[DIS] Mallard's Eye suggestion

[asuratva]

Hi,

Wanted to suggest improving viability of archers in swamp. Personally, I always preferred playing archer or mage, but currently, warrior is pretty much the only effective option for levelling at swamp.

My suggestion is to improve Mallard's Eye to make it armour-piercing. Specifically, focusing mallard's eye should allow arrows to ignore a percentage of target's def. Unlike crits which ignore all def, this should be easier to balance. Plus, without speed or warrior's high def, it should keep archer from being too broken.

Survivability requires running around and sniping from range. Getting hit by mobs is a death sentence

If people think this is a decent idea, might need to consider slightly reducing the range and/or speed of swamp mobs to allow archers a little breathing room to run around.

I've made a branch for testing at https://git.themanaworld.org/asuratva/t ... rds-pierce if anyone wants to see how it feels. The changes are all in src/map/battle.cpp around line 1709. Currently I've set it ignore ~ 70% of def at 100 dex and 9 mallard's eye, but that probably needs to be adjusted by someone who has a better grasp of balance.

[Hello=)]

Subject: [DIS] Mallard's Eye suggestion

Hi,

Wanted to suggest improving viability of archers in swamp. Personally, I always preferred playing archer or mage, but currently, warrior is pretty much the only effective option for levelling at swamp.

My suggestion is to improve Mallard's Eye to make it armour-piercing. Specifically, focusing mallard's eye should allow arrows to ignore a percentage of target's def. Unlike crits which ignore all def, this should be easier to balance. Plus, without speed or warrior's high def, it should keep archer from being too broken.

Survivability requires running around and sniping from range. Getting hit by mobs is a death sentence

If people think this is a decent idea, might need to consider slightly reducing the range and/or speed of swamp mobs to allow archers a little breathing room to run around.

I've made a branch for testing at https://git.themanaworld.org/asuratva/t ... rds-pierce if anyone wants to see how it feels. The changes are all in src/map/battle.cpp around line 1709. Currently I've set it ignore ~ 70% of def at 100 dex and 9 mallard's eye, but that probably needs to be adjusted by someone who has a better grasp of balance.

Oh seems I missed some really nice idea. I appreciate such efforts. And had share of ideas how to deal with issue. So far known things to try are...

1) Add some crit def on swamp mobs to reduce DPS of rage warriors a bit. I mean, to swamp specific mobs only.
2) Add some crits to banshee as they were. Its just been massively OP - but now it looks like massive overnerf.
3) Add some crits bonus to various "advanced" arrows like bone arrows. So far they make no point to buy them at 4-5GP each.
4) Perhaps make "compound" attacks, E.g. "enchanted" attacks. Technically there's quite funny construct, overrideattack() builtin. It mostly been meant for magic but it doesn't follows from anywhere magic should be the only caller. Like that, upon attack script is invoked, given target and can do whatever it wants.
5) Add some ranger-friendly mobs. That is - with strong - but short range - attacks, so warriors wouldn't like that while mages and archers would be okay since they do not have to constantly withstand hits.
6) I had idea to try items that "favor archers" - e.g. making Enchanter Amulet giving bonus to rangers "as Enchanters wearing it were". Perhaps even checking for magic presence. Item can run script on equip and its that script what decides what bonuses would be. Fact most items used primitive logic only means "feature were grossly underused". It can look up focus, make decisions, adjust values etc. Tentatively I want it to be like +18 hit -> warriors while keeping +22 to archers and give some cris bonus to archers. What would you think?

As for survivability. Consider simple thing. Archers were OP for a while, being highest DPS class in game AND unbeatable in PvP, esp optimized PvP setups. There were many reasons. One of them would be: warriors need to routinely withstand retribution, wasting stats point. Archers don't. Furtheremore, mages get huge malus for wearing heavy armor. But archers don't. Actually its maybe bug. I think at least heavy armors should have dex/agi malus. Most of archers not really using them anyway (due to lack of str to carry that?). But keep in mind archer in heavy armor is possibility. I think most archers try overaggressive setups badly lacking Hp.

When it comes to swamp in particular I would say key issue is pathetic DPS. I think fiddling with archer-specific crits bonuses on "archer" items like monocle or enchanter amulet can solve that, at least to some extent. And on my side of story I learned one thing. "The pure and simple truth rarely pure and never simple". On my side I would prefer to try gradual, "perfurb and observer" approach - tuning things instead of massive simple grand solutions - which tend to have a lot of woes attached.

On technical side I wonder: why reinvent, basically, another kind of crits, if there're many ways to recalibrate things using usual crits? Does this approach brings some advantage? Apparent downside it is +1 variable in already fairly complcated "equations" which are quite problematic to balance and +1 interaction - making "whole thing" even harder to estimate - esp various interactions in corner cases. But overall I have to say its nice, fresh look on problem and there's something in it. I just wonder what you think about mentioned ideas? On technical side I'm curious to implement "smarter items" (lore wise - "magic items having minds of its own, preferring to serve rangers, just as Enchanters were") - and "compound attacks" where script can make "dynamic" decisions. E.g. some enchanted arrows can have "magic" damage "stappled" - esp if archer can into weak magic, like some archers routinely practice anyway to do some heals.

NB nothing set in stone. Its just my ideas how to move forward on this issue.

[asuratva]

Subject: [DIS] Mallard's Eye suggestion

On technical side I wonder: why reinvent, basically, another kind of crits, if there're many ways to recalibrate things using usual crits? Does this approach brings some advantage? Apparent downside it is +1 variable in already fairly complcated "equations" which are quite problematic to balance and +1 interaction - making "whole thing" even harder to estimate - esp various interactions in corner cases. But overall I have to say its nice, fresh look on problem and there's something in it. I just wonder what you think about mentioned ideas? On technical side I'm curious to implement "smarter items" (lore wise - "magic items having minds of its own, preferring to serve rangers, just as Enchanters were") - and "compound attacks" where script can make "dynamic" decisions. E.g. some enchanted arrows can have "magic" damage "stappled" - esp if archer can into weak magic, like some archers routinely practice anyway to do some heals.

NB nothing set in stone. Its just my ideas how to move forward on this issue.

I heard someone (forgot who) saying that crits were too complicated to balance around. That was the impetus for the "armour-piercing" idea. Crits tend to ignore hit-rate and armour altogether. The "armor-piercing" does not ignore 100% opponent DEF, but instead a variable percent. I thought this might help with balancing somehow. But I guess I didn't put enough thought into it.

I like your idea of using overrideattack() -- it would be useful for implementing active skills for non-mages. As of now, characters only have passive skills. I was talking to ThinkSome, and he had the idea for a "knock-back" skill for archers that pushes the mob they are attacking a few tiles away from them, eg., they use small amount of SP to cast "Knock-Back" which for the next ~ 50 attacks has a 20% chance to push back the enemy mob. And we can check this only works with bows (or any weapon with range > 3). This would greatly increase survivability, and add some minor complexity to the class. - i.e., archers now have to decide whether to preserve SP for heals or use it more aggressively for "Knock-Back".

The new skills could also be focus-specific - eg., Knock-Back is only available to chars focused on Mallard's Eye.

[cuoco]

Subject: [DIS] Mallard's Eye suggestion

..
2) Add some crits to banshee as they were. Its just been massively OP - but now it looks like massive overnerf.
...

This is quite an interesting idea. Would love to see the Banshee Bow used a lot in the game again.

[Hello=)]

I heard someone (forgot who) saying that crits were too complicated to balance around. That was the impetus for the "armour-piercing" idea.

What's complicated for one, manageable for other. Person who can code builtin overnight after random chat is alive proof. Crit wise, mobs have notion of crit def. To me it looks like efficient handle to tweak crits - even somewhat based on location. E.g. its possible to tweak but not change candors much - due to different sets of mobs. It came to me recently crit def just virtually unused as tunable. In fact a lot of nice things were not used in TMW, hell knows why.

This said: balancing (MMO)RPG? Its not just complicated. Its insanely complicated to point it impossible in its full. Even for pro gamedevs. Just remember Starcraft I. Zergs were initially pathetic in air - to point it got fixed in addon. And that's pro gamedevs who had adequate architects, testing, all that. But sorting out "major" woes still within reach.

Crits tend to ignore hit-rate and armour altogether. The "armor-piercing" does not ignore 100% opponent DEF, but instead a variable percent. I thought this might help with balancing somehow. But I guess I didn't put enough thought into it.

To me it looks like it sorta compounds problem further. Just in case I'm not against this thing - so do not take it as "no" by any means. But as concrete thing, most high level archers preferred "speed" over "mallard". So far natural evolution path of archer been use mallard at low level to hit things better, go for speed on high levels. What if we assume this landed? What outcome you target as set point? High level archers moving to Mallard over "speed"? And who are users of "speed skill" would be? With its fancy quest it would be sad to make this skill unused/useless. So yeah, balancing things is complicated. And its where I'm puzzled.

like your idea of using overrideattack() -- it would be useful for implementing active skills for non-mages. As of now, characters only have passive skills. I was talking to ThinkSome, and he had the idea for a "knock-back" skill for archers that pushes the mob they are attacking a few tiles away from them, eg., they use small amount of SP to cast "Knock-Back" which for the next ~ 50 attacks has a 20% chance to push back the enemy mob.

To my taste... TMW can do a lot of cool things it not currently does, even with existing engine, as it is. There's many things nobody just bothered to implement. Not like if its hard to implement, just nobody bothered. "Advanced" attacks or more of fun spell are obviously in this list. I'll try to change that.

And we can check this only works with bows (or any weapon with range > 3). This would greatly increase survivability, and add some minor complexity to the class. - i.e., archers now have to decide whether to preserve SP for heals or use it more aggressively for "Knock-Back".

Actually quite interesting idea. But what I really do not understand is how "speed skill" fits this scheme.

The new skills could also be focus-specific - eg., Knock-Back is only available to chars focused on Mallard's Eye.

In theory it can be funny to have >1 focusable skills. Technically nothing prevents it. But balancing wise... you can imageine.

This is quite an interesting idea. Would love to see the Banshee Bow used a lot in the game again.

As such e.g. items can even check for presence of each other item and e.g. its possible to try to e.g. make what other games call "compound artifacts". E.g. behavior of one artifact may depend on presence of another. As such items invoke scripts on equip and these can do quite arbitrary tweaking/computations based on whatever. Key issue is: edge cases are hard to foresee. But players would find most crazy and exotic combo and (ab)use it in its full.

[Livio]

Subject: [DIS] Mallard's Eye suggestion

..
2) Add some crits to banshee as they were. Its just been massively OP - but now it looks like massive overnerf.
...

This is quite an interesting idea. Would love to see the Banshee Bow used a lot in the game again.

Uhh... I have to admit I didn't fully read this whole thread (I'm feeling pretty tired) but I guess I must tell you this:
When banshee bow wasn't nerfed I refused to use it for long time until I leveled enough to have enough VIT to spare for it and avoided to use the pots due to the heavy HP drain so I kept using the forest bow until level 95 IIRC.
After the nerfing, well... I admit I didn't complain too much since using the nerfed banshee bow allowed me to exterminate weak monsters way faster than before without getting killed in the process so I managed to get lots of money by hunting in the forest while feeding myself on fruits (and sold lots of roots to some mages too).

I never focused on Mallard's eye for too long since it's slow on attack and not every map allows you to fire from long distance. However, it would be cool a magic spell for that skill that allows to charge a slow and powerful shot just like #ingrav does for the mages. Just to feel like a real sniper...

[Hello=)]

In theory its possible to have quite arbitrary attacks doing "anything" - it not even have to be magic. There's some funny construct, overrideattack() builtin, that invokes arbitrary script, and script knows 1) who're you and 2) what you've been to attack. It can do really arbitrary logic at this point.

This been mostly made for spells, but it not really needs to be spell. For example on my local test I have "target sharing with summon" that scans map around of this spot, finds summon and requests it to hit target I've been about to attack. So its targeting system - even if client knows nothing of this. Like that it possible to compute AoE, or some unusual thing. But guess its not very nice to invoke scripts on very fast frequent attacks. Though it works quite well with #upmarmu it seems.