1일
* 팬텀 리워크
- 블링크 주문 삭제
- 맞은 적을 블링크시키는 블링크 공격을 함
- 하운트 주문에서 팬텀 빠짐(궁합이 안 좋아져서)
- 체력 30% 감소
- 이 공격은 개미맨을 이동시키지않음

2일
* 매뉴얼 아이템의 골드화
- 이제 습득시 적용되고 사라짐. 마법책처럼.

* 아레나 스프린트에서 볼라이트닝 삭제

* 멥틱주문으로 타게팅하면 대상이 혼란에 걸릴 확률 표시

4일
* 힘 가중치 단순화
Previously, the strength damage modifier had three forms depending on your strength, from 0-9, 10, and 11+.

The formulas were almost but not quite the same, damage growth was fastest in the 0-9 range with +7.6% per str compared to +5.12% per str in the 11+ range. (Note: these +% figures are the max increase, since the algorithm includes random2 the average increase is about half the quoted numbers.)

The way this formula changes based on strength isn't great because either:
1) If it should affect attribute increase choices, it's a non-continuous breakpoint and therefore spoilery.
2) If it's too subtle to change attribute increase choices it's probably pointless.

Change the damage modifier calculation to use a single continuous formula for all strength values.

Rather than bothering with randomising the multiplier each attack, use a non-random formula suggested by advil. Strength modification is 100% at 10 strength, with +2.5% for each additional point of strength (-2.5% for each point below).

Derandomizing the formula reduces the variance of melee combat slightly (there are plenty of other factors that keep the rolls high variance), and reduces the maximum possible damage at high strength values.

The new formula increases damage delt at very low strength slightly in absolute terms, since at low strength the damage for most typical weapons is low enough that percentage changes aren't very noticable.

At very high strength the new formula doesn't exactly approximate the average benefit of the old formula (the extra .06 of a percentage point of bonus per strength point over 10 becomes noticable in absolute terms at around 40 strength).

For typical strength values the average damage tracks to within the measurement error of fsim, and the reduction in maximum damage (used as a proxy for variance) is small (for example at most 1 point of damage for a war axe at each of 10, 15, 20, and 25 str).

In exchange for this slight number shift, derandomization should make the effect of high strength easier to "feel" for players (at least those using high base damage weapons).

* Dex 스텝다운 제거
This commit removes the hidden stepdown to bonus EV from dodging. No change for characters with 18 dex or less, but high dex characters will receive better returns on dodging skill.

* EV 스텝다운 제거
This was an extremely hidden mechanic and I suspect that most players didn't know about it. No change for "natural" evasion 30 or less, which is everything except high dodging high dex characters and spriggans in practice.

* 스프리건을 좀 너프
Since they're the most likely to get out of control with stepdown removals, they lose 1 base dex and 1 dodging apt.
- 기본 덱스 1 감소, 회피 적성 1 감소

* 브로드 액스 딜레이 증가 커밋 취소
In the commit message for 8172001, its pointed out that broad axes are too cheap relative to battle axes for their rarity.

Part of the issue on later analysis of skill points required, is that *shields* were too cheap to add on in a penalty-free way.

Commit 0b64aa2 increased the shield penalty. To swing a 20 skill broad axe at min-delay with a kite shield now takes 21,000 skill points on a human; compared to the 13,950 of a battle axe. For an 18 skill broad axe the cost is 17,850; still more than the battle axe (since the trade is for -2 base damage in exchange for the defenses of a shield, it should cost more), but not 1.5 times more expensive.

To repeat some of the discussion from 0b64aa2, prior to that commit the cost to swing an 18 skill borad axe at min-delay was 12,200, cheaper than the broad axe.

Addressing the disparity via broad axe delay was one solution, but made broad axes a non-use nobrainer for small species (the same base damage is available 1-handed at 16 skill from a morning star for them!); broad axes are still pretty expensive for a small species, but should be more attractive at 18 delay when the shield changes are also considered.

- dc official App