Re:Are Quests in MMOGs doable?
on
Quests
·
· Score: 1
I don't think the bottleneck is in the graphics. The bottleneck is in the fact that AI on such a grand scale is so difficult to implement and so burdensome on server hardware that it is exorbitantly expensive to implement.
Once server hardware is powerful enough to, e.g., implement a different genetically generated AI algorithm for each mob, be it townsperson or creature from the black lagoon, and progressively mutate these AIs overtime based on their respective aptitude (ie, monster that has AI that lets him live longer is more apt to procreate than monster that lives for 30 seconds and dies due to outside interference - ie other mobs)... well, once that's able to be accomplished, I think we'll start seeing more simulations done in world - rather than a static playground.
Jumpgate came first, then Vendetta (which was a clone of Jumpgate), and now ND is remaking JG into something that might (we can all hope) be fun to play instead of infinitely annoying.
VO and JGC both suffered from the same problem: a max population of about... 25 pilots.
The original had semi-frictionless space. No stopping unless you used thrusters in the opposite direction, or hit the 'break' thrusters which automatically slowed your velocity.
However, each ship had a limited top speed, boostable by after burners and other add-ons, that when completed would return your ship back to its original speed (not very cool imho).
The original Jumpgate did indeed have great Joystick controls, with real newtonian physics to play around with. It was fun blasting past someone on afterburners and spinning quick and firing at them as they try to turn around and get at you.
I'm less concerned about this as I am with NetDevil's poor quality of service with the original. They didn't have enough players to really have an interesting gameworld on its own, so they added really stupid-acting giant purple space fish (squids..eels..manta rays..snails..) with simplistic AI. And that's all there was to shoot, because they liked to ban people who actually tried to compete/enact large PvP dogfights.
Re:Non-Tech Percent of Web Traffic from Chrome
on
Google Chrome, Day 2
·
· Score: 1
What I want to know is why he linked to an (ugly) "review" that is just a recap of other reviews, and even (erroneously) calls the V8 JavaScript engine performance "Java results"... and is riddled with typos?
plus the Google Brand (and all the monitoring that goes with it
TFA:
Chrome has a privacy mode; Google says you can create an âoeincognitoâ window âoeand nothing that occurs in that window is ever logged on your computer.â The latest version of Internet Explorer calls this InPrivate. Googleâ(TM)s use-case for when you might want to use the âoeincognitoâ feature is e.g. to keep a surprise gift a secret. As far as Microsoftâ(TM)s InPrivate mode is concerned, people also speculated it was a âoeporn mode.â
Please, please, PLEASE buy a cultured diamond! Not only are they a higher quality, but they do not contribute to continuing degradation and violence in Africa (blood diamond wasn't exactly a myth).
Keep in mind that diamonds are actually a fairly common stone - and the reason they are so expensive is because of a tightly controlled supply.
I'm not that keen on the family-management aspects. I think it takes away from the adventuring draw of the game - a tedious punishment for players instead of something that drives the gameplay along.
Well, you're thinking of it in terms of evershit and world of waste.
The other part that holds this together:
In a day or two, you can get your character to 'average.' Enjoy 90% of the content of the game.
Each step you take past that has an increasing chance of killing your character. But equally large rewards for your descendants...
This accomplishes: a majority of the population is within an order of magnitude from each other. The powerful people - if they manage to stay alive - become famous.
If you look through my comment history, you might find some novel ideas on how to implement permadeath whilst keeping it "fun." Unfortunately, I'm not a subscriber and can't view my own full comment history. The gist of it:
1) why do players dislike permanent death? 1a) Because they lose hard work. 1b) Because their character might have been famous, and now they have to start all over again.
2) what can be done to rectify this? 2a) incorporate death somehow into the character development process, so it's a GOOD thing. Sometimes. 2b) their friends (and enemies) will still know who the player is behind the toon, because the bloodline's surname is the same.
Proposal:
You create not a character, but a bloodline. This bloodline has an abode (probably starting as a hovel with upgrade possibilities). Said bloodline can have several characters in it. Bloodline possessions are handed from one character to another.
The circumstances of a character's (permanent) death influence starting character attributes (if it were fallout, I'd call em perks). Say one of your characters dies fighting a dragon? Now you have the ability to create a character with the Dragonslayer perk (+ to damage against dragons, perhaps), to better avenge his forefather. This also can kick off dynamically generated quests.
Example: forefather that died fighting dragon had a pretty cool sword. A few months down the road, your current character gets wind of where his sword might be, and sets off on a quest to get it. The strength of his forefather's spirit/character imbued the sword with even more magic, etc.
Having a haven/home for your bloodline also allows you to more easily create different styles of characters and play subsequent characters easily.
Example: You build up your character to a coveted level of skill in swordmastery. Unfortunately, he dies. Your subsequent characters, if they so choose, can be built with a "Swordmaster Heritage" perk, for example, which makes getting up to a decent skill level in swordmastery a bit easier/faster.
See what I mean? Permanent death need not be an annoyance to be complained about. It can be integrated into the play workflow nicely.
It would be interesting if crafting in these games was changed from mundane drag item from backpack to anvil, click skill button until complete...
So that: for the casual players, they can gain fame and fortune by completing puzzles when crafting, supplicating and perhaps overtaking local NPC-driven economies; hardcore combat nerds can buy shit and go kill shit. Everybody's happy.
I really think you're thinking too narrowly on this, believing that all the finite states need to be preprogrammed/designed.
It would be more interesting if the developers took a lesson from machine learning, whereas:
simple rules combine to create complex behavior.
NPC a has wants: hunger, money.
NPC has skills: battle, crafting
NPCs whose generated AI algorithms do not complete its objectives (ie, dies of starvation, dies in battle), well, their AI dies with them. Successful NPCs completing an objective (find mate) replicate and use algorithms already defined by genetic algorithm systems to create a new AI for their offspring.
It's likely that if the rules and objectives are robust enough, and you have enough "craft" (not just swords and armor, but buildings and walls and such), and you let the system run itself for quite a while before inviting players in...
Well, I'm certain if the above is accomplished, you would end up with a complex and dynamic gameworld. No plotlines to write - the NPCs write them themselves.
Let's just hope they don't have bluetooth connections to each other - after a couple hundred of them huddle together, they can cast fifth level spells!!!
It's like in Cryptonomicon, but on a galactic scale!
Actually, this problem was interestingly addressed in A Fire on the Deep, by Vernor Vinge.
Cool book.
Both of 'em.
Yeah, but what self-respecting landlord would, upon a maintenance request for a leaky pipe under the kitchen sink, come in and: snoop through your financial documents, put on your wife's dress and dance around in it before putting it back, sniff your underwear, switch your toothpaste with your foot cream, and possibly - while they're at it - poke holes in all your condoms?
I know! And my point with this was they are often lumped in there by their teachers and fellow students: they are a genius with cars and how they work, but how often is knowing how to replace the alternator on an '87 buick tested in school? how often is instinctually knowing how the engine is put together in a new car that you haven't seen before tested? It's a great skill, but hardly ever recognized in formal schooling as such.
"Brute force all possible chains of action (grows exponentially)"
"Who dies first (have fun with all the level, item and skill combinations along with the problem of deciding who SHOULD win)?"
Ahhhhh.... thus comes the adage, "Keep it Simple, Stupid." Maybe if combinations of equipment/skillsets cannot be easily tested automatically (you've already defined the skills and the equipment, why can't you write an algorithm to test 'em against each other?)... you're thinking about the game mechanics all wrong. Maybe the game mechanics system sucks. There's a difference between a behemoth game mechanics system that attempts to predict everything, and all the subtle nuances and bugs that can introduce... and something simple and elegant that seems complex on the surface.
Slashdot Mirror
I don't think the bottleneck is in the graphics. The bottleneck is in the fact that AI on such a grand scale is so difficult to implement and so burdensome on server hardware that it is exorbitantly expensive to implement.
Once server hardware is powerful enough to, e.g., implement a different genetically generated AI algorithm for each mob, be it townsperson or creature from the black lagoon, and progressively mutate these AIs overtime based on their respective aptitude (ie, monster that has AI that lets him live longer is more apt to procreate than monster that lives for 30 seconds and dies due to outside interference - ie other mobs)... well, once that's able to be accomplished, I think we'll start seeing more simulations done in world - rather than a static playground.
Not only that, but they believe that if they learn these things before they're cleansed to a certain point, they die.
Jumpgate came first, then Vendetta (which was a clone of Jumpgate), and now ND is remaking JG into something that might (we can all hope) be fun to play instead of infinitely annoying.
VO and JGC both suffered from the same problem: a max population of about... 25 pilots.
The original had semi-frictionless space. No stopping unless you used thrusters in the opposite direction, or hit the 'break' thrusters which automatically slowed your velocity.
However, each ship had a limited top speed, boostable by after burners and other add-ons, that when completed would return your ship back to its original speed (not very cool imho).
And then ruined it...
Except Vendetta sucks =/
I'm pretty sure this is in 'noob mode.'
The original Jumpgate did indeed have great Joystick controls, with real newtonian physics to play around with. It was fun blasting past someone on afterburners and spinning quick and firing at them as they try to turn around and get at you.
I'm less concerned about this as I am with NetDevil's poor quality of service with the original. They didn't have enough players to really have an interesting gameworld on its own, so they added really stupid-acting giant purple space fish (squids..eels..manta rays..snails..) with simplistic AI. And that's all there was to shoot, because they liked to ban people who actually tried to compete/enact large PvP dogfights.
What I want to know is why he linked to an (ugly) "review" that is just a recap of other reviews, and even (erroneously) calls the V8 JavaScript engine performance "Java results" ... and is riddled with typos?
Um, taco..hello?
Bingo.
TFA:
To add my two cents..
Please, please, PLEASE buy a cultured diamond! Not only are they a higher quality, but they do not contribute to continuing degradation and violence in Africa (blood diamond wasn't exactly a myth).
Keep in mind that diamonds are actually a fairly common stone - and the reason they are so expensive is because of a tightly controlled supply.
Science: it works, bitches.
I'm not that keen on the family-management aspects. I think it takes away from the adventuring draw of the game - a tedious punishment for players instead of something that drives the gameplay along.
Well, you're thinking of it in terms of evershit and world of waste.
The other part that holds this together:
In a day or two, you can get your character to 'average.' Enjoy 90% of the content of the game.
Each step you take past that has an increasing chance of killing your character. But equally large rewards for your descendants...
This accomplishes: a majority of the population is within an order of magnitude from each other. The powerful people - if they manage to stay alive - become famous.
If you look through my comment history, you might find some novel ideas on how to implement permadeath whilst keeping it "fun." Unfortunately, I'm not a subscriber and can't view my own full comment history. The gist of it:
1) why do players dislike permanent death?
1a) Because they lose hard work.
1b) Because their character might have been famous, and now they have to start all over again.
2) what can be done to rectify this?
2a) incorporate death somehow into the character development process, so it's a GOOD thing. Sometimes.
2b) their friends (and enemies) will still know who the player is behind the toon, because the bloodline's surname is the same.
Proposal:
You create not a character, but a bloodline. This bloodline has an abode (probably starting as a hovel with upgrade possibilities). Said bloodline can have several characters in it. Bloodline possessions are handed from one character to another.
The circumstances of a character's (permanent) death influence starting character attributes (if it were fallout, I'd call em perks). Say one of your characters dies fighting a dragon? Now you have the ability to create a character with the Dragonslayer perk (+ to damage against dragons, perhaps), to better avenge his forefather. This also can kick off dynamically generated quests.
Example: forefather that died fighting dragon had a pretty cool sword. A few months down the road, your current character gets wind of where his sword might be, and sets off on a quest to get it. The strength of his forefather's spirit/character imbued the sword with even more magic, etc.
Having a haven/home for your bloodline also allows you to more easily create different styles of characters and play subsequent characters easily.
Example:
You build up your character to a coveted level of skill in swordmastery. Unfortunately, he dies. Your subsequent characters, if they so choose, can be built with a "Swordmaster Heritage" perk, for example, which makes getting up to a decent skill level in swordmastery a bit easier/faster.
See what I mean? Permanent death need not be an annoyance to be complained about. It can be integrated into the play workflow nicely.
It would be interesting if crafting in these games was changed from mundane drag item from backpack to anvil, click skill button until complete...
So that: for the casual players, they can gain fame and fortune by completing puzzles when crafting, supplicating and perhaps overtaking local NPC-driven economies; hardcore combat nerds can buy shit and go kill shit. Everybody's happy.
(of course, the huge hurdle here is having enough CPU horsepower to actually run all that AI concurrently)
I really think you're thinking too narrowly on this, believing that all the finite states need to be preprogrammed/designed.
It would be more interesting if the developers took a lesson from machine learning, whereas:
simple rules combine to create complex behavior.
NPC a has wants: hunger, money.
NPC has skills: battle, crafting
NPCs whose generated AI algorithms do not complete its objectives (ie, dies of starvation, dies in battle), well, their AI dies with them. Successful NPCs completing an objective (find mate) replicate and use algorithms already defined by genetic algorithm systems to create a new AI for their offspring.
It's likely that if the rules and objectives are robust enough, and you have enough "craft" (not just swords and armor, but buildings and walls and such), and you let the system run itself for quite a while before inviting players in...
Well, I'm certain if the above is accomplished, you would end up with a complex and dynamic gameworld. No plotlines to write - the NPCs write them themselves.
Let's just hope they don't have bluetooth connections to each other - after a couple hundred of them huddle together, they can cast fifth level spells!!!
It's like in Cryptonomicon, but on a galactic scale! Actually, this problem was interestingly addressed in A Fire on the Deep, by Vernor Vinge. Cool book. Both of 'em.
Yeah, but what self-respecting landlord would, upon a maintenance request for a leaky pipe under the kitchen sink, come in and: snoop through your financial documents, put on your wife's dress and dance around in it before putting it back, sniff your underwear, switch your toothpaste with your foot cream, and possibly - while they're at it - poke holes in all your condoms?
Not to mention it can interact with unit testing frameworks for java, c#, javascript, etc. Jawsome.
That's not accurate information, that's precise information! It's precisely inaccurate! :)
I know! And my point with this was they are often lumped in there by their teachers and fellow students: they are a genius with cars and how they work, but how often is knowing how to replace the alternator on an '87 buick tested in school? how often is instinctually knowing how the engine is put together in a new car that you haven't seen before tested? It's a great skill, but hardly ever recognized in formal schooling as such.
"Brute force all possible chains of action (grows exponentially)"
... you're thinking about the game mechanics all wrong. Maybe the game mechanics system sucks. There's a difference between a behemoth game mechanics system that attempts to predict everything, and all the subtle nuances and bugs that can introduce... and something simple and elegant that seems complex on the surface.
"Who dies first (have fun with all the level, item and skill combinations along with the problem of deciding who SHOULD win)?"
Ahhhhh.... thus comes the adage, "Keep it Simple, Stupid." Maybe if combinations of equipment/skillsets cannot be easily tested automatically (you've already defined the skills and the equipment, why can't you write an algorithm to test 'em against each other?)