[crossfire] Project: Slow down combat

[Mark Wedel]

Juha Jäykkä wrote:

> The idea of mana/grace regeneration needing rest is, again, one thing
> that has been used a lot in pen-and-paper RPG's over the time. I think it
> has proven itself a good solution. *BUT* in order to keep the magic users
> comparable to warriors in this kind of system, the magic users must be
> able to do a lot more damage than warriors per unit time. This is because
> the magic user can only cast a certain number of spells before retreating
> to regenerate the mana/grace while the warrior can keep up hacking and
> slashing. (Note that some systems have some kind of exhaustion or
> endurance for warriors as well, which means they need to rest as well. In
> that case the amount of damage/time may be closer to same.)

  Yes - some form of fatigue system has been suggested.  I'm not sure I'm crazy 
about that.

  One thing that I think could be quite easily done to help out mana users is to 
make weaker (and thus more common and cheaper versions) of mana regain potions. 
  Right now, there is the single 'magic power potion' which regains all your 
mana.  That is great if you have 500 mana and are high level so the cost isn't 
an issue.

  But if your level level and have 50 mana, you'd really just rather take the 
money from that and buy better spells, etc.  Adding in cheap versions of the 
potions that only give 50 mana max would seem easy to do, and make them 
relatively cheap.

  I'm not sure what to do for priests - with praying, they can get it back 
fairly quickly.  Having bottled grace seems a bit odd.

  I agree that spell casters probably need to kill things a bit faster than 
melee, but this gets trickier to measure.  For example, if the mage is using 
firebolt against a single monster, it being same speed or even slower would seem 
reasonable.  The reason is firebolt has the potential to hit many creatures, so 
if it is a hallway and you can hit 6 creatures with that firebolt, now on 
average kill rate, you are probably faster.  So it sort of becomes using the 
right spell for the job - burning hands on a single tile creature is not 
effective use - burning hands is useful because it hits many more creatures at 
once, etc.

  And mentioned elsewhere, to some extent, spell casters have some advantage in 
that they are typically going to be farther away (hitting creatures with range), 
which also generally means safer.

> 
>>   But this does lead to an interesting question - how do we deal with
>> classes that are not good at melee, especially hybrid classes?
> 
> Do we need to handle them any differently from basic classes? I do not
> think 50/50 fighter/mage has to be able to finish off a 80th level
> monster - the character is just 50th level, after all. In practise, this
> means twice the work for the dual-class character to reach 50/50 level
> when compared to a single-classer, but is that a problem? The
> dual-classer is basically playing two single class characters.
> 
> Paladins, rogues (do we have them?-o) and such are a more difficult
> question. They need careful balancing.

  Crossfire really doesn't have classes, it has skills.  But that question still 
applies.  There is also the warlock, which is supposed to be a fighter/mage type 
of combo.

  The answer may be that if you do diversify, you're not going to be as good as 
a specialist - that makes some sense.  A person that took that to an extreme 
(one handed weapons, 2 handed weapons, missile, praying, sorcery, evocation, 
etc) _should_ have a tougher time - if his top level in each of those skills is 
20, vs 80 if he specialized, being able to kill higher level creatures may not 
be possible.  Nothing preventing that character from killing level 30-40 
creatures of course.

> 
>> monster.  The simplest fix would just can't use a bow if next to
>> monster (and vice versa) whether it is attacking you or not.
> 
> I have a different idea about this. How about if arrows never hit a
> monster/*player* next to the archer? This should be easy enough to
> implement for single-square monsters, but how about bigger beasts, like
> dragons, I do not know. The nice thing about this is that it gives
> possibility of teamwork: put a fighter in front of the archer to keep the
> orcs from hitting the archer... and the archer won't hit the fighter
> either.

  I was sort of thinking the same way, but perhaps a bit differently - you can 
fire arrows (or bullets) through friendly creatures, but not enemy creatures. 
Thus, the same situation above applies (fighters in front row, spell caster or 
archers in the back).

  OTOH, this would also make things nastier for the players - the group of orcs 
could have ones in the back row firing at the player.

  I actually doubt the fact it was an archer next to the player was an issue - 
orcs, and many monsters, get generated with random stuff.  For orcs, most of the 
time it may be no weapon at all, or a crappy weapon.  But once in a while, they 
could get a good one - whether that is a bow or a sword or whatever, it makes 
that creature much more dangerous.

  Real problem here however is that in crossfire, there isn't any easy way to 
tell (you could examine the monsters, but spending the time to do that probably 
isn't good)

  Sort of goes back to Gros's point about seeing the characters (and then by 
extension, the monsters) weapon they are swinging, so you can what is being 
used.  I have some ideas how that would be done, but that isn't a small project 
(it would be right up there with letting characters dress themselves, etc).

> 
> I am a little confused about the talk about buffering keystrokes. Do we
> need to buffer them at all? What advantage does it give anyway? It seems
> to mostly make people die; have died many times myself casting a few too
> many spells in a row... Personally I hate it and use running all the time
> because it does not buffer (why?).

  This can be controlled to some extent in the command window option in the 
client - that basically limits the number of commands that can be pending on the 
server before the client starts dropping commands.

  The real problem with dropping commands is you may not really know what 
command is dropped.  If you're in battle, and do something like 'apply potion of 
healing' and 'attack east', dropping that apply potion of healing but still 
attacking east may not be the thing you want to do.

  Running doesn't really buffer because it operates different - the client 
basically says 'keep moving in this direction until I tell you to stop'.  So the 
server does that, looks at incoming commands and executes them (which may be 
stop running or change direction), etc.

  But if you're just doing general movement or other commands, they get sent to 
the server, and the server executes them as the character has actions to do 
them.  Most any player can type in commands faster than the character can 
execute them (especially true with compound bindings).

  The other reason to buffer at all is because you want the character to do 
something when he has an action.  If you set the command window to 1, what 
happens then is basically no buffering - the client sends a command, and will 
drop all subsequent commands entered until it gets confirmation of that first 
command being completed.

  The problem here is that due to time for packet traversal, there will be a 
time there where the character has an action, but doesn't have a command to do 
anything (the time it takes for the reply that the command succeeded to get to 
the client, and then the client to send a new command).  If you're playing on a 
local server, that isn't a big deal.  But if it takes 60 ms for a round trip, 
that is basically half a tick.

> 
> As what comes to the speed of the battle, I do not think it matters is
> clearing the newbie tower takes some time. I do not care if getting to
> level 2 takes a while etc. What I crave for is a sense of going forward,
> gaining something. This can be loot, levels, skills, spells; sometimes it
> can even be a piece of information eventually leading to some quest or
> just the history of the place. Slow pace is ok, I do not need to be 100th
> level in a couple of days.

  I agree.  IMO, getting the first few levels always happened really fast (level 
3 within an hour).  I think slower rate also helps out balance in other areas - 
if you can't kill things as fast, it means you can't get loot as fast either.


> Generators, of course, are a bit of a problem if killing monsters takes
> too long, but generators spawn-rate can be adjusted down accordingly
> (without need to remove the generators at all). I do not like generators
> but I do not think we should get rid of them - yet. They are too
> essential (they provide the challenge) in too many maps.

  A downside about generators is they also provide camping opportunities.  Maybe 
that is OK, but it means that a character can sit on a map and just wait for 
creatures to pop out, kill them, get loot & exp, and repeat for ever.

  I think I'd probably ammend what I said to 'generators should be used 
sparingly'.  I'm going to try reducing the number in the newbie tower, and also 
put them a bit more in the back and see what effect that has.

> 
> As what comes to spells, I am against big, medium and small versions of
> the same spells; just having the effect increase by caster level is
> enough. I also favour area effect spells and big numbers of monsters - in
> certain situations. This has to do with the above mention of spellcasters
> needing a higher kills/second ratio than fighters because their
> "resources" run out. (BTW spells like fireball are useless even now, they
> can kill nothing except critter: at low levels not even a fireborn can
> cast enough of them to kill any relevant number of orcs and when the
> caster has leveled up the player no longer wants to even hear the mention
> of an orc, which is not about the toughest monster one's fireballs can
> kill.)

  There was a suggestion that there should really only be one version that gets 
better as it goes along, and perhaps some way to modify them.  Eg, the 'lighting 
bolt' spell at high levels would act like large lightning, but it may be 
desirable to cast it as a 'small lightning' type of thing.

> 
> High level spells need a lot of thought. Things like frost nova should
> definitely be very high level or quest-spells. On the other hand,
> destruction is nearly useless at the moment but it is currently the
> highest-level spell there is... strange. I think some of the very highest
> level spells might not be combat-related at all, but this needs a little
> more dynamical playing world. (Thinking of combat spells one thought
> appeals a lot to me: earthquake... think of what that could do in a
> dungeon!)

  yes - destruction is a very old spell, and was never very useful IMO.