A small request: Kaeko's AV strategy needing critique~!

[Quizno]

Tell that to the people whom GMs threatened to jail for using terrain as an exploit.

Right or wrong this always makes me giggle. Only because "using the terrain to your advantage" was actually pointed out as something that you should try to do when the original (and now infamous) strategy guide came out upon NA release!

All praise Brady!

As a side bar, terrain exploits are routinely used on DL, on the Dynamis-Bubu boss, and in limbus (to name a few).

[Titanss]

I really like this strat. Now to find that many skilled whm's+rng's?

[ringthree]

The first thing I said when I initially posted on Kaeko's blog in response to this was that the strategy so far operates on several assumptions, for which I did not necessarily believe were well founded. Although all testing requires hypotheses which require assumptions, these assumptions should be based in knowledge that we have so far from empirical testing. Much of the hypothesis (any untested strategy is nothing more than a hypothesis awaiting experimentation) is based on the assumption that some of the most critical and important issues that cause the major problems with the fight are unimportant and are mostly dismissed out of hand or with only precursory discussion. The problem is that hypotheses based on assumptions that are not well founded may or may not work, but can cause indeterminacy in the results which limits the knowledge gained from testing the hypothesis.

For example, in handling Draw-in Kaeko says "It does have Draw In, but it only uses if someone attempts the LB "Wall of Justice" strategy." If this is the case, then that criticism is definitely unfounded, but more explanation would be beneficial in this case. For example, if it will only use Draw-in when it is completely unable to path to the target on the hate list then this would not be an issue, but if the Draw-in has a different trigger (e.g. it has some maximum pathing distance before it uses Draw-in) then that assumption would devastate the strategy. This is all before the issue of GM interference which, although Kaeko seems to look down upon, is a serious consideration when trying to develop a repeatable strategy instead of a trick strategy.

I do not know your experience with AV and testing it's regen, so I can not speak to it, but Kaeko again operates on a rather broad assumption that the repeated reports of massive Regen outside of ~30' are false. Kaeko does not even mention this Regen effect in his post and the only statement on it was in a response to a comment.

Kanican wrote: "I read those same reports and I frankly call bullshit on the Regen one. We've gotten AV to stop Regen before (waaaaay before this new SE video) and we were fighting it not anywhere close to the spawn area.

This whole idea of not leaving the spawn point is a bit weird to me. I read all the threads and to be honest, it is HEAVY anecdotal evidence at the very best. I just don't trust the testimonies where those were given in - not as a disrespect to those players, but just because there was so much going on at the time, it is very difficult to really show that it is true.

Nothing from our own fights suggest either of those to be true, so for now, I'm going on my own judgment on those."

I have to ask when the testing for this Regen was done by Kaeko, and whether it was done before or after the ninja fix to AV. The reason I ask is that my current operating assumption on the "out of spawn area massive Regen" is that it was added in the same ninja fix that prevented AV from being lockable and added Draw-in. If his testing was done before this time then his assumption may be flawed, and with the repeated reports of this much larger regen outside of an area in the proximity of it's spawn point I think the out-of-hand dismissal of this is not well founded.

Ultimately, and Kaeko readily concedes this, we do not know what will happen if this strategy does indeed work. And there is the rub. This is a holding strategy, it is not a method to defeat AV. It is a good, well thought-out and comprehensive holding strategy, but until we reach a point that the fundamental problems are resolved, it is nothing more.

One other unspoken assumption that I believe colors Kaeko's strategies is his tendency to implicitly, and perhaps unconsciously, always favor a mana-burn strategy. We all adapt to our surroundings, and I think this definitely influences Kaeko's strategic thinking usually for the better (his low-man strategies, for example, are excellent) but in this case I do not know if it provides any unique insight into this fight. The problems with AV are not changed except to perhaps provide a means to avoid one deadly move (and damage in general), but other issues remain. People have, even after the ninja fix, been able to reduce AV to the point where it will automatically use Chainspell and Manafont (assumed to be 79%). This strategy may, in fact, provide a way to avoid that one incredibly destructive move, but the fundamental problems with AV are not changed.

Ultimately, I personally do not believe that a "trick" strategy is necessary to kill AV. Is it bad to develop one? Absolutely not! I highly encourage it! But, in the end, I do not think that it is much more than the rest of the 100's of BG forum pages with suggested alternative strategies. This one is far more developed, and far more in-depth, but nonetheless it is the same. This is just my opinion, and it should never be taken as a suggestion to not test. We need more tests, more ideas and more observations, but knowing the limitations of our tests before they are conducted is not a bad thing.

[Linktri]

Our LS shadowbound Prudence for at least 5 minutes once. I don't know if that is a glitch or what, but we had people raise, get unweak and fully healed before Shadowbind wore off.

This strat could be used without the wall. In addition, if I'm not mistaken, AV can cast through walls. It was given draw-in to keep people from moving it all around the zone and exploiting the footprint trick. (Feel free to correct me if I'm wrong here.)

Also, if AV is reactionary, it is possible for it to use Benediction any time that a negative status keeps it from engaging its main target.

Nevertheless the concept behind this theory bears some merit and could explain how SE was keeping AV off the main group.

If the mages that were spamming 2hrs, were outside of the main alliance and somehow out of casting range but had hate on the main targets, could they pull enough hate to somehow interrupt AV? (Like the uggly pendant tonberry that casts death.)

[Lexa]

One of the few theories for AV that's come to light since the neverending pages of "HAY GUYZ I SAW A [insert job here] WEARING A [insert weird gear choice here] AFTER NOON ON EARTHSDAY FACING WEST IN THE SE AV VIDEO, I BET THAT HAS SIGNIFICANT MEANING!" that didn't make my brain hurt endlessly with the stupidity. I'd definitely be willing to give this a spin, but like others have said - finding the people, and having the people play their parts in sync, would be the real challenge.

[Suiram]

@ringthree, AV seems to only draw-in when--and only when--the person with the highest hate is physically located in "Northern Al'taieu".

@Linktri, while AV likely can cast through any walls that a player can cast through, one of the beautiful elements of this proposal is that hate switches before AV gets into a range where he could cast on the new target, so rather than moving further from that target along the line of his path between tanks in order to cast, he simply changes direction to return to a closer physical proximity

Also, questions of "what if he regens forever" and "what if he benedictions repeatedly" seem somewhat pointless here, since assuming that there are no prohibitive issues, at the very worst attempting this strategy will give people better insight into those very questions.

[Sci]

The strat is fine and would probably work if it wasn't for lag. Especially with the number of spectators AV would draw, AV will appear at max distance on a Ranger's screen one instant, and the next instant be past them and out of range. So unless there's something in the AI that'll prevent it from using Manafont when it gets near the WHMs, there's a good chance it might Meteor everyone on one side, unbalancing hate. Or an equally bad situation, if the RNG shadowbinds AV out of cure hate range of the Cure V target mule. Then there's no way besides AV damaging the WHM to bounce back hate. So in theory it seems it's a sound way to beat AV, but I just doubt it can be executed.

[Kaeko]

Thanks for making a thread Ashi

Ok to answer a couple of questions...

What happens if both WHMs in this hit the hate cap?

First off, if 2 WHMs actually manage to hit the hate cap on this (requires 25 successful back-and-forths by AV), they deserve some sort of medal. To answer your question though, you would be ok. The mechanics of hitting the cap work in a sort of "tie rule". If 2 players are at the top of the hate list having the same TE, the mob attacks the closer player. If both players are at the same distance, it attacks the one that most recently did something.

Say both WHMs are at 10,000 CE (cap). WHM across thew all throws out a Cure 5 and is now at 10,000 CE and 600 VE. His TE will be higher, so AV will switch directions. Once the TE wears and they are back at 10,000 CE each, I BELIEVE the check for distance is conducted first before the "who performed the last action", but I'm not 100%. Either way, just keep the WHM you want to avoid getting hit away from the wall so the other is closer and you'll be fine.

Regen when out of spawn point?

There was a report about AV having massive regen outside the 30' spawn point. I personally believe that is not true and base the strategy around that assumption. I haven't fought AV recently so I cannot say 100% for sure either way; HOWEVER, I'm sure most old school BGers remember the original kill strategies - most kills were done outside the spawn point area. I have personally be a part of a kill attempt and we did break regen outside the spawn point. Unless SE ninja patched this, I seriously doubt regen jumps outside 30'.

Benediction?

I have no suggetions on how to fix this. That is for the people suggesting things like to melee with the correct element of day staff, spamming 2hrs, fighting naked, etc. This strategy was meant to bypass the bad 2hrs which wipe everyone. I believe you can at least get AV to 2nd or 3rd form (79%/59%) with this. As the old BoL and Apathy attempts showed, at 79% it starts randomly spamming meteor even without Manafont. This fixes that issue. It doesn't fix all that will come up later but 1 fix at a time for now in my opinion.

Is this an GM-able Exploit?

Well it's definitely not how the dev team killed it so it wasn't 'meant' to be done this way; however, I don't think it's a true that this is a true exploit. The mob is allowed to fight back, and will wipe everyone on the order of seconds if you do not time things correctly. The mob can still reach you, hit you, etc. if you sit still. This was not true for WoJ or the Footprint.

Can you be GM'd for it? Give me enough time and I can always find you a GM dumb enough to ban you for just about anything they can't comprehend. Do it at your own risk I guess? I personally don't think it's our fault mobs make incorrect decisions around terrain though. That's like getting GM'd for taking advantage of the fact Kirin sucks at running around pillars.

Kill Speed?

Ruk thinks the kill method I mention is too slow. Having never gotten AV past like 88%, I wouldn't be able to tell you personally. At this point, I think any concrete, workable strategy that can keep you alive indefinitely and maybe even help you survive past the 79%/59% AI changes is progress.

[Therin]

I believe AV has 50% magic damage reduction similar to CoP wyrms. This would mean that Nether Blast will do about 250 damage. With ten SMNs you're looking at 2500 damage per minute. You're right in saying that if you can completely dismantle the regen, you can in theory kill anything - but I don't think anyone have ever completely dismantled AV's regen. Lower it, yes. Remove it, no.

Basically, in one minute, you have twenty ticks of regen. It's probably safe to assume that AV, as enfeebled as it can be, has about 50/tick regen (the same speed as Tiamat and other CoP wyrms). Non-enfeebled, I would imagine it's closer to the order of 250/tick. Basically, with ten SMNs you're looking at 1500 damage per minute, which is, at best, a 100-minute fight. This is assuming you can safely say that AV's regen is 50/tick and that nothing will screw up. One screw up and you're finished.

Sounds plausible, but ridiculously risky and almost impossible to pull off without something happening. 100-minutes of perfect play (and this type of fight would be astronomically more difficult than a 6-man Jorm or Tiamat or whatever, where simple screw ups can be fixed VERY easily with a crew of intelligent players).

[Kaeko]

Ah another thought...

I ready something about AV's 2hrs being on like a 1-1.5 minute timer? You can time thigns with Shadowbind to keep AV away from you during 2hr access if this is true and reproducable. If you can do that, you can just have the WHMs at the wall at all times and spam cures directly at each other. This would drastically simplify the strategy since you don't have to Bind at a precise range of the wall, just cure bomb players. Bind could be used at any time in this.

This is only if you can properly time when 2hrs go off. Bad one goes off, you Bind twice. Good one (like PD) goes off, you let it go?

Also, my biggest concern with this was how difficult it would be to pull off in practice. I personally think in theory it works, its just about how hard it is to pull off with actual people.

[Therin]

It's certainly possible in theory, and it's a GREAT strategy. I was utterly amazed when I read it at how much thought and planning you put into it - it's very inventive. Personally, though, I think it's nearly impossible to pull off in the real word. In theory you can probaby duo or trio Cerb, Khim, Proto-Omega, and a slew of other things - but as far as I know no one has ever successfully done any of that. ;P

[Suiram]

I think even with the mules you might not need to rely on shadowbind, if you can get the timing down. Depending upon how far you can space a WHM from the mule on their side that keeps the mule within 25' of AV's path to him/her, you might be able to time cures so that they land in that interval. Then if timing screws up and a WHM does take damage, you can then default to shadowbinds when he reaches the other side to give the WHM who lost hate enough time to cure 5 the mule as much as is needed to restore balance. The only issue in that case is determining how much is enough without going over, since I'm fairly sure that /assist during a bind won't adequately report who the target will be after bind wears.

[Seditedi]

I think you can still /assist a bound mob and get the correct target.
At least I've done it before while trio'ing Mother Globe, since bound mobs still turn to face their current target.

[Suiram]

I think you can still /assist a bound mob and get the correct target.
At least I've done it before while trio'ing Mother Globe, since bound mobs still turn to face their current target.

Well the problem is there is "regular target" and then there's "bind target" and I'm not completely clear on the rules that decide when the two coincide. For instance, if person A has the most hate, the mob gets bound, person B steps into melee range for one moment, then regardless of the distance between B and the mob, he will remain the target for the duration of the bind (or if someone else does something to become the bind target). However the moment bind wears, the target reverts to person A.

Of course that particular example wouldn't be an issue for what I was describing earlier, but I don't think that's the only way in which bind screws up hate for its duration.

[Seditedi]

Yea, that is correct, but I'd hope people fighting AV wouldn't be stupid enough to walk next to it while it's bound.

[Suiram]

Of course not lol But I guess what I'm asking is does the target change *at all* while bound except in that situation? For instance if two BLMs are fighting something, BLM A has the most hate by a slight margin, and binds the mob, BLM B nukes the mob enough to take hate but doesn't break bind, will the target be A or B for the duration of bind assuming no other actions are performed? I want to say it'll be A, remembering in particular when we used to manaburn Azure Experiments for divisor ring, but I can't say that with complete authority.

[Gergall]

I left this as a comment in the LJ:

"Another thing that you have to be ready for is when AV uses Virelai on alliance members and they go straight for the WHMs."

Given everything else that's going on, this would not be insurmountable. But if you neglect to consider it, it could break you.

[Spekkio]

I left this as a comment in the LJ:

"Another thing that you have to be ready for is when AV uses Virelai on alliance members and they go straight for the WHMs."

Given everything else that's going on, this would not be insurmountable. But if you neglect to consider it, it could break you.

wouldn't it only do that on the person who's top on its hate list? i'd imagine it's more likely AV would SV and never get a chance to cast, same as meteor.

[Suiram]

Okay so I just went and double checked, Person A binds a mob, Person B vokes, neither are ever in melee range at any point, person B does not become the target until Bind wears, I think it's safe to assume it'd be the same for cures. Again this doesn't really pose a problem for this strategy or the modification I proposed above, only for reestablishing a balance between the WHMs should something go wrong in either case (a bind too close to the WHM in the original strategy, or a mistimed cure that allows a WHM to take damage in one that doesn't rely so heavily on binds).

[Shinsei]

This was a very nice read, my LS might try this out depending on our turnout, I find it even more interesting cause before I read this, our leaders decided to have a SB rotation thing for his devastating 2hrs, so he can't get close to us in order to wipe us.

But that wall thing is pretty interesting.