Re: Thoughts on game balance

[Dave Fancella <david.fancella@comcast.net>]

On Monday 09 February 2004 05:11 pm, Jens Granseuer wrote:
> I'll first set a few points that I think are important when discussing
> unit balance.
>
> I think that each unit should be able to destroy each other unit.
> The obvious exception to this are Interceptors, Artillery and the like
> which simply cannot target an entire class of units. But it should never
> happen that a weak unit (e.g. Infantry) doesn't stand a chance of even
> damaging a stronger one (e.g. Heavy Tanks). CF had such a combat
> system in the first releases which was the main reason it got replaced
> later on. This also means you have to be especially careful when
> defining the weakest and the strongest unit. (Ideally this would be
> taken care of by the combat resolution algorithm. I'm still not really
> satisfied with how the current one works but overall it's ok.)

Agreed.  One thing that irritates me is that AIs usually attack with pieces 
against which the defender can't defend.  It is, however, a fact of modern 
warfare that most units can defend against most other units.  :)

> Then there's experience. If you have a very strong level 1 unit
> (being able to wipe out other units 1-on-1 in one or two turns),
> you get real monsters in later levels. This was the case, for
> example, in Battle Isle where a single artillery, surrounded by
> a ring of cannon fodder, could easily win or lose a map. (Admittedly,
> this could be fun if the artillery happened to be on your end...)
> In CF I did not want such units.

This is something I've only started experiencing (no pun intended), or rather, 
noticing.  :)  It doesn't look like CF moves pieces from one field to the 
next, though.

> Finally, we have the AI. Making the game rules more complicated
> usually makes the AI at least twice as complicated. This is somewhat
> unfortunate, but also provides a very good reason for keeping combat
> rules simple.

Personally, not that I'm volunteering to program the AI (although I might take 
a crack at it sooner or later, since I'm planning on trying to use parts of 
the AI to automate parts of the game for human players), but I tend to think 
that the AI is something that shouldn't define the rules.  That would be the 
tail wagging the dog.  Nevertheless, the games that are always the most fun 
have the simplest rules.  :)

> That said, I tend to agree on the aircraft being too weak,
> possibly with the exception of the interceptors, but I'm not so
> sure about the long-range weapons. Sure, they are very weak on
> the lower levels, but IMO become quite a bit stronger later on
> where they _can_ wipe out weaker units in two turns.
>
> > On defense, both artillery and air units seem to be stronger than they
> > should.
>
> Well, the reasoning for the artillery units was that since they
> are so slow and can't even defend themselves when attacked, they
> at least need some protection so they are not obliterated by the
> first gust of wind that happens to come along.

Actually, if you were after some sort of realism in this case, it is a fact 
that artillery get obliterated by the first gust of wind that comes along.  
:)  I recall a certain Swedish king in the 15th Century had a habit of 
capturing his opponent's artillery and then pointing it back at his opponent.  
That said, I agree that artillery should be able to defend themselves when 
attacked, but defense strength and attack strength are entirely different 
matters.

> > And the gunners themselves should probably
> > have hand weapons, but if they used them in defense it seems like they
> > should lose the turn's attack because they're too busy fighting defense. 
> > Maybe we don't want that kind of detail, though.
>
> The artillery units already have the disadvantage that they can either
> shoot or move, but not both. I think that's bad enough.
>
> > In any case, I figure that any
> > unit capable of firing at air units are going to have something portable
> > and powerful enough to take them out with one or two hits.  We're dealing
> > with highly advanced, right?
>
> Not necessarily true. E.g. I don't think Infantry should have an effective
> weapon against aircraft but they also shouldn't be completely defenseless.
> Likewise, quite a lot of units only have minimal effectivity against
> aircraft. This may be personal preference, though.

Hm, it may be a personal preference.  I don't know about the regular Army 
infantry, but I do know that US Marines, at least, carry some pretty advanced 
AA weapons.  Of course, the Marines aren't "infantry" in the traditional 
sense, and their mission is totally different than the regular Army's.  It 
might be interesting to have a Marine piece, though, that has abilities that 
regular infantry don't have.

> When talking about the number of turns it takes to detroy an enemy unit
> it makes a vast difference if you fight a 1-on-1 or use a second unit
> for support. Depending on the units the latter can easily double or
> triple the number of hits.

I just learned about support, actually, and have been using it fairly 
effectively (I won the Island Hoppers map using it, not that that map was 
particularly hard already).  But don't forget two things.  :)  First, when 
talking about the 'number of turns', that's just a rule of thumb.  As a 
tactician, you have to estimate how many turns it takes for a unit to destroy 
another unit, assuming a reasonable rate of success.  The other thing is that 
the support method that's there actually gives an even greater advantage to 
the army that has the most numbers.  I would tend to think that greater 
numbers is already an advantage in itself.  But I don't yet have an opinion 
on this because I only just learned about it.  ;)

> > Further, it
> > shouldn't even be possible for ground units to surround an air unit and
> > prevent it from escaping.  As it stands, if your gunship gets surrounded,
> > he's probably toast.  Air superiority is big in modern warfare, but is
> > mostly nullified in Crimson because of this particular trait of air
> > units.
>
> This is one of the reasons aircraft are maybe a bit harder to shoot down
> than they should. That air units can also be trapped is probably an engine
> deficiency if you look at it from the 'realistic' point of view. However,
> due to the superiour action radius of aircraft this is often the only
> way to get them at all (if you don't have aircraft yourself). Besides,
> it also keeps the engine simple ;-) So I think this is fine as it is.
>
> > What do y'all think about a second movement phase after the firing phase
> > is resolved?  I ask because that's how Ogre/GEV did it, you got a second
> > movement phase for GEVs (Ground Effect Vehicle, aka hovercraft).  That
> > made your GEVs very useful, if light, strategic pieces.
>
> In Battle Isle 2 (or was it 3, or both?) this was used as well, mainly
> for recon squads. I also thought about adding this ability but it
> doesn't work very well with combat resolution at the end of the turn
> as is used in CF.

Actually, I seem to recall that Ogre/GEV had a turn sequence identical to what 
Crimson has, except for the second movement phase.  First you move all of 
your pieces, then you resolve combat, then you get your second movement.  You 
also didn't have to choose targets as part of the movement phase, but I think 
Crimson's interface for this is the best possible interface.  ;)  It would 
work well, or at least it has in board games.  I'll see if I can throw 
together a hack that does it so you can check it out.  I feel particularly 
passionate about this one, because I'm a hit and run type of tactician, and 
there aren't any opportunities for them that I've found, yet.

> > I realize that in Crimson
> > the hovercraft can also carry troops, but there isn't a piece capable of
> > 'hit and run' tactics.  In a battle against superior numbers, which most
> > of them are, hit and run tactics are essential to being able to win. 
> > With them missing, well, maybe I'm just a poor tactician after all.  ;)
>
> We made sure that all currently distributed maps can be beaten with
> the current set of units and tactics available in CF (TM). ;-)

Just out of curiosity, how do you make sure?  I ask because it's possible to 
design a map that is so hard that most people will say it's unbeatable, even 
though it really isn't.  ;)

> > Two more things I'd like to see, though.  :)  Ramming, such as when a
> > powered unit like a gunship or a tank are so weak they can't mount a
> > serious attack, but they can use their vehicle as a weapon to make a big
> > explosion.
>
> There was a proposal to add something like this as a special unit
> ability which would allow for units like cruise missiles or so.
> Personally, I don't like this much, though I'd have to think about
> it some more to explain why...

I'll tell you why I like it.  I like it because it's romantic, and that's all.

> > And also
> > being able to split infantry and distributing the remaining power of the
> > infantry amongst the new divisions.
>
> I don't like this either. The entire splitting issue is mostly a
> matter of perception, IMO. Initially the idea was for one piece to
> really be one unit. However, we didn't want 'instagib' units which
> would be destroyed with a single hit, so we needed something
> equivalent to hit points. We could have simply used a scale of
> 1 to 100 to indicate unit health, but decided on the piece = squadron
> approach which seemed to fit better. The idea behind it still the
> same, though, and allowing a unit to split would undermine the
> goal we had in mind (and allow players to swamp the map with
> countless and highly annoying amoebas). The only valid
> application I see here is the option to merge two damaged units
> into one (though I must admit this isn't entirely logical on
> second thought and maybe not such a good idea, either).

Just an idea.  I was mostly interested in having this ability for infantry 
only, because it would complicate the game too much otherwise.

> > Oo, one more.  The suicide stand.
> >
> > :)  Basically, you tell a piece "Kill everything until you get killed". 
> > : So
> >
> > they get some sort of enhanced power because of the resulting surge of
> > adrenaline a real unit would experience, but in exchange for that they
> > are dead soon.  Like maybe give them 3 turns of high power, and then make
> > them really really weak afterwards.
>
> Seems I'm being pretty negative today. You were talking something like
> 'highly advanced' above, but this sounds more like the middle ages,
> gripping your banner and going berzerk, swinging wildly and clubbering
> friend and foe alike. It doesn't really fit the overall theme.

No, it doesn't, but it is something of a fact of military life in general.  
When someone knows they have to hold a piece of ground for a length of time 
but they can't possibly survive (supplies cut off, surrounded, etc., like 
Custard at Little Bighorn, or any number of examples even during more recent 
wars), they get a charge of adrenaline.  I like this idea because I had it, 
and also because I haven't seen it run in any game yet.  I'm not saying it's 
required, just that it's something I haven't seen and I'd be curious how it 
played out.

> > Just some thoughts, anyway.  Maybe it's obvious that I've been looking
> > for a good game similar to Ogre/GEV, and Crimson is the closest to it
> > that I've found.  Those games were really cool, if painful to play
> > because they were regular board games.
>
> There's painful and there's painful. As I said I don't know Ogre, but
> I've one such game hidden in my closet. I played it exactly once and I
> can't really be bothered to try again. Setting it up took something like
> 1 and a half hours, and playing was virtually impossible without
> rereading the manual for every second move or so. If you get it right
> from the beginning a game can probably last for days, but really...
> (in case anyone wants to know, it's 'Against the Reich').

Ogre and GEV, most scenarios, could be set up in about 5 minutes, depending on 
how skilled the players were and how easy the pieces were to find (I had to 
dig under my bed for a lot of them :) ).  The rulebook was approximately 2-3 
pages long, the little 3x6 booklets.  The rules could all be contained in 
your head.  There was an odds chart, but if you knew the formula you didn't 
need it.  You took a piece's attack rating and the other pieces defense 
rating, divided, and rolled the die hoping to get that number or less.  The 
rules were very simple.  Terrain wasn't covered, except for water, so all 
pieces had few terrain restrictions.  It was a great game because the rules 
were so very simple, but balanced just right to allow for many styles of 
playing, so every game was different.

> Aside from that, even though I seem to shoot many ideas down I
> really appreciate every single proposal that might make CF
> a better game so I hope this wasn't discouraging. Even rejections
> sometimes help or turn into something different. One thing that's
> pretty important for me is the KISS principle, and to deviate
> from that path I need very good reasons. The term 'feature-bloat'
> hasn't been coined for nothing. Then again, sometimes something
> just has to sink in for a while (I'll just mention
> internationalisation here).

Yeah, simple is best.  There's a reason I tell people that Chess is the best 
strategy wargame ever made, and while there are many flavors, when it comes 
down to it I still like to play Chess.  :)

> To sum up, I'm ok with trying to rebalance aircraft and artillery
> units. We'd need some testers and this should be done before the
> release of 0.4.0, though, which means pretty soon. Also, if you
> have ideas on how to improve the combat resolution algorithm I'm
> all ears.

I'll go download the preview release and start running it.  I'll be happy to 
test.  I haven't cracked open the code yet because I have so much other stuff 
on my plate, and it's not likely I'll be able to for awhile still.  Then I 
have to get acquainted with it, and all kinds of stuff.  :)  So for right 
now, I'm most likely to be able to test stuff and work on maps when I run out 
of maps to play.  ;)  Music will have to wait a little while longer, but I'd 
like to throw together some jams in the near future to give you an idea what 
I'm interested in.  Classical seems to go well with war games...

> I thought you'd never ask ;-)
> The only sounds currently available are interface sounds (like pressing
> a button). The basic infrastructure for background music, movement and
> shooting effects has been in place (albeit untested), and I'm really
> only waiting for something suitable to show up as my sound engineering
> skills really suck. I even have a few sound effects which probably just
> need a bit of remastering to be useful but I just can't manage. Doh.
> I'm still undecided concerning the style of music we should use. I've
> been thinking about something slow with lots of drums, march-like,
> though I'm probably heavily influenced by a mod from some old Amiga
> demo I liked very much and which IMO would fit very well, but I can't
> find it anymore and I also have no idea who composed it so it
> couldn't go in anyway. As we don't want to scare the first players
> away at the title screen it should be, well, a little more
> 'mainstreamish' than what you have on your website (no offence, I
> actually listen to metal myself, but I don't really think it would
> fit in this context). In short, yes, we need something, but we don't
> know exactly what, yet! It's a matter of proposing something and
> convincing the overwhelming majority that it sounds great, I guess.

I'll see what I can come up with.  What's on my website is personal stuff, but 
I can compose for many styles and for many goals.  :)  I actually play a lot 
of blues these days.  In any case, I'm also very interested in drums, but I 
think that's just because the Shadow of the Beast soundtrack was mostly drums 
and percussion, and it was awesome.  But I'll see what I can come up with, 
when I get a chance.

> Jens

Dave

-- 
Visit my website!
http://www.davefancella.com/?event=em

Don't worry if you're a kleptomaniac; you can always take something for it.