Awesome testing done on enmity

[Armando]

Re: Awesome testing done on enmity

No problem. If you manage to get some solid numbers for the damage-to-enmity relationship, consider studying Spirits Within to see if it follows the same rules. I specify Spirits Within because you can easily reproduce its damage each and every time. You could also study Ranged WS, since with capped R.Attack they always do the same damage.

[Ryoii/Nonomii]

Re: Awesome testing done on enmity

That is a great idea. There might be some additional enmity generated by a WS which is separate from the raw damage. Spirits Within would also allow for checking if some VE is generated along with CE. I believe testing a WS with additional effect stun would also be beneficial.

[Lunaryn]

Re: Awesome testing done on enmity

I'm actually curious about how Steal stacks up in this, seeing as it is, in my experience, the fastest way for a THF main to secure hate in the absence of traditional hate tools like Provoke or debuffs (or an opportunity to Sneak Attack). Maybe when I get WHM to 28 I'll see if I can't arrange some tests myself...

[Armando]

Re: Awesome testing done on enmity

Also, I'd like to point out that by testing against most WHM or RDM mobs, it's now possible to check if:
- Hitting the mob for 0 generates hate
- Hitting a Blink shadow generates hate
- Additional effects generate hate

It'd also be interesting to check if Spikes give hate, and if hitting a mob's Utsusemi shadow will reduce your enmity against it. All of those are pretty doubtful, but none of those tests would take particularly long, and it's not good to make assumptions either way.

Oh, and if you want to force misses against a mob for any reason whatsoever, try Blinding Potions. Those things are insane. They give over -150 Accuracy. I couldn't figure out the actual number.
------------------------------------------
Bump. Just pointing out that we have some confirmation that VE is affected by +Enmity.

(Anonymous) wrote:
Nov. 13th, 2007 04:57 pm (UTC)
VE and enmity gear

Noticing that voke is 1800 and depletes in 30s at a rate of 60E/sec it seems that voking a slept mob is useless in an attempt to build hate up, and you'd be better of just waiting until its about to be woken. However if enmity gear does effect the VE as it does CE voke should last longer. Test was a smn/whm pld/war drk/war smn earthen wards the party, and pulls with silence, DRK with 0 enmity, PLD with 37 enmity. 1.37 * 1800 = 2466 or a difference of 666 VE, aprox 11 seconds.

Test 1 (no enmity gear)
Smn pulls << takes hate
Pld vokes << takes hate
1 second later Drk vokes << takes hate

Test 2 (+37 enmity gear)
Smn Pulls << takes hate
Pld voke << takes hate
1 second later Drk vokes << doesnt take hate

Test 3 (+37 enmity gear)
Smn Pulls << takes hate
PLD voke << takes hate
11 seconds later DRK vokes << doesnt take hate

Test 4 (+37 Enmity Gear)
Smn Pulls << takes hate
PLD voke << takes hate
15 seconds later DRK vokes << takes hate

Clearly Enmity gear affects VE aswell, meaning with +37 enmity, voke will last 11 seconds longer then its timer, making it possible to build hate on a slept mob by voking it every 30s.

Thanks again for the great info, cant wait to see how you guys plan on doing dmg ^^

-----

kanican wrote:
Nov. 13th, 2007 08:36 pm (UTC)
Re: VE and enmity gear

The previous test on Enmity Gear was only "the effect of Enmity Gear on CE". We've found results similar to yours and Enmity Gear probably does the exact same to VE as well. We know it does not affect the decay rate, but it will increase/decrease that hate spike.

Will probably be in the next post whenever that is.

So we can just consider +Enmity to be a simple +% increase to all actions.

Also, it seems Foe Sirvente reduces the rate of decay of VE. It's speculated that Guardian has the same effect.

[Taskmage]

Re: Awesome testing done on enmity

Part V is up

[Ryoii/Nonomii]

Re: Awesome testing done on enmity

Taskmage and I performed some enmity testing using the Crawler's Nest NMs. Currently we only have rough estimates due to time constraints. I doubt we will get a chance to do this again for a week or two, so I wanted to go ahead and share our results.

Rough Damage Dealt CE Multipliers

Lv1 Enemy: Damage Dealt x ~13.33 = CE Gained
Lv45 Enemy: Damage Dealt x ~2.42 = CE Gained
Lv50 Enemy: Damage Dealt x ~2.18 = CE Gained
Lv55 Enemy: Damage Dealt x ~2.0 = CE Gained

Here is a decent formula which roughly fits the data above:
CE Gained per Damage Dealt = [85/(Enemy Level+23)]^2+0.85

I've yet to see anything squared in a formula for FFXI, but a linear relationship clearly did not fit. It would help greatly to get a rough multiplier for Lv75 Steelshells. At the very least, this little bit of testing has shown that CE gained per damage is dependent on the level of the enemy.

I would recommend the following method for obtaining rough numbers:
RDM #1 builds TP for a weapon skill
Both RDMs buff with Stoneskin and Phalanx
Character #3 spawns the enemy
RDM #2 sleeps the enemy
RDM #1 uses the following macro:
/attack
/wait 2
/ws "name" <t>
/wait 2
/attack
RDM #2 sleeps the enemy again to let VE diminish
RDM #2 slowly builds CE until they pull hate from RDM #1

From our limited testing without using weapon skills, it looks as though Spirits Within does not add an additional amount of CE beyond its damage.

[Armando]

Re: Awesome testing done on enmity

Very nice. Those multipliers make a hell of a lot more sense than only needing 135 damage to produce 1800 CE, lol. I'll be looking forward to your results on Steelshells.

[Taskmage]

Re: Awesome testing done on enmity

Added part VI. Actually went up on the blog two days ago.

[Ryoii/Nonomii]

Re: Awesome testing done on enmity

Enmity from Damage Dealt Update
Volatile Enmity ~= Cumulative Enmity x 3

This is only a rough estimate based on my testing with Lv50 Drone Crawlers. I was using Sleep and Dispel to build ~2560 CE. Then using Ryoii I would engage, Penta Thrust, and disengage using a macro that prevented additional damage. I would then use a timer to check how long Ryoii held hate and keep track of how many Shadows she lost. I tested previously that no one gets the extra claim enmity when fighting the crawlers spawned with Rolanberries.

Example:
Nonomii casts Sleep (320 CE)
Nonomii casts Dispel x 5 (1920 CE)
Crawler wakes up
Nonomii casts Sleep (1920 CE)
Nonomii casts Dispel x 2 (2560 CE)
Nonomii casts Stoneskin and Refresh (2562 CE)
Crawler wakes up
Ryoii uses Penta Thrust (981 DMG x 2.18 ~= 2138 CE)
Timer started
Ryoii looses two Shadows (2088 CE)
Ryoii casts Utsusemi: Ni (2089 CE and 300 VE)
Ryoii looses two Shadows (2039 CE and 300 VE)
Ryoii casts Utsusemi: Ni (2040 CE and 600 VE)
Ryoii looses hate to Nonomii
Timer stopped (108 seconds, 108 * 60 = 6480 VE)

Calculation:
When Nonomii Regains hate: Nonomii's Total Enmity (TE) ~= Ryoii's Total Enmity (TE)

Nonomii's CE + Nonomii's VE = Ryoii's CE + Ryoii's VE

2562 + 0 (decayed while enemy was asleep) = 2040 (CE left after Shadows are lost) + 600 (VE from casting Utsusemi) + VE from Penta Thrust - Time Passed x 60 (VE decays at 60 units per second)

VE from Penta Thrust = 2562 - 2040 - 600 + 108 * 60 = 6402

Ratio between VE and CE from damage dealt = 6402 VE / 2138 CE ~= 3

Comment:
Due to the fact that I was using SAM/NIN and RDM/WHM, I was only able to get a rough estimate. I performed the test 5 times, and all tests yielded ~3 for the ratio between VE and CE from the Penta Thrust damage.

[Ryoii/Nonomii]

Re: Awesome testing done on enmity

Enmity from Damage Dealt Update

Lv75 Steelshells: Damage Dealt x ~1.53684 = CE Gained

Here is an updated formula which factors in the data above:
CE Gained per Damage Dealt = [97/($A2+41)]^3+1

I'm don't guarantee the value to be exact, but I believe it to be fairly accurate. I was only using Ryoii (SAM/NIN) and Nonomii (RDM/WHM). I had Nonomii pull with Sleep, Dispel x 2, Sleep, Blind, Paralyze, and Slow (1483 CE including Claim). Once the enemy awoke a second time I had Ryoii engage, Penta Thrust, and disengage. I then timed how long Ryoii held hate and counted how many Shadows she lost to calculate her total enmity (TE). I then divided Ryoii's TE by 4 to get the CE gained by her damage dealt from Penta Thrust.

[Armando]

Re: Awesome testing done on enmity

Awesome job, Ryoii. That means that against a Lv.75 mob, a 300 damage WS generates the same TE as Provoke. That explains why my Vorpal Blades and Swift Blades (which only do moderate damage; bit under 300 on average) help me hold hate so well.

Think you could look into the CE+VE generated by cures? It'd be interesting to see the ratio of cures vs damage vs WS.

Oh, another thing to consider - assuming no resists, you can get the same number over and over again using magical WS such as Red Lotus Blade or Raiden Thrust. Don't count on doing that with G.Katana though since those WS are mixed damage, not magic damage.

[Ryoii/Nonomii]

Re: Awesome testing done on enmity

I was a little shocked by the huge amount of VE generated by damage. It certainly supports the argument for more DD oriented PLDs and NINs for tanking later in the game. All my Lv60+ PLD tanking experiences were prior to sushi and fighting VT enemies, and things have certainly changed to make the tank's damage dealt a huge factor in maintaining hate today.

Another player has already produced rough numbers for cures cast on a Lv75 character: http://wiki.ffxiclopedia.org/wiki/Enmity_Numbers. They show a 200 HP Cure III to produce about 145 CE and 880 VE when cast on a Lv75 character, which makes cures far less efficient than Flash, Blind, Bind, Poison, and Barelements.

Armando, how hard was it for you to spam Blind as a PLD/RDM? Have you considered giving this combination a try again? I just see Berserk and Defender going to waist these days, and I would think Phalanx and Stoneskin are so much more useful.

PLD/WAR
Flash (180 CE & 1280 VE, 25 MP, 0.5 sec), Provoke (1800 VE, 0 MP, 0.5 sec)
180 CE and 3080 VE = 3260 TE in 1 sec for 25 MP (Double Attack will increase enmity, but not very helpful with Joyuese)

PLD/RDM Option # 1
Flash (180 CE & 1280 VE, 25 MP, 0.5 sec), Blind (640 VE, 5 MP, 2 sec), Bind (640 VE, 8 MP, 2 sec), Barelement (300 VE, 6 MP, 0.5 sec) x 2
180 CE and 3160 VE = 3340 TE in 5.5 sec for 50 MP (closer to 5 sec of casting time due to Fast Cast)

PLD/RDM Option # 2
Flash (180 CE & 1280 VE, 25 MP, 0.5 sec), Blind (640 VE, 5 MP, 2 sec) x 3
180 CE and 3200 VE = 3380 TE in 6.5 sec for 40 MP (closer to 5.9 sec of casting time due to Fast Cast)

PLD/RDM Option # 3
Flash (180 CE & 1280 VE, 25 MP, 0.5 sec), Barelement (300 VE, 6 MP, 0.5 sec) x 6
180 CE and 3080 VE = 3260 TE in 3.5 sec for 61 MP (closer to 3.2 sec of casting time due to Fast Cast)

PLD/RDM Option # 4
Flash (180 CE & 1280 VE, 25 MP, 0.5 sec), Dispel (320 CE, 320 VE, 25 MP, 3 sec) x 3
1140 CE and 2240 VE = 3380 TE in 9.5 sec for 100 MP (closer to 8.6 sec of casting time due to Fast Cast)

PLD/BRD
Flash (180 CE & 1280 VE, 25 MP, 0.5 sec), Threnody or Requiem (100 CE & 240 VE, 0 MP, 2 sec) x 5
680 CE and 2480 VE = 3160 TE in 10.5 sec for 25 MP (large loss of time for combat without any benefit over /WAR)

RDM/WAR (with Ice Spikes for Paralyze)
Provoke (1800 VE, 0 MP, 0.5 sec), Blind (640 VE, 5 MP, 2 sec), Slow (640 VE, 15 MP, 2 sec), Barelement (300 VE, 6 MP, 0.5 sec)
3380 VE in 5 sec for 26 MP (closer to 4 sec of casting time due to Fast Cast)

[Armando]

Re: Awesome testing done on enmity

Another player has already produced rough numbers for cures cast on a Lv75 character: http://wiki.ffxiclopedia.org/wiki/Enmity_Numbers. They show a 200 HP Cure III to produce about 145 CE and 880 VE when cast on a Lv75 character, which makes cures far less efficient than Flash, Blind, and Barspells.

Hm. Although I have no doubt they're less MP and time-efficient, I don't think we have enough info to interpret those numbers yet. I mean, they're probably true, but I don't think we know for sure that the enmity generated from Cures depends on the level of the target and not the level of the mob witnessing the act, do we?

Armando, how hard was it for you to spam Blind as a PLD/RDM? Have you considered giving this combination a try again? I just see Berserk and Defender going to waist these days, and I would think Phalanx and Stoneskin are so much more useful. Also I believe you previously used Blind and Sleep, but it appears that it should be Blind and Barspells for the most MP efficiency.

It wasn't hard by any means. Holding hate was the easy part, provided I had Refresh. However, I did so much casting my damage was totally awful. I'd consider returning to the combination if people started fighting plain old IT mobs in traditional SC+MB parties. However, as things are now, I can hold hate just as well with /WAR and do enough damage to kill the mob faster to boot.

Previously, I fell back on Sleep and Bind, using Blind as a filler. My logic was that I had a gut feeling Sleep and Bind were getting me the most hate (I had good results casting those when I lost hate; now it's clear that the three of them get practically the same TE though) and Blind is so cheap and low-recast that I could spam it as needed in between. I tried Gravity a couple of times but it didn't impress me and it was slow to cast and expensive, but I dropped it. (Of course, now it's pretty obvious that its hate wasn't as high as you'd expect.)

I could see Barspells serving as a short term hate spike. However, while I agree that I could drop Sleep, I'd only drop it in favor of Dispel (which only costs 1 more MP, has a 10 second recast, generates the same CE and slightly more VE.) I think Dispel would be valuable when used in moderation simply because it generates CE.

Also: Even on /RDM, I used WS as a hate tool. RLB and Seraph Blade are actually slightly more efficient at 200 TP than at 100 TP, and 300 TP is just as efficient as 100, so I could hold TP as much as I wanted and I wouldn't be wasting damage. At 200 TP and with /RDM's MAB, RLB could do 200-250 damage in the 50's, and if Magical WS generate the same TE as physical ones, then that's nothing to sneeze at.

[Ryoii/Nonomii]

Re: Awesome testing done on enmity

Thank you for your perspective based on experience Armando. I've been itching to level PLD again some day, but my life just doesn't allow for it. PLD and RDM are the jobs I keep unleveled on Ryoii in case I find a RL friend who wants to play FFXI. BTW, congratulations on getting your Joyuese.

[Armando]

Re: Awesome testing done on enmity

No problem, and thanks.

So that this post isn't just postcount +1: would be interesting to use PLD/BLM from 40-49. Particularly because of Ice Spikes.