Here's what I got, it kind of is an obvious result but it may ease those who are curious like me:

Haste is indeed the strongest stats for Paladins, including the situations of my testing. Even though we don't have waiting times when we put HA on, haste still gives us a solid chance to fit one more attack (even in short durations).

Mastery>haste isn't totally bad. It basically can achiev a similar damage output in the Tendon situation (would need more extensive testing to get real numbers). But haste will, most of the times, still be a better choice. Mastery>haste only seems to be fairly equal in the situations where you have enough haste to fit one more attack with the aforementioned build. Barely losing an extra attack is a real hit to DPS, though. As latency is a huge factor, and having high haste with a mastery-build is tough, it's advisable to have as much haste as possible (currently) to high up your chances of having an extra swing.

Things to consider:

-> Delaying an Exo for CS is NOT a good idea, even with mastery stacking. That is, at least on our current tier. As Exo benefits from haste (and gets nothing from mastery), the highest your haste is the more Exos you should be hammering. More Exos>Stronger CS (on current tier).

-> Haste is even better in a situation without HA. As the improved AW is being the popular pick, I consider Haste would be even stronger on that.



There, it may not contain real numbers, but I consider the info gathered to be solid. Both builds have had a similar damage output in all of my testings, but with all things considered, having a bigger chance to strike an extra attack (and the fact that Exo>CS - even with mastery stacking -, and Exo does not benefits from Mastery) is a big reason to stick with Haste.

If you consider this to be garbage theorycrafting, feel free to hate it. But I think this kind of observations are the first step on determining old ideas and breeding new ones.

TL;DR: Both builds have had close enough overall damage on tendons, but Haste seems to be a better pick.




I'll try and run a few tests on SimC later to get real numbers for curiosity's sake.

EDIT:
Here are the numbers:

Haste build
Mastery build

Sorry, but it IS garbage theorycrafting, even if it's correct. And based on your actions, I don't think you understand what theorycrafting actually is. If you aren't backing up your anecdotal evidence with statistically significant numbers, you aren't theorycrafting at all; you're just beating up a dummy and saying "Yes, it blends!"

Here are the numbers:
Haste build
Mastery build

Messed up the trinket up-time, here's the fixed version:
Haste build
Mastery build

Basically what I said before.

Mastery builds are viable in these situations, but yeah, not really worth the effort of changing all the reforges.

I used all the Heroic tier pieces (including 410 shoulders). Trinket choice was Valor+Spine trinkets. The equipments are not from my armory, I made a full BiS set and reforged properly.

I can provide the full detailed reports in html if anyone desires to take a look at them.

EDIT: This was tested with Holy Avenger.

I don't want to be offensive, but this reads like a bad science pamphlet made by Big Tobacco or a Pharmaceutical Manufacturer.

1) Where is your data? You post pictures of SimCraft, but nothing about the supposed tests you performed. At least give us your logs.
2) What was your methodology to "separate the noise from the data"?
3) How did you determine what was noise and what was data, in order to perform said separation?

2 and 3 really smack of biased science. They can be restated very pejoratively as 'I threw out any data I didn't like or disagreed with my desired answer'. Again, that's an ugly way of stating it and not necessarily what you did. What you did doesn't even qualify as the lousy napkin math I occasionally post - that at least has hard numbers.

It's like saying "I surveyed the American public and 100% want to vote for: <politician>"... because I interviewed 2 whole people. Sample size matters, no matter what you do to 'clean up' data.

The observations remain useless. The useful thing you contributed to theorycraft was the original question: how does Haste vs, Mastery compare in some crunch situations. It's a good, solid, valid question. The type of question theorycrafters love to see (until someone asks repeat questions without searching threads). Your tests did nothing to prove, nor disprove, either option.

Asking questions is good!
Working properly to test questions is good!
Working to generate insufficient data that has no impact on the question and pretending it answers a question is bad!

So I was in LFR Terrace in beta with my guild and someone gave me symbiosis (Wrath for Rets) as a joke. I kept it and decided to try to be a ranged dps for fun. The amount of damage I could put it out at range was surprising, and wrath ended up being a large part of my damage. I then started experimenting with trying to replace other abilities with wrath and then finally trying to using wrath at melee range.

Logs containing these attempts are here: World of Logs - Real Time Raid Analysis

I realise its most likely not going to be optimal (and why would it be?) but I was wondering if anyone had even bothered to calculate how much dps a Ret can achieve at range or even at melee range using wrath. Could be useful in future if there are any fights involving flying bosses (similar to Atramedes).

The only problem with this is trying to convince a druid to give symbiosis to a ret!

stRafaello - Yes, please, share the details.

Hinkyz - that's an interesting question. We'd assumed Wrath would be useless (barring, as you said, some ranged fight where it actually let you hit a target). We could try some napkin math if we figure out the average damage of Wrath as well as coefficients (I'm not going to assume it's identical to Druid, but it could be) .

Spurs me to a related question - does Wrath apply/refresh Censure?

Also, you didn't notice anything wildly unusual like still being able to melee during casts*, right?

* Cast time normally castrates melee DPS since a surprisingly large portion of total damage is due to autoattacks and casts reset the swing timer.

I have a few questions I have been unable to find answers to either in this thread or through exhaustive searches on other forums so I'm left with trying to find the answers here.

What is the optimal opening sequence? Obviously pre-pot > But from there do you: 3hp>Inq to get it going then use cds OR would you build up 5hp then Inq at the beginning or 2hp Inq?

Does the priorty system of 5hp TV's change DURING cds or does that still remain in effect and why or why not? I have been throwing out TVs at only 3hp during CDs then building back up to 5hp TVs after the duration of SW or HA/AW is over. Is that optimal? Would you gain more TVs during CDs at only 3hp rather than building up to 5hp then TV?

When taking SW I have noticed that due to the length of the Inq time and SW time that I am am left with the choice to refresh Inq with my last global during SW or use TV. Which is optimal? When taking HA this seems to not be an issue as the buff durations of HA and AW are completed before the refresh of Inq is required.

I apologize if this has been answered before and I was just unable to find the information. Thank you guys so much for all your hard work.

With HA, I would think that the TV5 rule can be ignored since we would be wasting HoPo to max - i.e., start HA with 0HoPo, CS for 3, Jud for only 2 since cap is 5. With SW, though, I'd think that the TV5 rule should be maintained, however, since you have greater access to HoW even with fewer TVs. I could be wrong about SW, but I'm fairly confident about HA simply because of the wasted HoPo.

Grognard - for Wrath to be viable DPS on a non-gimmick mostly-ranged fight, it would have to be higher than many/most/all of our attacks. Using Judgement to stack/refresh Censure would run counter to this.

I find it really hard to believe Wrath could be so powerful - it wouldn't be affected by Inq, shouldn't be affected by SoL, shouldn't (though could) stack Censure, should reset swingtimer, and doesn't generate Holy Power. That doesn't mean it's not remotely possible.

Cartavian
Opening sequence - 3 HP, Inq, then CDs (at 0 HP if using HA, 3+ if otherwise to try to maximize TV usage). Going to 5 before Inq just increases the no-Inq window by a few GCD. I don't see a gain from that.

During CD - SimC by default doesn't change priority during CD. You can custom-set that with if statements around the CD being active. My spreadsheet is not capable of such on-the-fly logic. That said, just thinking logically, if you're using HA then every second attack should be TV, to avoid lost HP by hitting 6+. Other talents it's probably still advantageous to remain with the standard priority - 3HP if everything is on CD (rarely under SW) and possibly just at the end of the window to get just one more in under bonus (TV +20% better than something else +20%). Skipping 3HP TV just to build up to 5 shouldn't gain any additional attacks, and could lose a few as you leave an empty GCD (could be filled by 3-4HP), get to 5, then use TV when something else is off CD.

Inq/TV at end is debatable - 30% on Holy damage and 10% Crit for a few seconds/attacks vs. 20% bonus on a single TV. Gut would incline me towards Inq being superior, but it's possible the TV could win.

In general it makes sense that you could just follow the same priority during CD as outside of it. SimC could prove otherwise, but I suspect the DPS difference would be quite narrow.

3HP Inq not 5. I can't see any major practical reason to go to 5 HP before using Inq.

The only CD where the Priority system changes will be HA. You'll want to do HP-Gen -> TV -> HP-Gen -> TV -> HP-Gen -> TV. Don't do double HP generators under HA, you'll lose out on the gain of HP because you can't receive it, and as a general rule of thumb you don't want to use something that will generate HP when you have no room for it.

For SW you'll want to do the rotation normally otherwise though. Keep your CD's rolling, use TV at 5 HP unless you have a free GCD above 3HP!

SimC has that priority change for HA enabled.

Rafaello:

First, if you are trying to simulate a spine type fight, why are your sims only 10 seconds long? Shouldn't those be 20 second intervals?

Why no Guardian? Why no potions? Your reforging is not very precise either, you're a fair bit over hit and exp caps. You've got a str/mastery gem in the haste set as well.