q3aServ/cpma/docs/arQmode.txt

Challenge ProMode Arena: arQmode
--------------------------------
Date: 28 Nov 03

There is no "arQmode" any more. It's been tweaked and is now the official
ruleset of Challenge ProMode. This document exists solely for "historical"
purposes.


Date: 18 Feb 03

arQmode is an unofficial version of Challenge ProMode that incorporates some
of the changes we've discussed over the years but opted NOT to include in
ProMode itself for one reason or another. It's really just a place for me to
experiment with gameplay, and unlike ProMode the settings for arQmode can
(and most likely will) change on a whim without any Design review.
I'll refer to it as APM from now on to save typing.  :)

The strategic elements of modern ProMode (armour specifics, weaponspawns,
MH behaviour, sound system, etc) are so close to perfect after the last few
years of CPMA refinements that changes in that area would provide at best
miniscule improvement in return for gratuitous differencing from CPM, so
APM (like CQ3) generally leaves them untouched. Instead it focuses on melee,
and especially on the behaviour of the "top tier" weapons.


The APM changes are:

Biased RL Damage
Motivation: Reward use of high-skill weaponry.
Splash damage from the RL now follows a Q1/Q2-like scheme (though nothing
like as ubernoobed) where ANY hit does a minimum of 15 damage, and the
damage/distance relationship is biased accordingly, i.e. a shot that does
50 damage in CPM will do ~60 damage in APM. Note that although the base
knockback factor remains the same as in CPM, knockback on splash hits
will also be higher since the base factor is scaled by damage. RL speed,
base damage, splash radius, and max splash damage remain unchanged.

Decreased Splash Vertical Knockback
Motivation: Ameliorate the higher RL knockback.
CPM's vertical knockback base is 150% of CQ3's. APM's is 133% of CQ3's,
or 89% of CPM's. Since an "average" RL hit in APM is ~20% stronger than
in CPM, players were thrown around too much. While this change reduces
the "lift" from a perfect RL shot by a small amount, when coupled with
the previous change it increases lift on an overall basis for the vast
majority of shots, and you can still juggle people very nicely.  :P

Increased PG Damage and Knockback, Reduced Speed
Motivation: Reward use of high-skill weaponry.
The PG's *direct hit* damage has been raised from 15 to 20. The splash
damage remains unchanged, and the splash radius is slightly reduced.
These changes promote the PG to "top tier" / "power trio" weapon status.
Knockback has been increased implicity along with damage, but only
slightly: it's now roughly that of the CPM LG.
The projectile's speed has been reduced from 2000 to 1800.
This keeps the weapon extremely dangerous at close range (where it's still
effectively a hitscan weapon) but reduces its utility as a mid-range spam
weapon since good players will be able to dodge it to some extent as was
possible in Doom; and greatly reduces its value for "pure spam" cases on
CTF and other large maps. Speaking of which...
The one significant issue with a strong PG lies not so much with the
weapon itself but with the behaviour of a certain class of player: the
tendency for newbies to simply spam it until they run out of ammo. While
not really a problem in skill games, it becomes a very obvious one in CA
and CTFS, as those gametypes not only have 99% of the newbie population
but also provide the weapon and ammo for free. Fortunately, such people
are far more likely to play RA3 or Threewave than CPMA.  :P
Regardless, the ammo limit for the PG has been reduced to 100, which is
more than enough for DM or even TDM play.
The size of the plasma graphic has been reduced too, since I REALLY HATE
the way Q3 does them; and the shots themselves are in reality infinitely
small, contrary to their appearance (especially in VQ3).

Decreased LG Damage and Knockback
Motivation: Penalise use of low-skill weaponry.
The LG is 90% ping and 10% ability. Terrifying in the hands of even the
most newbie SLPB yet near-useless to HPBs regardless of their skill, it
has no place as a top tier weapon. The interesting impact of damage
changes to the LG is how that damage is distributed across health and
armour. In CPM, the LG damage is always split 2H 6A whether the enemy has
RA or YA: despite the lower protection percentage of YA, the rounding
applied to the factors involved makes both armours equivalent. Dropping
the damage to either 6 or 7 makes the armours distinct again, but the
health loss is the same whether the LG is 6 or 7: damage is applied as
1H xA against RA and 2H xA against YA for both.
I flipped a coin and chose 6.
To look at that another way: against a YA player, all 3 "total damage"
factors reduce health at the same rate: the only difference is how fast
they go through armour. Against an RA player though, either reduction in
LG damage makes it virtually worthless for actually killing people until
such time as the enemy's armour has been eroded. As such, the LG's role
is now very much restricted to that of a finishing weapon and a way to
"test" the enemy's health and then combo from.
Knockback has been reduced implicity along with damage, but increased
explicitly to compensate somewhat. CPM's LG has a knockback of 10 (units
per hit) and I originally set APM's to 8, which is JUST enough to hold
someone in the air, but it requires exceptional aim to do so. In the end
I upped it to 9 simply because it's just so much more fun at that when
people take jumppads.  :)
9 Mar 03: The only thing I DON'T like about 6 is that it means LG now
has the same hit tone with Quad as without, which is a bit of a shame.
So I got sneaky and increased it to 6.5 instead. Since you can't do
fractional damage, that means it stays at 6 normally, but with Quad it
jumps from 24 to 26.

Decreased Rail Damage and Knockback
Motivation: Penalise passive play.
Now half of its CPM value, enabling fast players to break through "standy
railer" defenses. It's extremely tempting to leave the knockback factor
at 1 but change the actual damage of the RG instead. Reducing it to 75
for example removes the ability to use it for skill-less spawnrapes but
leaves it strong enough to be used for "balance of power" shifts or to
finish an opponent with. It would however diminish the overall power of
the RG to a level that removes it from the top tier, thus leaving no
hitscan weapon in that class, which I think would be a mistake. The only
way to compensate for that would be to reduce the reload penalty or
reinstate cessrail: of the two, a consistent but faster reload is
probably still the more desirable option, but with the LG also modified
the classic cessrail combo is really a non-issue for APM already.
9 Mar 03: Okay, tempting enough that I'm going to try it.  :)
Damage is reduced to 80; overall knockback stays at half of CPM/VQ3.
I found it impossible to decide between a constant faster reload and
just re-enabling cessrail by default: there are pros and cons to both.
In the end I went with this: if you enable fastrail, you get the standard
FR timings of 1000 and 1500; whereas if you leave it disabled (default)
then the reload is dropped to 1250.
APM was already strongly RL-dominated, and I think that's even more the
case after this change. The RG still clearly has an effective role even in
DM though: it remains the weapon of choice for a couple of key situations,
so I'm willing to run with this for a while and see how it plays out.

Assorted BFG Tweakage
Motivation: Avoid Superweapons.
When I originally set the BFG to be a "120% RL", the plan was that we use
that as a starting point and then refine it from there. That never happened
in CPM, and the weapon remains +20% over the RL in all these aspects:
speed; damage; splash damage; splash radius; and implicitly, knockback.
(The knockback scale is the same as the RL's, but since the damage is +20%
the knockback automatically is also).
APM shrinks the blast radius down to 20% *less* than the RL. Interestingly,
because of the new bias on the APM RL the two weapons now cause roughly the
same damage for "average" shots, but the BFG causes more on "good" shots
and less on "bad" shots.

Biased Spawns (the only strategic change)
Motivation: Minimise streaks of "cheap luck".
Biased player respawns mean that a player CAN respawn at any spawnpoint,
but is more likely to respawn some distance from where he died. While
clearly still superior to baseq3's utterly predictable respawns, and
significantly less likely (note: NOT impossible) to result in unpleasant
"get spawnfragged in the same place 3 times in a row" than the "true
random" scheme that CPM uses, there are still elements to it that I'm
not really comfortable with. My personal opinion is that that there IS no
"best" way to handle respawns, because of the variance in map layouts etc
and in the events that lead up to the frag in the first place. Still, to
paraphrase the ELM docs: "All respawn schemes suck. This one just sucks
less than the others".  :)


Changes that STILL didn't make it:

Increased Ground Friction
Motivation: Improve player agility when on the ground.
Something that was supposed to be in *pre-CPMA* CPM even, but in reality
was never changed. Friction is 25% higher, allowing for faster stops and
increased dodging ability. It has the interesting side-effect of reducing
"launch speed" (i.e. the speed of the initial CSJ when starting a run) by
about 30-50ups, which makes certain now-common tricks like stairs->MH on
CPM3 significantly harder.
Final Verdict: Had we done this years ago, it arguably would have been
the better call, but at this stage it just seems limiting.
Moreover it forces players to relearn the initial CSJ and thus introduces
a conflict between CPM and APM movement, and a barrier to being able to
play both. Those factors more than outweigh the slight improvement that it
makes to the ground game.