Diablo 4 Boss Timer
Audio Alert:
Diablo 4 Boss Timer

Some thoughts on javazon setup


Well-known member
Guide Author
Mar 13, 2019
Original Thread by @ffs can be found here.

While javazon was always one of my favourites, except for running Pindle she's not a very good MFer or rune finder, which is why she didn't see much playtime anyway. However, I played her a lot since the third round of RFL, and had a lot of fun with this play style with her specifically for cow runs. It revolves around fast teleporting (for Amazon standards) and a lot of herding, aimed at killing only huge mobs to maximize the potential of the LF/100% pierce combination. Since she was below level 94 and couldn't deliver reliable numbers in terms of efficiency at the time, I continued to play her in order to get some numbers. In all this running, I've tested a lot of different setups and crunched a bit of numbers, and I realized that at least for me there still were many things to learn regarding this build.

Now, it's not like I have any crucial new information to present or something. But having played around with this play style for a while and having implemented some tweaks to make it work better, I just wanted to share a few thoughts on this. I hope this is useful to others, and maybe to get some tips on improvement from others as well. :)

First maybe some examples of short runs in different areas to get the idea:

All runs are /p7, gear used in all videos is shown at the end of last vid. As you can see this is running around with totally negative resists and no block on javelin switch. Also no CtA in latter two runs (actually not necessary for cows either). CS and Pits are not really optimal maps, but I'd like to keep my cow map for now.

So the setup I eventually settled for basically consists of these components:
  • 99% FCR teleport switch (with decent resists and ideally 86% FHR)
  • Defiance Merc with Infinity (though eth bugged Fortitude could be considered key as well)
  • As much damage as possible on javelin switch (ignoring resists) to deliver dead monsters quickly and keep MPK/LPK going
  • Enough life to be comfortable without CtA unless pre-buffing (1400+ for me, might be more or less for others)
  • Sufficient mana upkeep (>10 MPK/ML. MPK is better but combinations of MPK and ML work fine)
  • 35% or 55% IAS (depends on available gear and other stats)
  • Build is vit only, fully synergized LF/ChS with Pierce and D/A/E
Play style
I guess the first thing is to consider how essential teleport is for her. It's used for a number of things that are crucial for efficient and fast running: Herding where possible to maximize DPS as well as MPK/LPK, positioning her in order to create a good pierce line and to keep Infinity merc with her, as well as generally getting around more quickly. I've come to think that speeding up all of those things by hitting the 99% FCR breakpoint on switch is a huge improvement over 68% FCR, which is ultimately even worth dropping CtA for. Most of the time is spent on switch for getting around and in particular positioning, with defensive stats to keep her safe. Javelin switch just needs to kill as fast as possible, before switching back to teleport. Basically similar to Pitzerker in that regard. Pierce is her God, and the idea is to play in the spirit of this by bringing her into favourable positions quickly, helping her kill speed as well as survivability. Since she always benefits from leech and LPK/MPK when attacking, she is actually not that vulnerable while doing so, and after some adjustments to her setup, I've finally bought into the notion that defensive stats on offensive switch are in fact not necessary. So in short, it's about playing her with a fast (and safe) teleport switch, and a completely offensively oriented javelin switch to kill stuff instantly.

Gearing considerations
Getting the more standard (or "pretty much ideal") gear out of the way first:
Merc: Infinity in whatever base. Ebug Forti is also extremely useful in keeping him alive since CtA is usually not used, but Treachery is an alternative and will trigger the aura quicker for shorter runs. For headpiece I'm using Andy's + RJoF, but Delirium might be even better for Pits for added safety.
Aura is Defiance (see below).
Headpiece: Griffon's Eye - most practical socketing I think is IAS, since this (1) can never be wrong and (2) allows for smooth transitions and flexibility between 35% and 55% IAS setups, not having to use HLW for 35% setups. I'm using 15%IAS/3 MPK which I consider pretty much ideal (since not playing with 1.07 rings). Socketing RBF and getting IAS elsewhere is the alternative, of course.
Weapon: Titan's, upped especially when relying (at least in part) on ML
Armor: Enigma
Gloves: 3/20
Belt: Razortail
Boots: Silkweaves (again that's without 1.07 mpk rings). These are awesome as they don't only provide crucial MPK but also defense vs missiles which is handy in mitigating her weak spot vs ranged mobs.
Rings: One Raven is required unfortunately. Second ring depends on the solution for mana upkeep, generally either SoJ or ML/MPK/stat ring. If you have/want to use 1.07 MPK rings, these are ideal as they allow for the most flexibility in other gear choices.
Switch shield: Spirit Monarch, FCR roll is irrelevant.

For the maybe more interesting points.

Merc aura
I never saw this recommended as first choice, but I really think Defiance is the best option for her, especially without CtA. Upon switching to Defiance for cow runs to help keep merc alive, I just didn't bother switching back at first, but soon realized that this is actually a big reason why this aggressive gearing/play style actually works in almost every area of the game. Considering CtA is not used (unless pre-buffing where it makes sense, e.g. longer runs), the extra defense from Defiance is felt a lot. It reduces chance to be hit by ~25% in my case. Moreover, other auras are of little use. Holy Freeze is counter productive in herding and creating good pierce lines, and the advantages of Might when it comes to leech (which is less effective than MPK/LPK anyway) or merc helping with unbreakable immunes are small, and ultimately not worth it in comparison to the survivability boost provided by Defiance.

Since defense is reduced to zero when running (or more precisely, the game's chance to hit formula is not applied), a further little thing to make the most of Defiance and improve safety: When facing a potentially dangerous situation, it makes sense to walk (instead of running) when doing some side-stepping, in order to keep her defense up. This is a small thing but it adds a lot of safety. (For those interested: Good information on defense in general in this guide.)

Teleport switch
To speed up runs, which is kind of the point here, I've settled for good for the Spirit + Wizardspike switch. Wizzy delivers 99% FCR and adds massive resists on the switch, which is where they have the most utility. Also, while not essential, 86% FHR on switch helps, as without CtA it does happen from time to time that she is put into hit recovery.

On CtA: Of course, pre-buffing is always an option for longer runs. But personally, at around 1450+ life, I don't consider CtA to be necessary anymore due to the combination of 99% FCR, good resists and FHR on switch (where most of the time is spent for positioning and getting around), and in particular using Defiance merc. In areas with elemental damage like CS, Wizzy was actually quite a bit safer than CtA due to the resists and faster teleport. Fewer potions used, fewer critical situations in general. CtA might seem more appealing in areas with ranged mobs and less elemental damage (e.g. Pits), in fact cursed and/or fana ranged bosses are one of the few remaining dangers for this character. But 99% FCR means quicker re-positioning and combined with Defiance adds plenty defensive tools, which are actually quite safe to rely on.

Attack switch
I might be late to the party in this regard... but at least with this play style I've realized that indeed defensive stats are simply not needed on the attack switch. Paying attention to when she takes substantial damage exactly, it turns out she is rarely under much of a threat at all when attacking. This is especially true with regard to elemental damage sources, greatly diminishing the utility of resists on javelin switch. In general, it's either LF from safe distance or ChS versus only a few mobs or bosses. While attacking, she always benefits from LL from Titan's and LPK from Enigma, an amazingly useful stat that is actually crucial for this. However, when not attacking, she's teleporting, and when teleporting, having Spirit and Wizzy means plenty resists and 55% FHR, hitting the 6 frames breakpoint and putting the 86% (5 frames) breakpoint within comfortable reach.

For almost every area I tested with her, she can hang around with totally negative resists on attack switch. LL (Titan's) and LPK (Enigma) is more than enough to keep her alive comfortably as long as she's attacking and damage output is high enough to kill quickly.

As a consequence, it simply seems locigal to gear her very offensively on javelin switch, which in turn means it's kinda difficult to justify using anything but a 4x RBF Monarch. JMoD is of course ideal but other magic (e.g. +life) or even plain 4os are way closer second choice than I previously thought because block on main switch is actually rather luxury instead of necessity. Though I don't have a 3os HHG, that's a very good option of course and works well with Defiance due to its defense boost (in particular vs missiles). However, not sure it's worth it to miss out on the facet which is equivalent to roughly up to +3 skills. 2x RBF upped Moser's (my previous all-round choice) is also missing out on damage for defensive stats that are nice, but simply don't have a lot of work to do on the offensive switch with this kind of running. An interesting option is Lidless Wall + RBF because it enables setups that don't require many facets while together with Silkweaves take care of your mana needs completely, which means more damage from other slots. Damage output of such setups is between 2x RBF and 3x RBF shield. I've included a few calculations of DPS output of different setups below.

Ultimately, the point is that going full offense on shield is just a logical consequence of when and where defense is needed, and not a particularly "risky" choice as it may seem. It's again somewhat similar to Pitzerker: Here too, resists on kill switch are ignored. Only instead of howling mobs away to be safe, the javazon kills them. :)

Challenges for her
Frankly, I am still surprised how rarely she gets into real trouble.

First, WSK: From what I tested with her, this is the only area where this approach doesn't work without some adjustments. Here she needs to have very good lightning resists and/or absorb because of those nasty souls. But running WSK is not very good for javazon in general, apart from Baal runs when going straight to Throne Room. There are several mob types in WSK that are both ranged and lightning immune, not the best combination for her, some of which are moreover not broken by Infinity. She can do full WSK, but she can't play to her strengths and I'd just go with other characters for this task.

Otherwise, the most serious threat are ranged cursed/fana/might boss packs. They can dish out considerable damage from distance and are not easy to herd because they don't run towards her. From the areas I think should be run with her, these hang around in Pits in particular. However, resists wouldn't help here at all, though block and/or CtA would indeed. But it's not like she's in danger of being one-shot or something, and as stated above, the stats on switch and increased defense through Defiance add quite a bit of safety. After adopting that little walking instead of running trick as well for particularly ugly packs, I basically considered this problem solved. Sure, it leaves more work to do than just tanking with CtA and/or block, but it works perfectly well, and actually makes runs more fun giving me something to do instead of just spamming LF. For sake of comparison as far as Pits goes in general, running with her on /p7 here is definitely safer than running Pits on /p1 with Pitzerker.

The lack of resists while attacking is also noticeable when fighting immune lightning enchanted bosses that you want/need to kill, e.g. seal bosses in CS. There's the little technique of teleporting on top of them, let merc hack away, take a step behind the merc and start attacking (or let merc take care of it on its own). That way the merc takes the damage from those bolts. Also, being poisoned can be a little annoying when there are no monsters around to kill to keep up your life. However, usually Enigma's LPK takes care of that just fine with the next big group, and worst case scenario is to drink one rejuvenation potion.

Damage output of several gear setups
I wanted some numbers to compare different setups, so I've calculated a bit. Of course I just did it for myself to see what I should go with, but I thought why not share it here as well for those interested. Since this is with gear available to me, the numbers below are not necessarily accurate for better (or worse) rolls on individual items. Also this is for max LF dmg, since LF is used almost all of the time. In respect of ChS dmg, the attack speed difference between 35% and 55% is 12.67% instead of 10.13% for LF. As a rule for all setups, I required them to have at least 10 MPK and/or ML (at least 6%) in any combination.

Assumptions & gear used for all setups:
Griffon -16/+13 (15% IAS/3 MPK)
3/20 gloves

Inventory: 7 Skillers
Infinity Merc

Netting -101 ELR with Infinity. Damage output in following setups is in each case on mobs with 50% LR.

4x RBF shield setups (-18/+18 tested):
55% IAS:
  • Max LF dmg (LCS): 3366
  • HLW
  • ML ring (or 1.07 MPK)
  • Vs. 50% LR mob: 3366*1.69 [-69 LR mobs]*1.1 [55% vs 35% differnce] = 6274
35% IAS:
  • Max LF dmg (LCS): 3596
  • +2/8ML amulet
  • SoJ
  • Vs. 50% LR mob: 3596*1.69 = 6077 (3.24% lower)
Alternative 35% IAS with RBF Griffon's (-23/+18), ML ring and HLW
  • Max LF dmg (LCS): 3494
  • HLW
  • Rare ring
  • Vs. 50% LR mob: 3494*1.76 = 6149 (2.03% lower than 55% IAS, 1.18% higher than 35% IAS setup above
Side note (also regarding setups below): Personally I'm using the first 35% IAS variant and based the following comparisons on this for 35%. This is mainly due to the fact that I've got another Griffon's socketed with a 15ias/3mpk jewel, and a nice +2/8%ML/13@/+41 Mana craft amulet. This in combination with Silkweaves takes care of mana problems for good, and together with SoJ provides a much bigger mana pool. The extra ~1.2% damage provided by 35% IAS setup with HLW is not worth dropping that in my case.

3x RBF shield setups (ideally Head Hunter's Glory) (-15/+14 tested - don't have such a shield but simulated with other gear):
55% IAS:
  • Max LF dmg (LCS): 3263
  • HLW
  • ML ring
  • Vs. 50% LR mob: 3263*1.66*1.1 = 5974
35% IAS:
  • Max LF dmg (LCS): 3486
  • SoJ, +2/ML amulet
  • Vs. 50% LR mob: 3486*1.66 = 5786
In comparison to 4x RBF setups, running 3x RBF is a ~5% decrease in damage in my case. Note that this comparison is favourable towards 3x RBF because I simulated this as if the shield had -15/+14, whereas my 4x RBF shield is -18/+18 and thus further off perfect rolls.

2x RBF shield setups (ideally upped Moser's) (-10/+9 tested)
55% IAS:
  • Max LF dmg (LCS): 3134
  • HLW, ML ring
  • Vs. 50% LR mob: 3134*1.61*1.1 = 5565
35% IAS:
  • Max LF dmg (LCS): 3349
  • +2/ML amulet, SoJ
  • Vs. 50% LR mob: 3349*1.61 = 5391

Lidless Wall + RBF setups (-5/+5)
55% IAS:
  • Max LF dmg (LCS): 3239
  • HLW, SoJ
  • Vs. 50% LR mob: 3239*1.56*1.1 = 5573
35% IAS:
  • Max LF dmg (LCS): 3377
  • HLW, SoJ, -23/+18 faceted other Griffon's - this is because mana is taken care of by Lidless so ML on amulet is not required.
  • Vs. 50% LR mob: 3377*1.61 = 5436 (due to ability to use SoJ and RBF Griffon's, this is only 2.52% lower damage than 55%)

Spirit setups
55% IAS:
  • Max LF dmg (LCS): 3102
  • HLW, ML ring
  • Vs. 50% LR mob: 3102*1.51*1.1 = 5152
35% IAS:
  • Max LF dmg (LCS): 3301
  • +2/ML amulet, SoJ
  • Vs. 50% LR mob: 3301*1.51 = 4985

For comparison: Defensive/no RBF shield setup (e.g. SS, Moser's or HHG with Ums or PDs etc.)
55% IAS:
  • Max LF dmg (LCS): 2903
  • HLW, ML ring
  • Vs. 50% LR mob: 2093*1.51*1.1 = 4821
35% IAS:
  • Max LF dmg (LCS): 3102
  • +2/ML amulet, SoJ
  • Vs. 50% LR mob: 3102*1.51 = 4684

Conclusions: The difference in DPS output between 0x RBF shield and 4x RBF shield is just over 30% of total damage (for both 35% IAS and 55% IAS setups). The difference between 55% IAS and 35% IAS setups is around 3.25%, pretty much regardless of shield used, in case of setups tested. In other words, as long as 35% offers anything else (e.g. in my case ML on +2 amulet, ability to use SoJ...), it's probably better to stick with that. However for other setups the 55% DPS increase might be worth it. Ultimately it's a matter of taste, needs, and gear available.

Areas to run
Obviously, the total killer area for her is cow level, where she can fully realize her potential. In general, she can't really load up on MF because she's pretty gear-dependent. However, I've been running around 180% MF with Enigma, Gheed's and some SCs, and she can run any lvl85 area on /p7 extremely quickly, providing good drop chances for runes and the possibility for all S&Us... and at least for me offering a welcome change to the usual MF/rune hunting runs. ;)

Boy, this escalated quickly...
Initially I just wanted to share some stats and videos, but this got way out of hand in terms of length haha.. So I'm going to leave it here. I hope there's some information that's helpful to others, and of course any suggestions on improvement are welcome. :)

EDIT: Typos, included Spirit setups in DPS comparison.
Good job explaining everything about her. It was a good read.

I'd like to hear more about your merc's survivability in Pit. From my experience, it is not javazon who is in any danger, but rather merc is fragile AF. Why I didn't ever try CtA-less javazon is because I have no idea how to prevent merc from dying every 10 runs or so from an ugly archer pack, especially double archer packs with Fanaticism around. How do you manage that? I use 14 BO CtA AND Valk exactly to prevent merc from instadeath in such situations because I have no idea how to make him be close to those packs at all. Even with BO and Valk I tend to use slow missiles to not lose merc, they can instakill even that greatly buffed merc.

So, this is the point I'm interested to hear more about since problem of keeping merc alive without reseting run every 10 or so runs is an enigma to me (without CtA, with my style he is tankish and could last for at least 50 runs which is fine with me). Best answer I could get from other non-CtA pitamazons is to simply fury bosspacks and kill them fast. Well, tried it and it doesn't work like that, they kill merc much faster than I kill them if I don't watch out. Valk is amazing at soaking damage since it moves faster than merc and goes in front line first. For amazon herself that is not particularly problematic to do. It is true I never tried defiance merc and I will try it to see is this aura the reason that makes him sturdy enough to not need BO or Valk to keep him alive. So, do you have any special techniques about how do you manage to keep your merc alive in ugly archer packs situation, what do you do?


I might try something else about her, also testing Defiance merc to see how it works.
DON'T use Delirium on the merc...it will make herding nigh impossible...

MPK rings are not the only things that 1.07 could offer this gal... crafted Blood gloves (from a very high level character) can get 3/20 AND other tasty mods too...plus the automod CB + LL (potentially VERY high), while Greater Caster crafted gloves (from any character) can get 3/20 +other mods and will give automod MPK. Crafted Blood shields ALWAYS get 3sox from Larzuk, but do have insane STR reqs (Aegis base), but they can spawn -reqs mods...

@Gripphon Glad to hear you might play around with her and test some stuff, hope you can find some improvements!

Merc is super sturdy, really. I actually can't remember rebuying him apart from the two times I died since end of RFL. In my case Defiance and Chilling Armor (from Forti) brings his defense up to 22,759 with ebug Forti (3213 GH).

As far as Valk goes, to me she's like an ex girlfriend: Haven't seen her in a long time, don't see why I should, don't miss her, and I'm still disappointed with her. :)

@WoRG Delirium I had in mind for areas where you don't herd that much anyway, like Pits. Of course you're right it's obviously counter-productive in herding, so big no-no for cows for example. Even in Pits I haven't tested it yet because so far I'm perfectly happy with Andy's Visage, I just assumed it's more comfortable, but maybe it's time to do that and see how it compares in practice.

Also thanks for pointing out the other 1.07 stuff! I don't time-travel, but for those who do, some of those items are totally imba of course. What a surprise. :)
I've decided to give a shot to javazon which breaks every rule in the book:
- no 55 IAS
- no 100% pierce
- no faceted shield
- no javazon skills gloves

Does she work? Check video how she works. She has these stats:
- 35% IAS
- 77% pierce
- 280% MF
- 99 fcr

Griffon's (ias jewel)
2/9fcr/54life/80mana amulet
eth upped Titan's
Spirit Shield
Arachnid Mesh
LoH gloves
fcr/mpk ring (1.07)
War Travelers

CtA + Lidless

Merc gear:
Infinity, Andy, eth CoH

I think I learned map reasonably well, though there are still spots I have to investigate more about bosspack spawning (are they worth checking or skipping). Map is excellent for RFL running due to bosspacks spawning across 2 nests, but I don't think I can provide better performance on this map than on map I had before. Still, it is a good map, I might even keep it for now. Map shows 8.75 bosspack average in 64 seconds average time which is pretty solid, though I still have to improve on that. I don't believe this is best method of running after all.

I'll test more things. Last time I used 68 fcr and faceted shield, so I must check how that works on this map along with 100% pierce. I can immediately say there is big difference between 68 and 99 fcr. 99fcr allows me to teleport out of bosspacks almost freely while with 68 fcr that is nearly impossible to do. Even on video I teleported in some creeps few times and got out without asking for permission to let me go. I could also try CtA-less approach, though I still don't see how will I keep merc alive...
I have tested 3 different setups. Each setup has its positive sides and negative sides and all 3 setups ended to be identical in efficiency. At least on short term running.

Setup 1 (from video):
- 99 fcr
- 79% pierce
- 35 ias
- no switch needed for teleport!

Good and safe setup, though with this setup you pretty much don't wanna kill archers beside boss and you probably wanna skip archer champions if there are only 2 of those. They take too much time and are worthless targets in general. Also, lack of 100% pierce could be annoying from time to time, but at least you don't have to switch to teleport and you start attack immediately upon landing.

Setup 2:
- 68 fcr
- 100% pierce
- 55 ias
- HHN 3x facet shield

Griffon + ias jewel
eth Titan's
HHG shield + 3 facets
LoH gloves
Raven Frost
fcr/mpk ring 1.07
War Travelers

Excellent and very safe setup which kills even archers, but downside is it has 68 fcr which means zon has harder time teleporting out. To me this turned out to be in general best setup (use 4x facet shield if you have one).

Setup 3:
- wizzy setup as proposed
- 99 fcr
- 55 ias
- 100% pierce
- HHC 3x facet
- no BO

In theory, this setup should be the fastest. However, it didn't turn out to be for me because without BO you lose A LOT of safety factor both your merc and yourself. Everything is super easy until you face ugly combination of monsters and auras, then you pray they don't get you or your merc. All the advantage this setup offered me was lost due to less aggressive playstyle just to stay alive when ugly packs arrive, or so it was in my case. Also, my merc died once in ~10 runs as I expected and that puts game over to that run. In the end this setup did not make me fastest even when everything went smooth. But, if player has no issues with survivability whatsoever, this setup is the way to go since it offers everything you need to be as fast as possible. Like I said, in theory this setup is the fastest if it covers up your survivability issues. Just because I failed to prove it to be the fastest doesn't mean it isn't.

I'd say setup 2 is the best setup to go for when you are after leveling and MF. Setup 1 is equally strong and even easier to use, but you do level up slightly slower because you leave archers alive and not having 100% pierce could be slightly annoying, though it is not really needed for pit running. Fom MF aspect, setup 1 and setup 2 are similar. Setup 3 I would not recommend if you want to level up, you will face nasty safety issues and if you need to stay alive for hunderds of runs, I'd use CtA. Setup 3 is good if you can manage to keep merc alive 50 runs in a row easy because setups 1 and 2 can do that, and if you can't, then merc deaths will ruin any sort of advantage you might build by using this setup. Also, setup 3 is solid option if you don't care if you die here and there, aka you don't care about leveling OR you can manage staying alive with like 1500 life without death.

In the end, every player must test setups for themselves to find out the best. This is one of those characters you really have to shape according to your own needs and style. I would recommend to try all 3 setups or even something else as well.

Also one final note on equipment. I use LoH because it speeds up killing lightning immune bosses while 2 or 3 extra javelin skills are pretty irrelevant in my opinion. I think LoH also makes your javelin stronger than extra skills do since eth Titan's does really solid physical damage, bumping it up by another 350% makes it really good against immune archers and stalker bosses.

My 2€ on Pit javazon. I'm not gonna test her more since on this map there is no way to outperform my previous map. Here I'm ~10-15% less efficient due to less favorable map layout while using literally same setup as before. Rolling a new map is not an option because this one seems like a killer for RFL, so I'll just wait for new runefarming action for 1 year. Agreed.
@Gripphon Nice vid, as you can see, with 99% FCR you find nice runes! :) Going for 99% FCR on javelin switch while dropping 100% pierce is something that didn't cross my mind I have to admit. Interesting setup. But you lose out on sooo much damage from pierce and shield. Same for dropping 3/20 gloves, even on /p7 LIs drop fast. The best defense is killing monsters quickly after all, for both the javazon and the merc, so I take every bit of damage to kill them faster, hence why I bought into the all-offensive javelin switch.

In general I agree that setup 2 you described is very safe and solid standard variant. Question is really personally, do you want to have the 99% FCR, and do you want to do the stuff which is necessary to make it safe enough. It's not like switching CtA for Wizzy and you're good to go. Some things have to be adjusted like Defiance + ebug Forti merc, as well as play style.

Regarding merc safety, I don't think the CoH setup you mentioned in your first post does the trick. Forti is superior. Don't know how much defense CoH (I assume it's ebug) gets your merc, I imagine it's somewhere between 3500-4000? So assuming a player considers 4000 defense on non-Defiance merc + CtA to be sufficiently safe, then so is my setup. This is what you want to look at (for lvl 94 merc):
  • Chance for non-Defiance merc with 4k defense to be hit by archer in Pits: 100*4998[=Dark Archer AR]/(4998+4000[Defense assumed]=8998)*2***/(**+94[attacker & defender levels]) = ~54%
  • Chance for Defiance merc with ebug Forti (in my case 3213 GH) and triggered Chilling Armor to be hit by the same monster: 100*4998/(4998+22,759[!]=27757)*2***/(**+94) = ~17%
  • That is a decrease of ~69% (!) of hits taken by merc
  • 6 BO CtA gets my merc from 2119 to 3591 life, which is an increase of ~69%.
  • Yes, that is some irony. :)
So basically merc in my setup is pretty much exactly as safe as non-Defiance merc with CtA and (ebugged) CoH. With non-ebug armor, it's in fact safer than non-Defiance/non-ebug Forti merc + CtA. CoH also has PDR and leech, but the leech is irrelevant for this kind of gameplay, especially in those dangerous situations (ranged mobs) we're talking about here. So only advantage of CoH + CtA is that additional 8% PDR, which I don't think can be considered decisive. Also I should note: In case of Fanaticism pack (lvl 11 aura adding 85% AR) the CtA variant comes out on top: Here, Defiance reduces chance to hit by 59% in comparison to non-Defiance/non-ebug Forti setup. Also for those using beta CtA that is more beneficial of course. But in general it's very comparable merc safety. Of course if you want a super safe merc and don't care about 99% FCR, you can go 68% FCR with Defiance, ebug Forti and CtA. But in my opinion that's just unnecessary and I'd rather have the 99% FCR.

As for keeping the Amazon safe without CtA, it's just a combination of some practice and implementing those tweaks I mentioned. But more safety can always be added by using Slow Missiles vs especially ugly ranged packs or Delirium (Pits only), which should take care such issues for good.

All of that being said, I don't want to say my setup is objectively clearly the best way to play her or something. I just wanted to present it as perfectly playable, super fun and efficient. But I agree completely that javazon is a character where very many things come down to personal taste like personal notion of what is sufficient safety, attack speed considerations, utility of resists etc... so indeed everyone has his own personal ideal setup in the end. :)

EDIT: That stupid ** filter is annoying. Those are two 8s of course.

I'm using eth bugged CoH because I don't have eth bugged Fortitude, though I think it is time to make one now. I did find Lo as well yesterday, it suggests it is finally time to make this eth bug step. I hope I have armor somewhere.

ffs said:
  • Chance for non-Defiance merc with 4k defense to be hit by archer in Pits: 100*4998[=Dark Archer AR]/(4998+4000[Defense assumed]=8998)*2***/(**+94[attacker & defender levels]) = ~54%
  • Chance for Defiance merc with ebug Forti (in my case 3213 GH) and triggered Chilling Armor to be hit by the same monster: 100*4998/(4998+22,759[!]=27757)*2***/(**+94) = ~17%

That's plenty of stars in calculations hahahaha. Gotta love that 88 censure. Oh wait...

There is one more setup I didn't give a try. For start I think I should really go make that 4 faceted shield. Secondly, trying the typical 99 fcr with CtA, but using CtA switch to teleport and have 4x faceted shield on kill switch. It will lack 100% pierce though.
For calculations, use this page because it contains precise data on everything.

Dark Archer minion and bosses have 5283 AR on hell, and that is on p1. With each extra player setting monsters get 1/16 extra damage and AR, so on p7 bonus is 6/16 aka 0.375 or 37.5%. This means their AR on p7 is 7264.

This means before they had ~70% cth against my merc, and that is if they weren't Fanaticism or Blessed Aim. OP. With ebugged Fortitude chance to hit drops to 56%, or with Chilling armor on it drops to ~39%. If merc has Defiance (+ say 2 extra skills), then it adds extra 260% defense, or merc would have ~19370 defense without Chilling (archers have ~27% to hit merc), or with Chilling activated as well, merc would have ~25560 defense (archers have ~22% chance to hit him). In comparison to other merc type (non-defiance), Defiance offers extra ~107% effective life if Chilling is not activated or more realistically ~77% when it is activated. This assumes nearly perfect Archon plate Fortitude.

Pretty solid survivability bonus to merc. If I don't use Defiance and use ebug Fortitude, my merc would be ~79% sturdier. OP. This is pretty much what I like a lot. Defiance provides easier teleporting out of creeps because they have harder time hitting you which is a solid bonus as well. Might, on the other hand, is also harder to trigger since merc is not hitting much, sometimes it does not trigger till middle of the run already. Similar problem is with pitzerker, I have witnessed runs when Might didn't trigger at all.
Oh yeah I forgot about the player settings, well... But it's weird that the AR stated on arreat summit is different as well. Anyway, right now I'm too lazy to edit calculations to that effect. ;) For sake of completeness though, adjusting calculations above accordingly, Defiance/ebug Forti setup is 63% (instead of 69%) fewer hits taken in comparison to non-Defiance/non-ebug Forti setup, in my case (still assuming non-Forti setup as 4k).
Interesting calcs on the Defiance merc and survivability... always nice to see something different

Pretty sure that the Defiance guy would be less useful to a pitZerker, tho...Shout is even better than Defiance at boosting def, and a Zerker will max it for the synergy to Zerk

Upon testing last setup: 99fcr CtA switch + 4x facet shield kill switch with Arachnid (no 100% pierce), it turned out equalishly strong as all other setups. All setups are quite similar for me in the end.

I also payed attention to when does Defiance actually activate. It is pretty much same issue as with might, I managed to do full run without activated Defiance and it takes like half a run for it to activate. I believe it is like that with all auras that need some hits to be triggered.

Defense is nearly useless to zerker and his merc because
1. when you zerk, you have 0 defense
2. howl prevents monsters from attacking, so defense is pretty irrelevant for merc as well

Shout is not really important for similar reasons, though it is definitely fine for HC, but is mostly used only to prevent stucking among monsters, but it should not happen at all once player learns his map.

Might aura, on the other hand, could mean a difference between 1 shooting devilkin and chasing him around because he survived a hit. It also makes merc more likely to kill bosses and champions faster, sometimes he kills them before you pull your sword out.
Been drooling at this tread as todays few runs gave me a juicy Lo rune. However my setup is faar from complete so heres what left.

I HAVE NO SILKWEAVE! :( and curently no mf char as my sorc is in Baal mode setup and i got no tele barb or w.e its called. I also dont do timetraveling. I'm not sure how hard it is in 1.07 to get a high mpk ring but i would like one....forgot to add here that I have no active trading profile either..

Regarding shield...so a simple 4 socket shield all with facets is superior to a Spirit?
I got 2 free socket slots in my Titans and the Griffons however just as i checked i only got 10 IAS :D and thats from a rare glove with 2 java skills and some mana...so that means ill be lookin at IAS jewels to fill those sockets.

The Defiance merc aura worked out great, thanks for the tip. Went with Holy Freeze up intill now and herding the cows is much faster now and they barely are able to touch me..so thanks again for that.

You can try to use Lidless Wall as main shield since it has up to 5 mana per kill (make sure to use one like that). You would probably want some mana leech amulet or ring too. Speaking of mana leech, it goes with physical damage, so for gloves I would recommend to use LoH since it will pump your physical damage nicely and make you leech mana nicely. In the end, I do use LoH over 3/20 gloves myself, though it is debatable which one are better. 3/20 make slightly faster killing of creeps while LoH makes faster killing of immune bosses, unless you tend to skip those.

LoH would also solve IAS issues since it offers 20, +15 Griffons and you are done. Titan's does not have socket however since it is javelin. Alternatively, you could use Highlord's for IAS and then with IAS Griffon's it is pretty irrelevant do you have or not IAS in gloves slot. But I'd still recommend LoH for mana leech until you farm Silkweaves, then you can experiment with gloves and use javelin skills ones if you prefer them.

Best solution is to develop "telebarb" for MF. There are 2 telebarbs: zerker and singer. One is bloodthirsty, other is Jessica Rabbit in disguise.
@Codoh Glad you give it a shot! Shield with 4 facets is ~18% higher damage output than Spirit. How the hell did you manage to have Griffon's but no Silkweaves hehe... Anyway, faceted Lidless Wall is an interesting MPK source with higher damage output than Spirit as well. Note that Titan's can't be socketed btw. :) 3/20 gloves can be bought from Anya with some (or a lot of) patience. :)

(EDIT: Damn, @Gripphon is just too fast for me.)

ffs said:
This is what you want to look at (for lvl 94 merc):
  • Chance for non-Defiance merc with 4k defense to be hit by archer in Pits: 100*4998[=Dark Archer AR]/(4998+4000[Defense assumed]=8998)*2***/(**+94[attacker & defender levels]) = ~54%
  • Chance for Defiance merc with ebug Forti (in my case 3213 GH) and triggered Chilling Armor to be hit by the same monster: 100*4998/(4998+22,759[!]=27757)*2***/(**+94) = ~17%
  • That is a decrease of ~69% (!) of hits taken by merc
  • 6 BO CtA gets my merc from 2119 to 3591 life, which is an increase of ~69%.
  • Yes, that is some irony.
~69% decrease in hits taken is much more significant than ~69% maximum life increase from Battle Orders: since reducing chance to hit from ~54% to ~17% effectively reduces average attack damage to the same amount, it more than trebles (54/17) effective life.

Let's say a character has 1000 life and each attack inflicts 100 damage, so 10 hits kill: with 50% chance to hit, on average it takes 20 (10/0.5) hits to kill; with 20% chance to hit, on average it takes 50 (10/0.2), or 2.5 (50/20) times as many (so reducing chance to hit from 50% to 20% can have the same effect as increasing 1000 life to 2500, or +150% maximum life).

Gripphon said:
For calculations, use this page because it contains precise data on everything.
The AR figures on that (and every other page) don't take into account a bug which doubles AR of monster missile attacks, so although Unique Dark Archers and their minions should have just 5283 base AR, they actually have 10566 (so 14528 at /players 7).

Replace 4998 with 14528 in ffs' original calculations and the difference is between ~76% and ~38% chance to hit, effectively doubling the number of hits to kill: ~100% increase in effective life is still better than ~69% more maximum life, but the difference isn't quite as significant.

ffs said:
Oh yeah I forgot about the player settings, well... But it's weird that the AR stated on arreat summit is different as well.
The Arreat Summit's figures are based on monster level in MonStats.txt for all difficulties, which is accurate for Diablo II and bosses but not for all other monsters in Lord of Destruction beyond Normal, since monster level in Nightmare and Hell is determined by area level (so normal Dark Archers in the Pit in Hell are level 85, not 83, whereas those in the Tower Cellar Levels 1-5 are levels 75-79).

onderduiker said:
The AR figures on that (and every other page) don't take into account a bug which doubles AR of monster missile attacks, so although Unique Dark Archers and their minions should have just 5283 base AR, they actually have 10566 (so 14528 at /players 7).
Those archers are quite OP it seems. Do all other missile monsters also have doubled attack rating? Like, Council members in Travincal have double attack rating than listed?

onderduiker said:
~69% decrease in hits taken is much more significant than ~69% maximum life increase from Battle Orders: since reducing chance to hit from ~54% to ~17% effectively reduces average attack damage to the same amount, it more than trebles (54/17) effective life.

Let's say a character has 1000 life and each attack inflicts 100 damage, so 10 hits kill: with 50% chance to hit, on average it takes 20 (10/0.5) hits to kill; with 20% chance to hit, on average it takes 50 (10/0.2), or 2.5 (50/20) times as many (so reducing chance to hit from 50% to 20% can have the same effect as increasing 1000 life to 2500, or +150% maximum life).
Wow, I can't believe I failed to realize that haha, thanks a lot for pointing this out! :) Guess that's really what explains the safety benefit of Defiance setup.

Also thanks for explaining the AR issue, good to know.
Gripphon said:
Those archers are quite OP it seems. Do all other missile monsters also have doubled attack rating? Like, Council members in Travincal have double attack rating than listed?
The bug applies to missiles not monsters, so monsters that have both melee and ranged attacks don't get their melee AR doubled as well... although I think in most, if not all, cases where this is true, the ranged attacks always hit (no hit check) anyway (Succubus and Stygian Fury bolts, Act V Claw Viper missiles, and so forth).
Estimated market value