Increase witch rage barrier

[GoogleCodeExporter]

= Requirements =
The witch should become aggressive (go after the survivors) from a further
distance than it currently is.

= CVARS =

l4d2_srs_witch_aggro_distance (0 to disable, ingame units of distance)

= Implementation Suggestions =

Investigate all of the z_witch_* cvars, one of them can surely provide this
functionality for us (e.g. z_witch_threat_normal_range,
z_witch_thread_hostile_range).

Original issue reported on code.google.com by gorik.sm...@gmail.com on 11 Jan 2010 at 8:14

[GoogleCodeExporter]

= Last Notes =

If it is possible to implement this as a L4D2 cvar change only, do not write 
any new
code for this. Instead note which cvar we need to change and what the values 
mean,
and close this issue.

Original comment by gorik.sm...@gmail.com on 11 Jan 2010 at 8:26

[GoogleCodeExporter]

Original comment by gorik.sm...@gmail.com on 11 Jan 2010 at 9:56

[GoogleCodeExporter]

= Cvar Investigation Findings =

z_witch_wander_personal_time - Time it takes when within a wandering witches 
personal
space for witch to startle. Default 10 seconds.

z_witch_wander_personal_space - Wandering witch personal space distance. 
Default 240.

z_witch_personal_space - Personal space distance for sitting witches. Default 
100.

z_witch_anger_rate - Rate for witch to alert from the time it is hostile. 
Default 0.2
(1.0 is instant i.e. if you are within its threat range it will startle).

Crouching witches do not have a 'personal space time' cvar, if you are in their
personal space they will start to startle based on the anger_rate (unless 
scratched
by SI).

z_witch_threat_normal_range - Distance you can be within range of a witch before
making her hostile (i.e. susceptible to SI scratches and anger_rate to startle 
her).
Default 300.

z_witch_threat_hostile_range - Distance you can be within range of a witch 
whilst
engaging in hostile action (i.e. firing your weapon) before making her hostile 
(i.e.
susceptible to SI scratches and anger_rate to startle her).

(Confusion whilst investigating this particular cvar as I assumed the 'hostile'
referred to the witch rather than the survivors actions).

The 'threat' cvars only relate to crouching witches. Valve simply increased the
personal space for wandering witches to counter act 'threat' so that you might 
not
accidentally make a witch hostile (susceptible to SI scratches) if they wander 
near
survivors. 

Ultimately the deciding factors in range for startling a witch is decided by 
either:
a) entering her personal space for too long (10 seconds for wandering, 
specified by
the anger_rate for crouching (~3 seconds as default)).
b) Entering her 'threat' ranges and having her stay hostile for a time 
specified by
anger_rate whilst continuing to stay within her 'threat' range.
c) Entering her 'threat' ranges and having her scratched by a SI.

=Requirement Solution=

Increase the personal space for sitting witches from 100->240 (to match 
wandering).
Reduce personal space time for wandering witches from 10->3 (to match sitting).

It might also be worth considering increasing the personal space for wandering
witches to make them susceptible to SI scratches from a bigger range to punish 
sloppy
survivor play, as there is no 'hostile' function like that of a crouching witch.

Other interesting witch cvars:

z_witch_damage_per_kill_hit - Damage per tick to in capped survivor. Default 
30. This
is the same value as in L4D1 but the witch hits survivors on the ground 
considerably
slower in L4D2 than in 1. I've failed to find any cvar which changes her swing 
speed.
Perhaps consider changing this value to sync with similar timing for a witch 
kill as
in L4D1. I propose 75.

Original comment by phil...@gmail.com on 11 Jan 2010 at 4:06

[GoogleCodeExporter]

Can you also please note what exact CVAR values would correspond to the 
solutions (so
it's crystal clear for everyone reading this).

Original comment by gorik.sm...@gmail.com on 11 Jan 2010 at 8:23

• Changed state: Started

[GoogleCodeExporter]

= Solution =

To simply fulfill the requirements of this enhancement I propose the following 
changes:

z_witch_personal_space 240 (Default - 100)
z_witch_wander_personal_time 2 (Default - 10)

= Further Considerations =

Might it be worth considering increasing her 'threat' threshold and/or her 
anger rate.

z_witch_threat_normal_range (Default - 300)
z_witch_anger_rate (Default - 0.2)

At this value a player can run straight past a witch and be out of her threshold
again before she is fully hostile (according to her 'slow' anger rate). If you 
were
to either increase her threat range up from 300 so that you are within range of 
her
for longer OR increase her anger rate up from 0.2 so that whilst you are within 
her
threat range she will startle quicker to negate survivors running past.

Currently the difference between witch personal space (startle distance) and 
witch
threat range (enraged distance) is 100->300 for sitting, or 240->300 with the
proposed changes. As long as the personal space does not overlap the current 300
threat range then it might not be worth changing. The main benefit of 
increasing the
threat range is the distance at which survivors will enrage the witch leaving 
her
open for SI scratches to scare survivors.

Original comment by phil...@gmail.com on 11 Jan 2010 at 9:23

[GoogleCodeExporter]

(CCing Donahue)

Donahue, would increasing the 'threat' threshold and 'anger rate' be within the 
scope
of the original requirements?

Original comment by gorik.sm...@gmail.com on 11 Jan 2010 at 11:10

[GoogleCodeExporter]

Hi.  Yes, I think so.  I'd also be curious as to what the range is that the 
witch music 
triggers at.  Bottom line, we need to make the witch back into something that 
should be 
feared or at least treated with caution by survivors.

Original comment by mdona...@gmail.com on 11 Jan 2010 at 11:56

[GoogleCodeExporter]

It depends on the understanding of the scope of the requirement. One of the
misconceptions about the wandering witch, as the prime example, is that she 
wouldn't
startle even if you are right next to her. In reality her startling range is 
actually
greater than that of the sitting witch and of the L4D1 witch. Instead it is the 
time
it takes her to startle when within her range that affects the witch being 
seemingly
nonaggression.

However, the wandering witch does not have a 'enraged' threshold, like that of 
the
sitting witch (as set by the threat_range cvars), presumably as Valve believed 
it
would be unfair if it wandering within range behind you survivors not notice 
(If you
were shooting your weapon and it wandered within 600 range of you then she will 
begin
to enrage).

Default is at 240 range with her taking 10 seconds to enrage. These settings 
make it
impossible to fail a crown even with SI trying to interfere as you have 10 
seconds to
clear any attack before even attempting to shoot the witch (shooting near her 
has no
affect on her enrage as she has no threat threshold).

I believe dropping the time to enrage to 2 seconds will make it more difficult 
to
back out of a crown and give S a chance to affect the crown attempt (this value 
will
make the startling time when within range of the witch similar to that of a 
sitting
witch). However if you wish to make her range greater than 240 then I propose a 
value
no higher than 300 (to match the enrage threshold for the sitting witch). Again 
if
you wish to make it higher than 300 then you will want to consider increasing 
the
thresholds of sitting witch so each witch it no harder than another.

The z_witch_anger_rate cvar does not fit the scope of the original requirement 
of
increasing her enrage range. I have suggested the 2 second
(z_witch_wander_personal_time 2) change for wandering witches to bring it in 
line
with the 0.2 (z_witch_anger_rate 0.2) default for sitting witches. Witch anger 
rate
and personal space affects how long you can be within the other thresholds and 
how
long she it takes her to enrage.

Original comment by phil...@gmail.com on 12 Jan 2010 at 2:56

[GoogleCodeExporter]

Do people think both the sitting and the standing need to be buffed? After my 
testing
I believe that it is only the wandering witch that brought about discussion of
change. The sitting witch still works exactly the same as L4D1. If you wish to 
buff
the sitting witch personal space then you will probably also need to change the
sitting witch's threat thresholds, depending upon the value you decide for the
wandering witch personal space to keep them all in line with eachother.

I think requirement needs to be reworded to involve both the sitting and 
standing
witches, what are peoples thoughts?

Original comment by phil...@gmail.com on 12 Jan 2010 at 3:02

[GoogleCodeExporter]

My personal opinion is that they should be treated a bit differently, inasmuch 
as the 
"range" should be different.  Conceptually, the survs stumble across a sitting 
witch,and theoretically, a roaming witch stumbles upon the survs.  One idea 
would be 
that relationally, the sitting aggro range is greater, but time to activate is 
slower, 
and for the roaming witch, it is reversed, activation range is smaller but time 
to 
activate is quicker.

Original comment by mdona...@gmail.com on 12 Jan 2010 at 7:18

[GoogleCodeExporter]

Reassigning to Fission for feature correctness.

Put this in a config and verify it does increase the witch rage.

-----------------------------
sm_cvar z_witch_personal_space 240
sm_cvar z_witch_wander_personal_time 2

Original comment by gorik.sm...@gmail.com on 14 Jan 2010 at 12:49

• Changed state: Fixed
• Added labels: Test-Feature

[GoogleCodeExporter]

Original comment by fissionm...@gmail.com on 15 Jan 2010 at 7:15

• Changed state: Verified