Testing of Saotome Haidate, go!!!

[Wraith]

Ok, so I've heard many people who say that the relic legs do nothing when seigan is active. I want to believe that SE couldn't be that dense, so I've decided to parse in a controlled environment both with and without the legs, and see what results I get. By the way, I apologize if this has already been done and I didn't see it.

To start, my SAM is currently level 72, and I'm fighting the lesser colibris right outside al zahbi. The only thing I am changing from one parse to the other is switching from Saotome haidate to Shinimusha haidate. The following are my results:

With Saotome Haidate:
Hits taken: 11.9%
Evaded: 52.9%
Parried: 6.5%
Anticipated: 21.6%
Countered: 7.4%
Counters/Anticipations: 34.3%

With Shinimusha Haidate:
Hits taken: 16.1%
Evaded: 45.1%
Parried: 8.2%
Anticipated: 24.3%
Countered: 6.3%
Counters/Anticipations: 25.7%

So far, these numbers are taken from 350 total attacks in both settings. This isn't really enough to prove anything to me yet, but I'm thinking about finding a way to get 3 additional agility into my gear when I'm wearing the shinimusha haidate, because the 3 agility from Saotome seems to be skewing my results slightly. It seems kind of hard to believe that 3 agility can cause a 9% difference in evasion, though.

Any comments on my procedures and how to improve it would be very appreciated.

[Martinius]

this has long fascinated me as to whether in situations where i pull hate whether i should macro in saotome haidate when putting seigan/third eye up.

my advice is to buy a cheap lvl 54 agi +3 NQ ring since you're clearly not parsing accuracy, etc, so the loss of a sniper's will be irrelevant.

however, i think it's not the 3 agi that's making the world of difference, i think its the 'enhances third eye' as your anticipations/counters were up.

[RKenshin]

I'm not exactly sure what your controls were but I think it'd be hard to test in this sense do to parts of the Third Eye duration being wasted away from fight to fight. Unless you're doing this entire test on one mob (which is likely possible if you use an earth staff), killing a mob in the middle of a Third Eye duration will skew results as well as starting another fight in the middle of the duration. Same applies for not activating Third Eye as soon as it's ready during a fight. I'm not sure as to whether you did this though.

But, anyway, that just gave me an idea for an easier and more controlled way to do this test... Only downside being I haven't had the time to. If I get a chance I'll give it a shot though.

The idea is just getting like 30 Mandragoras from the Boyahda Tree, full damage reduction/defense gear, RDM with phalanx II, and using the results from that. Getting hit something like ~1000 times every few minutes would make it incredibly accurate and time effective. The only downside being (in theory due to the nature of Third Eye) that the anticipate %s would be relatively small so the difference wouldn't be as apparent. But perhaps it's better to just use a single mandragora if this is the case. As long as Third Eye was activated the second the duration was ready each time, it would be a more controlled test.

[Phraust]

Don't the mobs' behaviour change based on their HP% too? I vaguely remember reading something about mobs getting generally stonger/faster/more accurate towards the end of a fight. If you spent more time at the end of a fight, it could skew your results.

I'd try to drop your accuracy and attack and "fight" a mob with higher defense and low variance in levels, like a Steelshell.

[RKenshin]

Steelshells have a relatively high level range. IIRC level ranges tend to get bigger as you move up from level to level... With mobs you fight at level 1 having the lowest variation and mobs you fight at 75 having the highest.

Also, as far as I know the only thing that changes as a mob's HP gets past a certain low-point is how often they use TP attacks. In some cases they also gain new TP attacks and/or use certain TP attacks more frequently (like, Cerberus).

[Rasantath]

steelshells are 74-76. this makes it very easy to know exactly what level mob you are fighting.
(the 74's will con dc, the 75's em, and the 76's tough)

[Sesshoumaru]

steelshells are 74-76. this makes it very easy to know exactly what level mob you are fighting.
(the 74's will con dc, the 75's em, and the 76's tough)

Take some Bst friends along and have them convert all the mobs for you and have them do the checking.

Testing in this game really requires major effort from several people I say 2-3 bst converting mobs into EM for you will ensure that you have the most stable sample of mobs.

[Wraith]

I'm sorry, I should have been more specific. First of all, I started each fight out with Third Eye fresh. Also, I don't believe that the status of Third Eye when the mob dies will be of any consequence. That's a random error that should average with increased data.

[VZX]

steelshells are 74-76. this makes it very easy to know exactly what level mob you are fighting.
(the 74's will con dc, the 75's em, and the 76's tough)

They are 73-76
74 is the most likely DC you find, 73 is a bit rare

It's actually easier to tell their level than any other guinea crabs
if it's DC : it's either lv73 or lv74
if it's EM : lv75
if it's T : lv76

[Aireiya]

I'm sorry, I should have been more specific. First of all, I started each fight out with Third Eye fresh. Also, I don't believe that the status of Third Eye when the mob dies will be of any consequence. That's a random error that should average with increased data.

I think your biggest problems are:

1) You're not accounting for lvl ranges of colibri outside alzahbi
2) Seigan third eye seems to change in effectiveness depending on how soon you get hit after activating third eye. i.e. more anticipates when hit earlier. This might just be my imagination, but if it's true, then this is a variable that will be pretty hard to take into account when doing this test.

[RKenshin]

2) Seigan third eye seems to change in effectiveness depending on how soon you get hit after activating third eye. i.e. more anticipates when hit earlier. This might just be my imagination, but if it's true, then this is a variable that will be pretty hard to take into account when doing this test.

It's true.

[Wraith]

1) You're not accounting for lvl ranges of colibri outside alzahbi
2) Seigan third eye seems to change in effectiveness depending on how soon you get hit after activating third eye. i.e. more anticipates when hit earlier. This might just be my imagination, but if it's true, then this is a variable that will be pretty hard to take into account when doing this test.

Both of these are random errors. The more data I take, these should average out and not cause a large difference.

[XxSlusherxx]

this might be broken now with Saigan in the game, but I was always under the impression the haidate added a chance to counter, at least I rember hearing that at some point, not a sam myself.

[RKenshin]

That's basically what it did before Seigan was added. It added a chance to counter after the first anticipate.

After Seigan it's been a mystery whether or not it does anything/is broken or not until now, where it appears to improve your overall chance to continually anticipate.

[Wraith]

I think the most important thing to look at is the counter/anticipates. My thoughts are that it could increase the chance for you to counter which seigan and TE are up.

[eva00r]

You can fight lesser colibri at Laypoint. The level range is a bit closer, only 64-65 around that point, and I don't think you need to control monster level if you take large enough samples. Kinda ... 100 colibris per setup.

[Zekyr]

My question, once any point found to the haidate is found (if any), is do I need to leave it on after using third eye? Or can I just just bring it along, swap it in when I hit third eye (like AF1 head + AF2 hands for med), then swap back to TP gear? If all it does is add like +5% chance to counter, that's fine, I just want to know if it is required to leave it on to get the enhancement or if you can have it on for ability use then keep the enhanced version.