don't think Adobe's too worried about losing Flash. They've been working on browser-based Javascript alternatives for quite some time and overall, since their acquisition of Macromedia, they've been improving standards compliance in Dreamweaver to a significant enough degree that you can see vast improvements in the quality of the code it produces between versions.
Although their interfaces are overengineered and clunky and their DRM continues getting more and more draconian, the actual output of their software continues to be better for their consumers. They produce some solid XHTML by default with a lot less of the bloat than Macromedia's initial versions were drifting towards.
No doubt they'll be in a position come CS5 or CS6 to be producing these HTML5 canvas documents with Flash itself and focus on filling in gaps with the Flash plugin which aren't currently covered with standards - such as local file access, UDP sockets, etc.
Not to kiss the ass of Adobe, but they're on top of the game because they belong there - they know how to compete and stay relevant.
An additional though related to that - a really solid browser-based layout engine would relieve some of the pressure for nonprofits and small papers to continue clinging to products like Quark and InDesign or being too reliant on Apple and Microsoft products. If I was Google, I would already be offering to hire that gentleman for just such a venture.
Now that is impressive. Not only impressive, but exciting for those of us who have the unenviable task of trying to merge web CMS with print production.
Embedded EPS display of ad proofs would be just the beginning... Merge that with the H&J algorithms from *TeX and you've got the potential to actually do document layout in the browser - and the browser could print the resultant page to PDF and report the element properties via AJAX to a database backend.
When you're trying to demo a new technology in a flashy way, it helps to go with relevant. The zeitgeist is all in favor of microblogging and Twitter, as valueless as it seems, is attracting positive attention to the social aspects of the internet, which means the right kind of people (journalists) are looking in its direction. Hence, you produce a demo which uses Twitter and has shiny graphics and light lounge music and you'll have mainstream hipster journalists drooling and pointing. Which is good for everybody who has an interest in seeing HTML5 take off quickly.
Look, this is very simple. This uses one full core (more or less, 60% on a C2D T7800 in Chrome 3.0) because it's running in realtime. Just read the source (although it's a bit condensed.)
It begins rendering the next frame as soon as the previous one has ended then scales the movements based on the framerate. This means that even though what you're seeing may look like about 30fps, it may be running in the hundreds. The apparent framerate will only be as high as the resolution on your screen as the smallest unit of apparent movement is going to be the pixel without some sort of subpixel rendering.
If you find that raster graphics running at a couple hundred frames per second on a variably-sized canvas without use of a polled timer is ridiculous is "absolutely ridiculous" I invite you to learn something about graphics programming and try to do better.
Good day.
Why would you advertise Internet Explorer anywhere? It's a broken piece of shit, and if you were able to run Internet Explorer, you'd already have it as it's pre-installed.
Actually, as per the language quoted in TFS, the concept of the notification page is infringing as well. You (the administrator) are transmitting a message to a user communication device (the web server) with grouping information (directories), user contact information (their login info) and response data (a return email address).
The whole thing is so vague...
It might run "Pitfall!" though.
Having browsed their methodology section, they took the 150 most popular games across all systems, selected the highest selling titles for the nine major game systems (PC, Playstations, Xboxes, Gamecube and the portables) at the time (March 2005-February 2006) and then compared them against the US population... Not the respective populations of the countries which produced them.
Ignoring that the top ten is mostly american football and soccer titles, the rest were Star Wars titles, games with no discernable main character (GT4), Pokemon variants and GTA: San Andreas.
Each game was weighted by the number of copies that it sold, meaning that a game selling 4 million copies figured twice as heavily in the computations as a game selling 2 million copies.
This methodology virtually ensures that they'll get figures with almost the exact relative composition of professional sports averaged against southern California, as the sports games and Grand Theft Auto: San Andreas outsold nearly everything else by a factor of 5-10.
Looking at the rest of the huge grossers, we get some more Japanese RPGs (with largely white-ish asian-ish compositions, it's hard to tell with JRPGs), World of Warcraft, and The Sims 2 and all of its variants, all of which skew the results severely as the characters are either created by the people playing them or they are of indeterminate or irrelevant race (Why are the Orcish people of the USA so underrepresented?)
In short, it's a decent idea for a study, but horribly executed. If they really want to do this for North America, they need to choose their sample better. Or generalize it to studying the markets that produced the titles, then stating how close the diversity of each market's games are to the population of that market.
AI cant make a conscious decision that is not preprogrammed.
Definitely. The job of the AI designer is to come up with a set of default behaviors and reactions which make the AI appear to be doing so.
You may not be able to make an AI figure out intent, but you can train them to recognize erratic motion - players in a pure deathmatch game don't often stop or double back quickly without an obvious reason, so something like that could trigger the bot to go into "cautious mode" and fire, say, a grenade to the entrance of that corridor then try to circle around. About 90% of the time it'd look like the bot was paranoid, but the few times it worked, the victim would be completely convinced.
I agree that we're a long way from being able to solve the problem of actual rational AI. I think we first have to figure out how logical frameworks and learning work before we can begin tackling making computers think and reason like people.
Fortunately, it's often a lot easier to make them LOOK like they're thinking and reasoning like people.
Indeed.
In fact, feeding bogus data to the AI is one of the realistic ways to limit, say, a racing game's agents - if they don't see the post in front of them because they aren't spending enough time per frame watching the road and are instead eyeballing their opponent, they're going to crash, just like any human. So you simulate that by using player proximity and the "erraticness" of the other opponents to model distraction and modulate the AI's awareness of dynamic obstacles and hazards.
Our illogical behavior is largely deterministic as well.
We tend to behave illogically only in response to specific stimuli (fear, anger, hunger, lust) or when our system is under strain (fatigue, extreme hunger or thirst, neurological stress), nearly all of which can be simulated effectively enough for a game simulation.
So now we examine the character of our illogical behavior - we prioritize actions inappropriately, mistake one input for another of a similar kind, suffer from reduced reflexes or recognition time, respond with an inappropriate reaction to a familiar stimulus, fail to suppress responses which we would ordinarily not allow in ourselves due to social strictures or personal beliefs, or simply fail to notice things in an appropriate timeframe.
What about that couldn't be simulated with an extremely simple system of defaults with a laundry list of pre-programmed failure behaviors? Illogic may be more complicated to simulate in a limited domain of actions than logic - the elevator can go up, down, or stop, but it's hard to make it change its mind when it's tired of going up - but illogic is really easy to simulate when the expected domain of AI activity includes nearly any action. This is the sort of condition found in most sandbox games - you expect the pedestrians and enemies to behave in an almost random fashion because you expect humans "in the wild" to be unpredictable. This means that anything short of obviously programmatical behavior or obvious illogical *group* behaviors will seem fairly realistic to the player - especially if the AI isn't just "instanced" appearing and disappearing with a short library of functions, but instead is programmed with agendas, no matter how simple (go from residential A to commercial B, append grocery bag model to arms, use carrying walk animation, return to residential A).
The AI in Ultima 7 was praised as being exceedingly lifelike because the AI had agendas, day/night schedules, and would respond to stimuli like violence, the appearance of a monster, etc in a variety of ways depending on their character role. This sort of realism (if not actually passing whatever Turing-test-like metric you employ when observing it) will serve to satisfy the requirements of suspension of disbelief on the part of the player.
One of the best things about video games is the potential to surprise the player with unexpected behaviors. The first Quake bots, even though following fairly simple nodegraphs, would continually surprise players by behaving in a fashion seen as "unpredictable" simply because the player themselves had not been taking the most efficient routes between "pickups."
The first "learning" Unreal bots would actually remember routes that bots saw players take and append nodes and traversal instructions so that it could follow or use that route in the future. As a result, you think you can evade a bot by leaping out a window they never go near then are alarmed to find that not only does it follow you out, it uses the same route to escape you in the next round.
The emergent behaviors of The Sims have been pleasing and surprising gamers for nearly 10 years now, all based on fairly simple wants/needs systems along with some basic stimulus response. The Sim is less intelligent than the average cockroach, and yet they are still capable of behaviors which seem satisfyingly realistic, at least in the short term. If a Sim is too tired to make a full meal, it might just grab a bag of snacks from the refrigerator. They might fall asleep on the couch instead of going to bed. These are all failure states in an ideal AI's daily routine, and yet they give the human touch - with very little computational cost.
The point here is that AI doesn't need to be perfectly human to be humanlike, and it's far from impossible to simulate illogical behavior - you just have to program some chaos into the system by which the AI selects actions.