Enmity testing - Still Going!

[RKenshin]

Would it be possible for hate to be based upon dHP% rather than it being based upon level?
It would show a similar relation since as level increases so does HP.

Could try finding two mobs with different HP but the same level, problem would be doing the same amount of damage to both to compare hate values. Perhaps a blu with 1000 needles, or breath damage, or spirits within, something unmodified by mob stats.

We would also run into precision issues, because our smallest count is 1%hp, unless we can predetermine mob hp, or kill with a parser running to determine HP after the fact. With that, one would have to be careful of only using parse-able damage though.

Just another possibility I guess. Anyone have a reason why this is not plausible?

It's not HP based. You can see it pretty easily since something like Tiamat with 100,000 HP has very similar reaction to the same damage as say an IT++ Aern, even though they are similar level but drastically different HPs. I'll probably run some formal test just to disprove the idea that HP matters though.

Yeah I agree, and can definitely confirm it as consistent with my own experiences as a DD/tank for everything.

I was thinking about this topic especially the other day in Salvage, and payed a bit closer attention. Rampart mobs in Bhaflau are a great for seeing this, where one DD may take off like 50%+ of a mob's HP and you can still get it to look at you without doing more than 10-30% of the damage + a provoke on the same mob. And a provoke definitely isn't worth taking off 20-40% of a mob's HP. If a DRK took off 20-40% off of Tiamat's HP within a few seconds, it'd take a lot more than a provoke to get it off them.

This might actually just be easy to test on a mob's pet since they have low HP. Have a PLD Sentinel + Invincible + JA/spell spam (but not engage), and see if a melee can gets it off the PLD before killing it. On DL, a PLD will do the same but lose hate almost instantly once melees pass ~2-3k damage delt, despite this being only about 5% of DL's HP. Taking off 99% of a wyvern's/pets ~800 HP likely wouldn't get even a look though, if I'm remembering this stuff right.

But yeah, it'd be a rather safe bet over it not being HP% based. Or well, if this really needed to be confirmed.

[Jayhawk]

Remember though that possible hidden effects for items with long descriptions are not limited to Enmity-related effects. A hidden effect for Skadi's Cuirie could be anything like Haste+ or Triple Attack+. It might be useful to perform a whole set of tests for these items with long descriptions.

[shihei]

Ares Body Hidden Effect: Lowers Gil on hand.



JK


It really is Odin Perpetuation cost-1

>.>

[Lilswitche]

Looking at a few items, haven't checked a whole lot. Normally emnity+/- and haste+/- are usually the last stats given in the item description, and possibly "Enhances --- Effect" (although I need more verification on that), so to me those seem like the most likely candidates for stats that would not be displayed given there aren't enough lines in the description.

If anyone finds some items that show anything contrary to this lemme know so I don't need to go out of my way to test certain items.

[Kaeko]

The "not enough description space" argument may still be valid if we look at how those items appear in the JP version maybe?

Would seem odd that Marduk Feet get get the extra line and Marduk Legs didn't... Maybe they confused stats on Feet and Legs (in which case the feet do not have -4 enmity)? Just another thought.

[Zhais]

score, that's a damn nice thing to know :D

[Kirschy]

I figured I'd put this here in case you haven't tested it yet. When I was doing haste tests, I only messed a little with slow%, and thought it was additive. (ie; If you had 10% in haste gear, and a heca piece /w slow +5%, it would give you 5% total haste.) This happend to be wrong, and was hard to tell at low values. The following happend to be true:

New Recast = Recast * (1-haste%) * (1+slow%)

It might be worth it to check /w higher values of combined Enmity+ and Enmity- and see if it's additive or multiplied. Because of truncation it might not be noticble if you counteract +4 enmity merits with a piece of -4 enmity equipment... but at higher values it might skew results.

I just want to add that I'm increadibly thankful for your tests, and applaud you on the work. I know how time consuming it is, but all of this is very useful and appreciated by many.

[Not Kuno]

Maybe it was a typo and the -4 enmity on the shoes shouldn't be there. Anyone have those to test out?

[Grey Jorildyn]

All these typos would be easily avoided if the textboxes were autogenerated from the item code. SE has such archaic ways of doing things.

[Spekkio]

a thought if you wanted to get figures on the enmity values of the teleport spells: all you need to do is have someone without one of the past crystals cast that recall spell. assuming they're all the same, it should get you a figure for all of them. not that this is a USEFUL value, but it's something for the sake of completeness.

[Gustave]

Anyone tested Ares yet?

[Ashira]

Anyone tested Ares yet?

Yes, it was tested by Tekki, all controls given to him by Kaeko in thread and myself in IRC; it appears to have no Enmity bonus on body or legs.

Need someone to maybe test Marduk feet and see if it is indeed some sort of mistake and the rating was put on the wrong piece.

[nitsuj]

I don't want to ask for more work to these wonderful tests, but I'd be interested in seeing the results of testing for hidden Enmity effect on the "Assault Breastplate".

In fact, I think anyone on the Unicorn HNM scene would love to see it...

[Kaeko]

I don't want to ask for more work to these wonderful tests, but I'd be interested in seeing the results of testing for hidden Enmity effect on the "Assault Breastplate".

In fact, I think anyone on the Unicorn HNM scene would love to see it...

People on Unicorn think it has hidden stats? It's kind of tough to get one of these here.

[Ashira]

I don't want to ask for more work to these wonderful tests, but I'd be interested in seeing the results of testing for hidden Enmity effect on the "Assault Breastplate".

In fact, I think anyone on the Unicorn HNM scene would love to see it...

People on Unicorn think it has hidden stats? It's kind of tough to get one of these here.

I happen to know someone who has one, Kae.

[nekura]

I don't want to ask for more work to these wonderful tests, but I'd be interested in seeing the results of testing for hidden Enmity effect on the "Assault Breastplate".

In fact, I think anyone on the Unicorn HNM scene would love to see it...

People on Unicorn think it has hidden stats? It's kind of tough to get one of these here.

Not really "people on Unicorn," just one guy:

http://i103.photobucket.com/albums/m...XI/eastsig.png

(Screenshot courteous of Theone)

[LeifofKujata]

wha wha now? O.o

[Lordwafik]

I doubt there's pieces with hidden enmity on them. Marduk's -4 is so specific that it's pretty obvious there's a mistake in which piece got the -4 on it since the shoes claim to be -4, unless the shoes were also tested and I missed it.

[Mattthemaster]

I have a breastplate and haven't noticed any hidden effects but I can test some shit if someone asks.

Edit: No change in defense based on hp unless its totally hidden.

[Kaeko]

As far as the Marduk thing, I am personally leaning towards just a typo before marduk legs and marduk feet. Have not been able to test this. If some1 with marduk feet wants to help that would be great. I seriously doubt SE puts like hidden enmity mods on random items though.

I actually would appreciate some feedback on something. I'm trying to figure out the idea of "hate resets" and how to force one to occur without dying... I know the following ideas did NOT result in a reset...

The following series of events all did NOT result in a hate reset when the person doing these actions was not at the top of the hate list...

Zone > wait 5 minutes > Rezone
Zone > /shutdown > Rezone
Zone > Zone to 2nd zone > Rezone > Rezone
Zone > Mog House > Change Job > Rezone

If the person at the top of the list zones he'll get a reset. Any ideas on how to try to force a reset when you're not at the top of the hate list? I'm kind of particularly interested in this for another random AV strategy.

**************************

Also, on a completely separate note, I want to try to write a clear reasoning on how experimental testing works. I see a lot of people on other forums misusing the information and making false claims about the testing saying it's like infallable or exactly how things work in game.

I have a hard time sort of explaining the idea that it is not as simple as showing how things work, but more like developing as close a model as possible to what you actually see in game and that it is impossible to really know what's going on without looking at the actual code.

On the opposite spectrum of people - the ones that completely disregard the information based on some anecdotal personal experience about hate - I want to try to present a convincing argument on why experimental testing on a bunch of lvl 0 bees can tell you so much about what would happen at say Nidhogg.

I think the information presented as is is pretty dry and the arguments as to the true purpose and methodology of experimental testing is lost on a lot of people. I don't want people to think the information is perfect and exactly how the game works (it's just a model); on the other hand, I don't want some to disregard it simply because they think testing on some lvl0 bees can't tell you anything about things at lvl75. For anyone scientifically oriented, I'd really like to hear any convincing arguments on how this process works so I can use it maybe.