headers
Andrew Sayers | 7 Sep 2005 18:40

Gameplay guidelines

What constitutes good and bad gameplay?  Obviously this is hugely
subjective, but I'd like to try and work out some guidelines for the way
we would like a game of Globulation 2 to play out, so we can start to
think more about what the game should be like.  I'll try and present
some general topics for discussion before I give my opinions on them,
but there are probably a lot of issues I've not thought about, so please
add ones you think of.

The two big issues that occur to me are competition vs. co-operation,
and thrills vs. skills.  There's also an issue of game length.

There can be good games anywhere along the competition/co-operation
spectrum.  For example, first-person shooters can be great, but tend to
be entirely competitive, while sim games can be great, but are totally
co-operative.  Ultimately, Globulation 2 is a competitive game in that
each team wins or loses on its own - alliances aren't considered.  On
the other hand, alliances can exist and make for much more interesting
gameplay.  It's also competitive in that the aim of the game is always
to beat the other team, rather than (say) to reach some part of the map.
Glob2's scripting language can handle some non-competitive goals, but
this is quite rudimentary and rarely used.

Thrills vs. skills is an area of gameplay that gets less attention.  Is
it better to have a game where the gameplay is more subtle and varied,
where every move has a thousand counters, and every counter a thousand
weaknesses, or a game where you build up a great army and smash your
unsuspecting enemy in one strike?

Finally, game length seems at first like a minor side-issue, but I think
it's a good way of getting at the way a game should work.  Obviously,
(Continue reading)

Permalink | Reply | 19 comments |
headers
Eli | 7 Sep 2005 22:06
Favicon

Re: Gameplay guidelines


On Sep 7, 2005, at 12:40 PM, Andrew Sayers wrote:

> What constitutes good and bad gameplay?  Obviously this is hugely
> subjective, but I'd like to try and work out some guidelines for the 
> way
> we would like a game of Globulation 2 to play out, so we can start to
> think more about what the game should be like.  I'll try and present
> some general topics for discussion before I give my opinions on them,
> but there are probably a lot of issues I've not thought about, so 
> please
> add ones you think of.

> The two big issues that occur to me are competition vs. co-operation,
> and thrills vs. skills.  There's also an issue of game length.
>
> There can be good games anywhere along the competition/co-operation
> spectrum.  For example, first-person shooters can be great, but tend to
> be entirely competitive, while sim games can be great, but are totally
> co-operative.  Ultimately, Globulation 2 is a competitive game in that
> each team wins or loses on its own - alliances aren't considered.  On
> the other hand, alliances can exist and make for much more interesting
> gameplay.  It's also competitive in that the aim of the game is always
> to beat the other team, rather than (say) to reach some part of the 
> map.
> Glob2's scripting language can handle some non-competitive goals, but
> this is quite rudimentary and rarely used.
>
> Thrills vs. skills is an area of gameplay that gets less attention.  Is
> it better to have a game where the gameplay is more subtle and varied,
(Continue reading)

Permalink | Reply | 18 comments |
headers
Bo Lorentsen | 8 Sep 2005 17:34
Picon

AIWarrush has small error i CVS

AIWarrush.cpp
AIWarrush.cpp: In member function 'Order* AIWarrush::maintain()':
AIWarrush.cpp:368: error: expected `;' before ')' token

As I can se it the source line sould look like this :

        if(numberOfBuildingsOfType(IntBuildingType::FOOD_BUILDING) * 18
                                <

numberOfUnitsWithSkillGreaterThanValue(WALK,0)+numberOfUnitsWithSkillGreaterThanValue(FLY,0))

Now some glob2 testing :-)

/BL
Permalink | Reply | 1 comment |
headers
Eli | 8 Sep 2005 21:21
Favicon

Re: AIWarrush has small error i CVS


On Sep 8, 2005, at 11:34 AM, Bo Lorentsen wrote:

> AIWarrush.cpp
> AIWarrush.cpp: In member function 'Order* AIWarrush::maintain()':
> AIWarrush.cpp:368: error: expected `;' before ')' token
>
> As I can se it the source line sould look like this :
>
>        if(numberOfBuildingsOfType(IntBuildingType::FOOD_BUILDING) * 18
>                                <
>                                 
> numberOfUnitsWithSkillGreaterThanValue(WALK,0)+numberOfUnitsWithSkillGr 
> eaterThanValue(FLY,0))
>
>
> Now some glob2 testing :-)
>
> /BL
>

Grr... It looks like I forgot to test that one change before I  
committed it. Fixed, and I will try to make sure to actually remember  
to test this stuff right before I commit it.
Permalink | Reply | 0 comments |
headers
Andrew Sayers | 9 Sep 2005 17:49

Re: Gameplay guidelines

> Well, good gameplay currently is setting up a settlement that works 
> effeciently. You can't win if it takes five minutes every time you want 
> to build a new building, and that is quite a plausible situation if 
> you're a bad city designer. I've had situations myself where my Inns 
> were in such locations that the wheat couldn't be reached practically 
> and so units would starve, even when I had twice as many inns as I'd 
> usually need for that population. I think I'll have to modify AIWarrush 
> to take this into account. (On the other hand, a particularly good 
> settlement can upgrade a school to level 3 in under a minute.)

I agree that enjoyable town-building is a big part of the fun of Glob2,
and I think it's something we've not paid enough attention to in the
past.  It's what you spend most of your time doing, so we should spend
at least as much time making that balanced as anything.

<snip>
> Cooperatively, there would nothing to fight, and despite the importance 
> of building, fighting is what actually drives the game. 'Backstabbing' 
> just seems completely pointless to me in this game (it's not selfish, 
> it's spiteful) and teams winning on their own just furthers the 
> 'tacked-on'-seeming nature of the alliance system. I think alliances 
> should be fixed, (i.e. you cannot make and break alliances.)

I agree that the team nature of games should be built up more, but
fixing alliances doesn't make any sense: if that's the way you want to
play, you can just have two players on the same team.

If Glob2 is supposed to be about experimenting with the value of teams,
we ought to give players the data to conduct their experiments.  I love
the statistics you get at the end of each game (even if I don't
(Continue reading)

Permalink | Reply | 17 comments |
headers
Stephane Magnenat | 10 Sep 2005 21:46

Re: Gameplay guidelines

Hi everyone,

I'm very happy about those discussions, because they make the game live. Let 
me add some hints:

* One of the core idea behind glob2 was to have more strategy by having a 
dynamic environment and much less micro-managment. I think this key concept 
is important. Bases building is critical and is clearly part of the fun.

* Ideally, and it is what we tried to achieve by the war/conversion/prestige 
concept, there is several strategy leading to victory and not only war. Of 
course, even without war the game is competitive. Nevertheless, with a better 
scripting there could be interesting player vs environment scenarios.

* A team has lost when it is mathematically not able to recover. So glob2 will 
make loose a team if it has units in an inn surrounded by wheat.

* Explorers behavours should be fixed in latest CVS, they should eat and heal 
as any other unit.

Thanks, have fun,

Steph

--

-- 
http://nct.ysagoon.com
Permalink | Reply | 16 comments |
headers
Eli | 11 Sep 2005 11:30
Favicon

Redesigning War Flags

What is the purpose of a War Flag? Conceptually, it is an order to 
"Attack this stuff." Therefore, it is micromanagement to designate any 
more settings than what stuff to attack, although it may be desirable 
to determine how strong the attack should be. Furthermore, an attack by 
a group of troops should not continue as an attack by a string of 
troops, because the player may want one or the other, and the other is 
doable by guard areas and is generally not too powerful.

Currently, you can group troops properly by moving a War Flag back and 
forth between here and there, but that is more micromanagement. 
Similiarly, it becomes necessary to move a war flag at its target to 
make the globs attack what you want them to. (On a side note, guard 
areas must be constantly updated for a proper attack as well. All glob2 
warfare, currently, is impossible to win without micromanagement.)

My new design is simpler to use and more complicated to implement. Idle 
warriors will gather in an Inn (or designated building) whenever the 
number of warriors at the flag is less than half the desired number. 
When their numbers reach the number allocated by the flag, they will 
march out towards it, regardless of whether that oversupplies the flag 
with warriors. Also, when they get there, they will patrol the area 
designated, they will patrol around it by going towards unvisited 
locations within its radius until the whole circle has been covered, 
and when this is done, they will ignore previously visited locations 
and repeat this movement. This last bit should go for Explore Flags as 
well.

Now it's up to the rest of you to find the horrible flaws in that 
system. ;)
(Continue reading)

Permalink | Reply | 5 comments |
headers
Stephane Magnenat | 11 Sep 2005 18:04

Usefull library

We could use this lib so that glob2 supports zip file insertion into vfs:
http://packages.debian.org/unstable/libs/libzipios++0

Steph

--

-- 
http://nct.ysagoon.com
Permalink | Reply | 0 comments |
headers
Andrew Sayers | 11 Sep 2005 18:17

Re: Redesigning War Flags

I think it's too early to be looking at the "ultimate perfect war flag".
Advancing one aspect of the game like this far ahead of the others only
leads to problems in the long-term.

For example, we've discussed on IRC about turning guard areas into idle
areas, where any glob can go when it has nothing to do.  I've also
suggested (and again, it's too early to be doing this) having some way
of automatically defining town boundaries and having globs patrol them.
I'd also like to see the alliance system beefed up so that globs are
will chase one team's attack once it's been routed, but leave another
team to its own devices.  I also suggested a long time ago the idea of
"danger gradients".  If that can be made to work, it would give us many
more possibilities for game development.

The other problem with this idea is that it means war flags work in a
radically different way to other types of flag.  Besides being hard to
implement, this means players have to spend longer learning how the game
works.

	- Andrew
Permalink | Reply | 3 comments |
headers
Eli | 11 Sep 2005 18:56
Favicon

Re: Redesigning War Flags


On Sep 11, 2005, at 12:17 PM, Andrew Sayers wrote:

> I think it's too early to be looking at the "ultimate perfect war 
> flag".
> Advancing one aspect of the game like this far ahead of the others only
> leads to problems in the long-term.
>
> For example, we've discussed on IRC about turning guard areas into idle
> areas, where any glob can go when it has nothing to do.  I've also
> suggested (and again, it's too early to be doing this) having some way
> of automatically defining town boundaries and having globs patrol them.
> I'd also like to see the alliance system beefed up so that globs are
> will chase one team's attack once it's been routed, but leave another
> team to its own devices.  I also suggested a long time ago the idea of
> "danger gradients".  If that can be made to work, it would give us many
> more possibilities for game development.
>
> The other problem with this idea is that it means war flags work in a
> radically different way to other types of flag.  Besides being hard to
> implement, this means players have to spend longer learning how the 
> game
> works.
>
> 	- Andrew
>

All points are correct. In that case, the simplest way is probably 
best: A war flag works like a clearing flag.
(Continue reading)

Permalink | Reply | 2 comments |

Gmane