|
Sapphire -> RE: =AQ= Spring Balance Update (6/10/2023 7:13:01)
|
quote:
2. Retaliation: Like Samukematsuri, you get rewarded for being hit. I had an idea where getting hit allows you to build damage/resist/source/other. Perhaps a No-Guard like effect where if the enemy hits you, then your next attack is more likely to land (reduced MRM or get bonus BTH on the next attack).
I think warrior needs something of flavor to add and provide some sparkle. Something somewhat similar to this, or something a la Titan's Mug misc baked into warrior somehow w/o causing imbalance IMO would be perfect.
I think it should maybe be a turn-based mod that the more damage you take that previous turn, the higher the damage mod goes for your outgoing damage. The -10% damage reduction aspect should play nicely. Maybe it's a small additional lean baked in, scaled via strength?
--------------------------------------------------------------------------------------------------------------------------------------------
In regards to guests, 1 of two things needs to occur.
1. Change every MP summons guest to SP. Rename the old summons to "Legendary" or something. Then allow people to just sell them off. If you're talking about a token/GGB item then maybe there be some sort of buyback. Since this isn't likely to be feasible...
2. You need 2 standards , depending on calls or summons.
The reason: MP guests for Mages replace the spell. Since Mages are assumed to use the spell slot , replacing say, an Ice Spell with an Ice Guest means that a Guest Summons should be assumed, too.
So you have standard 1 which is "Spell Replacement Theory". Isn't it an assumption that you have all 8 elements covered at base assumptions?
SO you must treat Guest Summons simply as a spell replacement. This means You need to try and match damage and cost to spell output over 20 turns, if possible.
However, 800% damage divided by 20 turns is 40% Melee/turn. And 653MP*4=2612 2612/20=130.6 Mp upkeep.
So let's roll with that temporarily. 40% Melee for 130.6 MP upkeep if we view this as spell replacement and thus, assumed damage.
quote:
Mage-> 800% (4 spell casts) + 1200% (16 weapon attacks) + 400% (20 turns of pet attacks @ 20% Melee) = 2,400% Melee.
Lucky Beast Mage-> 0% (0 spell casts) + 1500% (20 weapon attacks) + 800% (20 pet attacks @ 40% Melee) + 800% (20 guest attacks @ 40% Melee) =3100% Melee. Same MP.
Ranger --> 2000% (20 rounds at 100% Melee) + 400% (20 rounds of pets @20% Melee) = 2400% Melee (Matches Reg Mage)
FDBR--> 2000% (20 rounds at 100% Melee with bows) + 800% (20 rounds of pets @ 40% Melee) 2,800% Melee. Now, at base assumptions here the use of the spell slot for a Ranger isn't assumed, however, they have CHA investment. The fact that CHA investment isn't translated over onto assuming they're using the spell slot for a guest is absurd. If you include this even at 40% melee as above for the Mage, you have another 800% So FDBR skyrockets to 3600% Melee, beating everyone by a mile.
So what we have is
1. Pure builds have lower, standardized output. But the stat revamp allows them to train END, so where CHA provides more damage, END provides more HP's. IMO, this should be a worthy assumption, and accepted trade off.
2. Mage based Beast builds get to choose between spells or guests. Spells are assumed, yet guests are not. However, they occupy the same slot. This logically should mean that if spells are assumed for a Mage, which is the base build, then a replacement for a spell in favor of the guest should also be assumed for MP upkeep guests. It's logical. Here, you have a decision. Do you want more up front damage that will end up less by the end? Or more of a sustained damage model that will eventually surpass the spell model? To me, it might be a worthy and acceptable trade off. It's 2400 damage but a bulk of it is front loaded vs 3100 damage but only if you actually go 20 turns, which isn't likely in reality.
So my conclusion here is this: If you use the widely accepted, yet widely scrutinized 20 turn model, the numbers are off. Also, since guests are housed in the spell slot, IMO< a MP upkeep guest should be assumed in the model because it's literally replacing the spell. To me, this is common sense.
Also, when you compare the two Ranger variants, the spell slot isn't assumed in the current model, but for a CHA investment to me it's absurd to think the spell slot wont be assumed to use a Call guest to actually USE the guest. The fact that assumptions in both cases do not account for a guest I call into question. Because *everyone* is using them.
Let's say you move guest damage back to 60% Melee. The LBM's total damage becomes 3500% and the FDBR's damage becomes 4000%.
We now have Mage vs Ranger (and warrior) both running at 2400% Melee. So pure builds now run 2400% melee and likely train END for defense.
We now have LBM pushing 3500% Melee and FDBR (and likely warrior lean based Beast warrior) running 4000% Melee. These builds assume LUCK and not END.
In order to bring down that 4000% Melee from a FDBR to match the mage's 3500% melee, whom is FORCED to REPLACE their spell slot because it's ASSUMED, you'd need to drop a call version of the guest to 35% Melee/turn.
This reduces total melee from a guest from 1200% to 700% over 20 turns and now you have both LBM and FDBR at 3500% Melee.
So having two separate standards for CHA and builds to compare with, why not:
1. Assume non CHA builds run END (Not to increase monster damage, but to use as a comparative standard as to say, CHA vs END is Offense vs Defense
2. Assume CHA investment = using the spell slot for guests
A. In the case of Mages, training CHA assumes "Spell Replacement Theory"
B. In the case of Rangers/Warriors, you lower guest damage to 35% Melee
3. This means summons guests do 60% Melee @ 130.6 MP upkeep
4. This means call guests do 35% Melee and the upkeep should be VERY low, if not outright free.
If you moved guests down to 40% melee, the gap between LBM and FDBR is 3100 vs 3600 and it's the same 500%. Reducing 500% then likely needs to come from Ranger BOWS, so then bows will do 75% melee and now we have a match. This route has good points as well as issues. It makes the guest the same value in Melee%, bridges some of the gap between Pure build vs CHA build, and 40% melee for calls version is higher than the 35% in the other version. The issue with reducing bows is for FD non CHA trained Rangers. They just lost a lot of damage.
I think the first model is better honestly.
You could also take model 1 and "split the difference". Maybe Bows move to 90% and you lose 200% Melee there, then the other 300 comes from reducing 60% guest to 45%. This also balances the model.
Either way if you went 35% Melee or 45% melee (with also reducing bow damage, which might be the better route ) Then Call guest upkeep should just be dirt cheap, if not outright FREE because for Summons, it's replacing the spell and thus should be assumed. That means for DEx+CHA or STR+CHA then using the spell slot for a guest should also be assumed.
And what you have is a new assumption in terms of "Do I want offense or do I want defense, ie INT+END or INT+CHA......DEX+END or DEX+CHA??
Now, I personally think Luk needs greater play into BTH, slightly, and take it from Main-stat. This means no luck investment reduces BTH by enough such that it's noticeable if you go INT/END/CHA, for example. Which is the backlash build, and you aren't going to be relying on accurate attacks.
The totality of this model IMO works way better with updated and logical assumptions.
So to recap:
*New standard assumptions
*Cha=Offensive Decision vs END=Defensive Decision
*Luck=Increase BTH slightly on weapon attacks/spells/skills/pets/guests, reduce accordingly from Main-stat and pet/guest formulations.
*Investing in CHA=You're assumed to use guests
* 2 Guest standards
*Summons Guests assume "Spell Replacement Theory" and = 60% Melee @ 130.6MP/turn (same as 4 spellcasts in 20 turns)
* Reduce Bow damage to 90% Melee
* Call Guests are 45% Melee and FREE cost, or dirt cheap (maybe 5% Melee)
You could then use CHA style bonus differently. Maybe Call Guests get 60% Melee effects power. Maybe summons guests get 15% upkeep reduction? Call it a day.
At the end of the day, You need 2 standards and use some logical assumptions for CHA, or you *can't* balance it. (Like assuming you're using guests if you trained CHA, duh!) One for Summons Guests, and one for Call Guests. And you need to assume the use of guests if CHA investment. And you need to assume you are going offense (CHA) or defense (END) This should be the new assumption model. Or forever be broken. I'm sure people will hate this. But w/e
|
|
|
|
