News for hackers. Stuff that matters.
Re:Peter Molyneux: My Next Game a 'Significant Scientific Ac... by Tichy on Monday May 19, 2008 @05:42AM (#193950)
This morality thing was already done nicely in the early Ultimas (I think starting with IV, or V). I remember the one dungeon where you are attacked by a monster called "little children". Before you had been trained to never fight children because it immediately makes you lose your virtues.

I don't think Molyneux approach sounds very scientific, though. What is scientific about giving game figures an artificial idea of morals? Scientific for me would be to not make those inbuilt, but let them evolve and see why certain morals make for more successful societies than others.

Sorry to rant about this, but it is one of my pet peeves with games (like civilization). They claim to be simulations, but they are not - they are just projections of the game designers limited perspective and morals. Of course the same could be said for many books and movies, too. I don't like it there, either, but somehow I feel that with games it is another dimension. The pseudo-simulation somehow makes the opinions of the game creators seem more legitimate.

Re:Peter Molyneux: My Next Game a 'Significant Scientific Ac... by teamonkey on Monday May 19, 2008 @06:50AM (#193966)
"What is scientific about giving game figures an artificial idea of morals?"

I got the impression that the "next" game he was talking about, the scientific achievement that would be on the cover of Wired, would be the one after Fable 2.

Re:Peter Molyneux: My Next Game a 'Significant Scientific Ac... by Tichy on Monday May 19, 2008 @07:32AM (#193990)
True, but he didn't mention anything new, so I assume he is just taking AI to the next level or whatever...
Re:Peter Molyneux: My Next Game a 'Significant Scientific Ac... by schtog on Monday May 19, 2008 @09:39AM (#194047)
but how would that be done.

no matter how you do it there has to be some starting point, initiation. and whatever values you place there wont be neutral right? or can they? it seems that even if you make it very basic they will still include some interpretation of the world from the designers and thus be more likely to evolve ina certain direction.

to be a perfect simulation would need the game designer to be some sort of a god no?

Re:Peter Molyneux: My Next Game a 'Significant Scientific Ac... by Tichy on Monday May 19, 2008 @09:45AM (#194054)
Not saying it is easy, but as an example I could point out "Conway's Game Of Life". At least it doesn't claim to be anything else than what it is...

What about a simulation of pool? I guess one could argue that the physics can never be 100% correct, but in that case I wouldn't feel so bad about it, somehow.

I guess I feel that it would be more interesting to have emergent behaviour from simple rules than to hardwire the rules on a high level, but of course I can not give a clear cut definition...

Re:Peter Molyneux: My Next Game a 'Significant Scientific Ac... by schtog on Monday May 19, 2008 @10:53AM (#194090)
sure i see your point, i guess in the future there will be middleware.

like molyneux himself points out it is too expensive. and i mena people kind of keep reinventing the wheel(like in a lot of other areas in programming) so one could imagine people creating worlds both for science and computer games that have evolving populations of something and then game designers can take these when they are at a certain stage of development and build a gameworld on top of them.

uh, kinda :)

Re:Peter Molyneux: My Next Game a 'Significant Scientific Ac... by Tichy on Monday May 19, 2008 @11:23AM (#194123)
Definitely, the way there are tools like Poser today for creating animated 3d people, there might be behaviour design tools in the future.