COMMITTED: Bard Pacify songs being "improved"

[sorvani]

Edit: just found SE_ChangeAggro, going to add that in and see if it handles it. wanted to post here before edit lockout.

Bard pacification songs are being improved by singing mods. This is a bad thing.

Problem was being discussed over on the PEQ forums.
http://www.peqtgc.com/phpBB2/viewtopic.php?t=11898

Here are the results of #showstats with no modifiers or AA

NPCID: 294000 SpawnGroupID:61376 
AggroRange: 5 AssistRange: 10

Here is where i started adding modifers

                                       Aggro | Assist 
No Instrument Mods + No AA =               5 | 10 
Equiped Epic 1.5 (Singing Mod 22 No AA) = 11 | 22 

Started Buying AA's in this order:    
Singing Mastery 1/2/3/4 =            6/7/8/9 | 12/14/16/18 
Ayonae's Tutelage 1/2/3 =           10/11/12 | 20/22/24 
Echo of Taelosia 1/2/3 =            13/14/15 | 26/28/30 
Equiped Epic 1.5 =                        21 | 42
Epic 1.5, All AA + Code change below =     5 | 42

I changed spell_effects.cpp with mixed results. Adding the SE_Lull exception did not change anything. Adding the SE_ChangeFrenzyRad keeps it from modifying the Aggro radius, but does nothing for the assist radius.

int Mob::CalcSpellEffectValue(int16 spell_id, int effect_id, int caster_level, Mob *caster, int ticsremaining)
{
	int formula, base, max, effect_value;

	if
	(
		!IsValidSpell(spell_id) ||
		effect_id < 0 ||
		effect_id >= EFFECT_COUNT
	)
		return 0;

	formula = spells[spell_id].formula[effect_id];
	base = spells[spell_id].base[effect_id];
	max = spells[spell_id].max[effect_id];

	if(IsBlankSpellEffect(spell_id, effect_id))
		return 0;

	effect_value = CalcSpellEffectValue_formula(formula, base, max, caster_level, spell_id, ticsremaining);

	if(caster && IsBardSong(spell_id) &&
	(spells[spell_id].effectid[effect_id] != SE_AttackSpeed) &&
	(spells[spell_id].effectid[effect_id] != SE_AttackSpeed2) &&
	(spells[spell_id].effectid[effect_id] != SE_AttackSpeed3) &&
	(spells[spell_id].effectid[effect_id] != SE_Lull) &&
	(spells[spell_id].effectid[effect_id] != SE_ChangeFrenzyRad)) {
		int oval = effect_value;
		int mod = caster->GetInstrumentMod(spell_id);
		effect_value = effect_value * mod / 10;
		mlog(SPELLS__BARDS, "Effect value %d altered with bard modifier of %d to yeild %d", oval, mod, effect_value);
	}

	return(effect_value);
}

[sorvani]

ok adding in SE_ChangeAggro did not fix it. Any suggestions on how/where to find the effect?

[realityincarnate]

I think SE_Harmony is what you're looking for. I was just looking at this the other day because mana regen is getting instrument modded incorrectly too. I hadn't noticed the aggro range problems, though.

[sorvani]

it is, i just figured out how the spell_effects and the database work together.

and EVERYTHING is getting modified as far as I can tell except for the attack speed increases that were already excepted.

world is building building now and then i'll test it out.

[sorvani]

Originally Posted by realityincarnate

I hadn't noticed the aggro range problems, though.

If you do not have a singing mod equipped and do not have AA yet, then you will not have a problem.

Been complaints from bards on the PEQ forums ever since the pacify level caps went in.
I thought this might be an easy thing to track down once i had my own server turned up and had GM commands to check the stats. A simpler thing to cut my teeth on you might say.

[sorvani]

Working on my test server. unified diff sent to gaeorn.

int Mob::CalcSpellEffectValue(int16 spell_id, int effect_id, int caster_level, Mob *caster, int ticsremaining)
{
	int formula, base, max, effect_value;

	if
	(
		!IsValidSpell(spell_id) ||
		effect_id < 0 ||
		effect_id >= EFFECT_COUNT
	)
		return 0;

	formula = spells[spell_id].formula[effect_id];
	base = spells[spell_id].base[effect_id];
	max = spells[spell_id].max[effect_id];

	if(IsBlankSpellEffect(spell_id, effect_id))
		return 0;

	effect_value = CalcSpellEffectValue_formula(formula, base, max, caster_level, spell_id, ticsremaining);

	if(caster && IsBardSong(spell_id) &&
	(spells[spell_id].effectid[effect_id] != SE_AttackSpeed) &&
	(spells[spell_id].effectid[effect_id] != SE_AttackSpeed2) &&
	(spells[spell_id].effectid[effect_id] != SE_AttackSpeed3) &&
	(spells[spell_id].effectid[effect_id] != SE_Lull) &&
	(spells[spell_id].effectid[effect_id] != SE_ChangeFrenzyRad) &&
	(spells[spell_id].effectid[effect_id] != SE_Harmony)) {
		int oval = effect_value;
		int mod = caster->GetInstrumentMod(spell_id);
		effect_value = effect_value * mod / 10;
		mlog(SPELLS__BARDS, "Effect value %d altered with bard modifier of %d to yeild %d", oval, mod, effect_value);
	}

	return(effect_value);
}

[sorvani]

I am not sure how many bard songs should not be affected by mods, but if it is a lot should new DB columns be considered? or existing columns made use of for bard song effect restrictions?

Also on the SoD and UF clients there is a problem with things like run speed and resists showing correctly server side, but the client is getting rollover or bad data of some kind and kicking out slow run speeds and low resists. Would it be possible for a limit to be put on what value is sent back to the clients to prevent the client from getting really out of sync with the server?