World: r3wp

And maybe I should finally solve the level 45 :/

ok... so you're the author... sorry, I mixed you up with Volker... 
he was trying to understand what I was talking about.

hehe

I always wanted to create a RTS game where each god starts off with 
nothing  really ... nothing... all he has are a few basic physic 
rules to work with, a selection of pictures to use for this or that 
(supply your own if you want  :-)  and basically, mix and purchase 
direct statistics and give them names... so that one person might 
create a civilisation which is centered on a religion called lets 
say... "Flying spaghetti monster"  an another which goes with long 
rang ballistics... really it would be a lot of fun, since you'd have 
NO bearing on your opponent's strategies, and since each one really 
is only purchassing raw statistics which are weighted full using 
sampling and probabilistic curves and such... the game would be utterly 
fair.

yet  when you are attacked by a "mighty whoopy cushion warrior" you 
might not know in the begining what they can do  :-)  so you actually 
have to learn to cope with the other's raw tactical might.

add an AI to this and you can get pretty cool long term solo playing 
with an infinite range of playability.

I guess a few oddball capabilities might be to purchase some game 
effects like motion speed, range, terrain type for your world, for 
which being need to have a specific means to cross... like noode 
forests and acid baths  :-)

obviously, the in game beings would be controled by scripts you supply 
as macros which can use inputs to determine their immediate surroundings... 
I guess purchasing evolution would simply be to allow more and more 
actually inputs... and will allow  you to create better agents, since 
they will be able to take more optimised decisions.

things like visible opponent count, then visible damaged opponents, 
visible meaner than myself opponents, etc.

or uknown terrain... as in, if you encounter unknown terrain... stop 
moving... it might be dangerous... and then  if you encounter terrain, 
report back if you seen someone move through it, so it become registered 
as known safe terrain, etc.

since game play is about improving your agents, and their statistics, 
this is done slowly and true evolution occurs.  you are free to create 
forks at any point too... but having many agents to update will obviously 
mean you loose time, but have better specialised agents... hehe all 
part of the strategy.

I must say, that the original Rebol version of the re-box is from 
Dockimbel (http://softinnov.org/rebol/rebox.shtml) I just took his 
level data and made a little bit more complicated graphics. And the 
avatar and some of the ferniture used is from one of my favorite 
Amiga games - DreamWeb - http://www.adventureclassicgaming.com/index.php/site/reviews/221/

http://en.wikipedia.org/wiki/Dreamweb

Anyway, why somoone would like to do remakes of an old games? We 
have emulators and the old games are available for download without 
problems now, like from this place: http://eager.back2roots.org/A-Z.HTML
If someone want to do a game in Rebol, it should be a new game which 
will use all the features of the network - it should be multiplayer 
online game - imagine Altme, but with avatars going around:-) I'm 
just waiting for R3 where I will be able to use custom fonts, better 
sound and modules  (for example). The problem I see is, that most 
of us here are coders, not graphic artists:(

altme with walking avatars, did i mentioned my *BONG spamm blocker 
activated*

old games because  of gameplay, which is hard. With network now of 
 course :)

a 3d planet defender with multiplayer attackers and defenders would 
be a total rush!

imagine having to take care not to crash or hit your allies... which 
are scrolling madly, just like you around the planet.

always looking out..

(obviously)

I mean the camera's POV should alway be pointing outward of the planet.

that works in 2d too. although the 3d-engine from the contest may 
 be able to do it.

that game with statistics, do you have an idea about implementation? 
something  for a small prototype? maybe with 1 or 2 d-physics for 
a start.

I haven't had a lot to time to put on the development side of that 
concept.  so I never really tried to go into such detail.  I have 
to much to do these days, but as part of the time I put on "fun" 
coding, I eventually will get to it... especially since most of my 
low-level apis are comming to fruition  :-)

each avatar is somehow liquid? ;)

how'd you guess  :-D

;-)

imagine that with liquid, you could actually share an in game avatar's 
data right on your desktop and plug that within a little app so you 
can let your game play until the avatar has something strange to 
report...    :-)

Yep. Only problem is, whatcounts as strange in the world you described?^^

something which is a result of your agent's processing identified 
as strange... basically, his inputs serve as senses or nerves and 
their processing create reactions.  the term "strange" can be relativised 
within your agent's logic but is not a term within the application 
itself.

it is you the god which defines a meaning to the term "strange" by 
causing some state or event to occur based on your agent's perception.

you could always call it "hocus pocus"  if you like that instead 
 ;-)

But with all that flexible changing rulesthat will be hard ;)

this actually means we can play both sides in our own language  and 
have to learn what the other's mean.

obviously within each successive game session or "world" there needs 
to be consistency, so that your agents can aquire a sense of meaning 
base on how they perceive your agents.

yes... this is like programming the matrix.

everything is open, and the world is only as real as you make it.

but its doable.  every game out there is an implementation of this 
concept... with preset rules, preset AI, and loaded graphics.  basically, 
all that is needed (like in the matrix) is to define the construct 
(laws of physics and rule set which can be exposed and perceived 
by the agents).

When changes have cost, that  consistency wouldcome. you take 50 
rules  from the last  game and can only afford 1  a day.

you can even tailor-in rule bending like magic, miracels or science.

scientific gods? Are you mad? ;)

or each victory allows you a point count (like xp) where you can 
add it more agent macro code bytes  :-)

I even tought of allowing individual agenst to be editable.  and 
then allowsing classes to be copied... obviously... cloning from 
a class is less costly when generating a new agent...

but since we are purchasing statistics in the end... the game should 
even out  :-)

so more tactical people will create highly specialised units.  average 
people will create average all-rounders... and savages will just 
create hordes of dumb agents which try to overul the terrain, and 
add in any archetype...

even things like morale can be tailored in, so that your agents automatically 
regroup or fallback.  maybe sending runners to fetch another regiment.

all part of your tactical might  :-)

How do the statisticswork? deciding how often you stumble?

now that is the concept.  implementation is obviously a question 
of many micro-evolutions of the engine's  capabilities... each little 
step adding a multitude of options.

for combat this is easy...   % of chance to hit... and number of 
hp damage, with probability of higher damage costing more.  so a 
one in six damage costs less than a 4 in 6 damage.