Relic of Xuen

[Ghostmind]

Hi all,

I'm just wondering if the Relic of Xuen (agi version ofc.) is actually Bis for rogues atm? All specs that is.
Bottle of Infinite Stars perhabs better for muti?

Someone told me that Terror in the Mists (Terrence of endless spring, 3rd boss) is the only trinket that is better then Xuen for muti rogues.


What's your thought on that?

[adain878]

From what i understand while maybe not BiS its top 3 for pretty much every class that can use it.

[Sesshou]

Hi all,

I'm just wondering if the Relic of Xuen (agi version ofc.) is actually Bis for rogues atm? All specs that is.
Bottle of Infinite Stars perhabs better for muti?

Someone told me that Terror in the Mists (Terrence of endless spring, 3rd boss) is the only trinket that is better then Xuen for muti rogues.


What's your thought on that?

I think Terror in the mists is only better if its the heroic version (going from EJ). Honestly it makes sense that the relic is OP. Its Agi with an Agi proc. It may be lower ilvl, but agi is worth a ton more than secondary stats. Considering how much you would have down before you can get a heroic terror, you might as well think of it as bis for this tier.

[Zhinkis]

From what i can tell it's:
Assa: Bottle HC>Bottle>Terror HC>Relic
Combat: Bottle HC>Bottle>Terror HC>Relic
Sub: Bottle HC>Bottle>Relic>Terror

[Deleted]

From all the testing I've done albeit in sims I'm getting the following results for Assassination:
Bottle heroic>Bottle normal>Relic of Xuen>Terror heroic>Bottle raid finder.

I'm fully convinced Terror isn't a bad trinket but you might as well stick with the Relic.

[Mizzeeh]

The problem with the stat weights on ElitistJerks is that it assumes(overly generously, as actually stated by the guy that did the math) that the trinkets procc instantly when they come off CD(and you get full benefit from them, but that's an entirely different thing that I don't plan on going into).

This creates all sorts of inaccuracies, most of which result in fairly minor changes in value(think ~ -100 EP on most of them). But there is one noticeably large inaccuracy it creates and that is in the value of Relic of Xuen.

The reason for this is that unlike the other (good) trinkets, Relic of Xuen proccs on crit, which means that it takes >5x longer to procc for most of us than the other trinkets after the ICD comes up, as a result the actual uptime of the procc is quite a bit less than estimated on ElitistJerks and the trinket is actually worse than they state.

When I realized this I did my own math on the trinkets based on Assassination EP values and estimated attacks/second and crit chance in raids. I won't bore you with the details(because the values won't fit for everyone but they don't change dramatically depending on gear so it should be just as relevant as the list on EJ), but my findings were by and large:

#1 BoIS(HC)
#2 BoIS(N)
#3 TitM(HC)
#4 BoIS(LFR)
#5 TitM(N)
#6 RoX

That said, these are still BiS values, at current gear values Crit a fair bit(~10-15%) worse compared to Agility than it is in BiS gear, so right now RoX might beat out TitM(N), but as you close in on BiS gear TitM will beat RoX out for Mutilate, at least.

I did not do the math for Combat, but considering that Crit and Agility values are almost the same in Combat and Mutilate, these values are still almost definitely correct(if anything, TitM would be a little better, relatively).

[Stede]

The problem with the stat weights on ElitistJerks is that it assumes(overly generously, as actually stated by the guy that did the math) that the trinkets procc instantly when they come off CD(and you get full benefit from them, but that's an entirely different thing that I don't plan on going into).

This creates all sorts of inaccuracies, most of which result in fairly minor changes in value(think ~ -100 EP on most of them). But there is one noticeably large inaccuracy it creates and that is in the value of Relic of Xuen.

The reason for this is that unlike the other (good) trinkets, Relic of Xuen proccs on crit, which means that it takes >5x longer to procc for most of us than the other trinkets after the ICD comes up, as a result the actual uptime of the procc is quite a bit less than estimated on ElitistJerks and the trinket is actually worse than they state.

When I realized this I did my own math on the trinkets based on Assassination EP values and estimated attacks/second and crit chance in raids. I won't bore you with the details(because the values won't fit for everyone but they don't change dramatically depending on gear so it should be just as relevant as the list on EJ), but my findings were by and large:

#1 BoIS(HC)
#2 BoIS(N)
#3 TitM(HC)
#4 BoIS(LFR)
#5 TitM(N)
#6 RoX

That said, these are still BiS values, at current gear values Crit a fair bit(~10-15%) worse compared to Agility than it is in BiS gear, so right now RoX might beat out TitM(N), but as you close in on BiS gear TitM will beat RoX out for Mutilate, at least.

I did not do the math for Combat, but considering that Crit and Agility values are almost the same in Combat and Mutilate, these values are still almost definitely correct(if anything, TitM would be a little better, relatively).

Poison and bleed ticks don't have a chance to crit to proc?

[Sesshou]

Poison and bleed ticks don't have a chance to crit to proc?

I would think that they do. They are both on the melee table. Absolutely nothing to support that guess though as I won't have my dmc till the 4th.

[Squirl]

Poison and bleed ticks don't have a chance to crit to proc?

They do. Looking at some logs now to determine how close to ICD Relic is proccing for me...

[Deleted]

After some non-sim napkin math it seems like Terror in the Mists heroic's proc is still very weak, but the passive agility is making it edge out over Relic of Xuen. My initial thought was that Terror's proc is easier to waste with phase changes and such because the internal cooldown is way longer, but because the value of the proc is so weak it seems like heroic Terror is actually better, since you don't waste much if you can\t use the proc to it's fullest.

Even when considering full T14 BiS gear, heroic Terror really isn't far ahead of the Relic in terms of stat weightings. But regardless I'm starting to change my opinion and I may very well have been incorrect in my previous post.

[Verain]

I finally got this card last night, and I was shocked at the time it took to proc.

Since the proc chance is the same as the strength one (20%) but on crit instead of hit, this trinket seems a lot weaker than advertised. I can go quite awhile after ICD expiry before seeing the proc, and I'm pretty sure that this is about three to five times as long as a warrior, DK, or ret would have to wait, so I'm having a hard time thinking that our trinket is really up with theirs.

This is just based on attacking some mobs and a bit of time spent on a target dummy, but it's still odd enough. I'll have to figure out the average time I'm seeing between procs to really have much to offer.

---------- Post added 2012-11-04 at 03:38 PM ----------

They do. Looking at some logs now to determine how close to ICD Relic is proccing for me...

Do you have these numbers?

[Xanjori]

Just a quick look at my last Garajai, from 220seconds of combat my Relic had a 27% uptime (60 seconds, 4 procs). My LFR Bottle had 46% uptime (5 procs). What was interesting was that Dancing Steel (ive only got it on one weapon) had a 75% uptime (23 procs).

Most of the fights I did have the same kinda uptime, 25%~ for Relic, bottle 45%~. Normally its between 1-3 procs less depending on length of fight from Relic.

[Deleted]

I've bought this trinket today, shit better die in raids.

[Verain]

Just a quick look at my last Garajai, from 220seconds of combat my Relic had a 27% uptime (60 seconds, 4 procs).

Hrm. That's an odd amount of time to check (remember ICD 55). I mean, if I had a 55 second fight, I would definitely expect to see exactly one proc, even with the lower proc chance, and I couldn't see more than one. Still, that's within the realm of acceptable. I'd like to see it more on like a full five minutes. But thank you, that's reassuring!

My LFR Bottle had 46% uptime (5 procs).

With best possible performance, you were expecting another proc in 5 seconds here, so again, we're so close to the end of the final period that it's not as helpful at detecting an anomalous proc chance. However, the bottle definitely works as expected- when the ICD is off, I can open a regular mob and it almost always procs quickly. The relic will be lazy sometimes. In any event, the longer ICD on the relic mitigates the lower proc chance.

[Xanjori]

Hrm. That's an odd amount of time to check (remember ICD 55). I mean, if I had a 55 second fight, I would definitely expect to see exactly one proc, even with the lower proc chance, and I couldn't see more than one. Still, that's within the realm of acceptable. I'd like to see it more on like a full five minutes. But thank you, that's reassuring!

I took Gara because it was the only fight I sat on the boss full time, I figured WotE or Elegon would have lightly strange readings as there is time stood around, or time swapping targets.

On Vizier tonight, for the 350 seconds I was alive on our kill I had 6 procs of Relic vs 7 procs of Bottle. There is some downtime again on this guy and I really need a kill where I stay alive

Id say with a 55second CD thats a decently quick amount of procs.

[Jiigeri]

From all the testing I've done albeit in sims I'm getting the following results for Assassination:
Bottle heroic>Bottle normal>Relic of Xuen>Terror heroic>Bottle raid finder.

I'm fully convinced Terror isn't a bad trinket but you might as well stick with the Relic.

Also seen the same results over several +25k iteration sims. I'm propably sticking with the relic aswell and in 5.1 after upgrading my weapons I'll be saving the valor points for my trinkets, most likely (if it's usable on Relic of Xuen, which seems likely).