Re: [glob2-devel] Full Nicowar behaviour assessment

[Bradley Arsenault]

On 3/4/06, Kieran P <address@hidden> wrote:
>
> Before you continue, please keep in mind that I have no knowledge of c++ so
> I havn't got a clue wether something is possible or near impossible.
>
>
> > > For Big Pond, it did very well. It won. No suprise as its a bigger map
> than
> > > the others. It used more spacing than Garden 3 did.
> > >
> > Heh heh, I do 99% of my testing on Big Pond so its no wonder it plays best
> here.
>
>
> Lol. Hmm.... a quick thought. A map could be 512x512 and nicowar could loose
> if it didn't have space right? So we should probably stop refering to
> nicowars success based on the size of the map but the space available for
> expansion? Your call.
>
>
> > Sometimes this is exactly the difficulty. Nicowar builds its buildings
> > following fairly strict rules. It always leaves enough space to
> > upgrade the building + 1 square away from other buildings (not
> > ressources). So yeah, castor places Inns in pairs, like it does other
> > buildings when it has the space. I have seen it place barracks,
> > racetrack, and swimming pools in pairs). I have been thinking about
> > sperating spacing into a building specific attribute, so that a
> > building can be placed really close to an Inn, but has to be more
> > spaced for central buildings like the Swarm.
> >
> > The when you can't find enough room part is iffy to implement. This is
> > exactly why I praise the Castor programmers like I do, because they
> > are quite good at doing "threshhold" based algorithms like this, where
> > as I tend to avoid threshholds all together, as they come with a vay
> > of inaccuraccy.
>
>
> Well, a few things to say and expand on. Group things that dont need space
> to upgrade or can upgrade atleast once without space. For eg, in all my
> tests of Nicowar, the inns are pretty much the last thing to get upgraded.
> Infact, I have never seen a lv3 inn before. So build them in groupd of 4 (in
> a square shape) to save space. They can still be upgraded to lv2 for more
> room since only lv3 needs that extra room for upgrading. As far as
> hospitals, shocking. Scattered everywhere. Group them in 4 (just like the
> inns). From the test I did with Big Pond earlier, the grouped hospitals
> would have saved enough room combined to make 2 swimming pools. Barracks and
> schools dont need extra room for upgrades. Group them in pairs (schools
> right next to swarms preferably, so newly created members can jump stright
> in hopefully). Doing these things will create lots more room, but still keep
> things functional, accessable, and upgradable.


You don't know how Nicowar places its buildings, so I'll let you go on
this one. But I'll give you one thing: having only inns placed near
other inns and all other buildings spaced is impossible in the current
system, and would be an expensive algorithm regardless. I can fake it
and allow any buildings, including inns, to be placed close to inns,
and to have upgrade spac ing requirements dropped for inns, but no
building to building specifics for now, maybe later.

>
>
> > I don't need to copy castors placement code, my existing code works
> > just enough. The towers thing has been iffy in the past, it has almost
> > uselessly takin up workers when Nicowar could sufficiently defend with
> > warriors, including on maps like Garden 3. The change to have towers
> > constructed is incredibly simple, a couple of header file changes, but
> > requires much experiementation.
>
>
> Yes, towers can take away free workers. But since Nicowar isn't smart enough
> to make the war flag chase enemies and eleminating them from coming right
> back, or determin like entrace ways (see G2, wild river, stone bases, and
> the sand square) its the only other means of quick defence (and holding them
> off till warriors arive) if groups of castor invade.

Like I said, experiementation is needed.

>
>
> > Your suggestion is, then, to give more units to upgrading and less to
> > constructing, yes?
>
>
> Well, sort of. A fully upgraded city should be established before venturing
> out. Doing so will mean you have better warriros and workers as soon as you
> can (for protection and speed). They'll work more effeciantly and faster
> when the time to expand comes. If you expand right away, not only will the
> buildings be vunrable (because the workers at that stage wouldn't have
> learnt how to upgrade buildings to make them stronger) but the rest of the
> civilsation will be too. So yer, atleast till and upgraded barracks and
> school at minimum are completed, hold of on expansion.

So how do you propose I stop expansion, and how do you propose I have
it decide when to start expanding again. If it did 0 expandsion it
would never build another building, because to build any building it
has to "Expand" to some degree.

>
>
> > I don't control the retrive thing. If the workers can grab the fruit,
> > they will, so there is no control whatsover. The exploration flags are
> > put on the three nearest fruits of each type, which is a
> > reccomendation made from many people on this thread.
>
>
> Stick a forbidden zone on all fruit exploreres find and then they wont go to
> them :P
>
>
> > Converting is a very important aspect of the game, it doesn't cost
> > anything but the harvesting of fruits. I don't see conversion as a
> > negative aspect in *any* circumstances, considering that if it didn't
> > convert enemy units, it would go on happilly. If you do convert enemy
> > units, but they starve to death, you haven't lost anything, you still
> > have all those swarm made workers you would have if you had done no
> > conversion, but your enemy has lost something, his workers have just
> > converted to his opponents team and died, so no losses here or there.
>
>
> Well, I always though that to convert units, you need fruit. To get fruit,
> you need workers. To get workers, you have to pull them away from other jobs
> that are more important, like upgrading to get a small army (10-15 globs)
> for defence. Stop converting, and your globs wont have to work so hard.

Ahh, this is where fault takes place. Stop collecting fruit, and your
globs won't work so hard. Fruit is collected automatically when
available. And the converting comes as a side effect, continue
converting, and you get more globs. Stop converting and send your
units to the swarm, you get more globs. The difference is that when
you convert, your enemy also loses a glob, its not just you gaining a
glob.

>
>
> And now for some new discussion.
>
> First a query. I have not seen the code to work out how many users to place
> working on a swarm/inn implemented yet. Swarm is still at 5 and inn at 1 :(
> I was wondering if you still going to implement it? Incase you forgot, for
> the swarm, 3 initial workers + .25 for every sqaure it is away from wheat,
> and for inns, same calculation but add 1 for every 20 total colony members.
> Next, and this is for very unique maps, but if Nicowar runs out of space on
> a very cramped island (such as newislands_green_blue), and it hasn't built a
> pool, it cant get off the island so it starts to die off when wheat becomes
> scarse and inns are depleted because of lack of room. For a human, this is
> easily prevented, but Nicowar hasn't been programed to do this yet.
> Solution: Either remove buildings to make way for a pool, or use a clearing
> flag (and remove any forbidden zones) specifly designed to make room for the
> pool (atm, if I manually place the clearing flag, seconds after nicowar
> places inns and swarms on the empty places when it should be a pool (and
> there was enough room for it)).

The Inn code is good already, it doesn't assign just 1 unit per Inn,
look more closely, it is more intelligent than that. The swarm code
still needs to be implemented, yes.

The pool thing takes onto a whole new catagory in building desire,
desire by design. This hasn't been implemented yet, as its fairly
complex. It would allow for such jests as "if the highest reachable
gradient from the starting point without traversal over water is x,
and our opponents center is not within this point, then increase
demand for a school to be constructed." The clearing area thing is
also more difficult, I would have to do a catagorization of the team
buildings within the gradient traversal, then group them according to
destance from eachother, and try to find the group that eleminates the
lowest possible buildings that still provides enough room for the high
demand building.


>
> The crash bug still exists. And I know you said you couldn't do anything
> about it unless you had a windows developer, but its really starting to
> annoy. And it only happens when running a game with Nicowar (or making a map
> with 32 players and trying to run it). I can't get gdb setup on windows
> unfortunatly, so if you can find and suggest a good debugger, I'd be happy
> to install it and give you the output. Most of the crashes involve something
> to do with user input. For eg, the last crash happened when I tried to
> remove a forbidden zone I'd placed because nicowar was getting itself killed
> (workers were within tower range).
>
> I notice Nicowar doesn't use markets. I'm wondering what they are actually
> for? Would they be usefull or worker waster?

No one should use markets. Markets are useless.

>
> I just played 7 Nicowar against me (set with nicowar team help), so 8
> nicowar in total. I got wiped out very fast. Its like they targeted me, not
> each other (even though they were on seperate teams). 5 teams had opened
> their inns and stole 35 units in less than 4-5 mins of starting the game.
> They also advanced a lot faster for some reason. I dont know how, but I had
> 3 inns when everyone else had four. I still had 3 inns when 2 teams had 6-7
> inns. Maybe its just me, but shouldn't they all stalemate? Their programming
> is the same :P
Of coarse they gang up on you, thats the normal thing. Try it with 7
castors, its the same. The "Custom Game" menu automatically allies all
AI against the human, its not Nicowars choice at all.
>
> Can the label "Nicowar (unstable)" be cut down to just "Nicowar" please? For
> several reasons. The alliances screen's checkboxes gets covered by the text
> because its too long. When selecting Nicowar in custom game, the text runs
> off the button. Also unstable makes it sound like it crashes all the time,
> which it doesn't. Just once or twice on my system.

Its not stable yet, so it should keep the warning that it is unstable.
Otherwise, we are giving people a reason to make formal complaints and
start disliking the game, but if we say its undstbale, people won't
use it unless they really want to, and won't care if it crashes.

>
> _______________________________________________
> glob2-devel mailing list
> address@hidden
> http://lists.nongnu.org/mailman/listinfo/glob2-devel
>
>
>