I've been playing rogue for quite awhile now, even to the point of maxing it out (100 power). So now I bring this question to the forums, what do you think is more beneficial to rogues, Crit or tempo? I personally think tempo with assassins third skill, makes assassin just as powerful as ninja. But I can't seem to find any overly great benefit when using Crit on either ninja or assassin. Any rogues out there, what do you think is more beneficial?
I've also managed to max my rogue, and I'm still on the fence about Crit and Tempo. To me, it seems, the deciding factor is how well you can dodge attacks (I'm assuming the role of Ninja; I have only a vague idea of how to play Assassin). If you manage to dodge everything perfectly and proc your passive almost every single time, I would say to build Tempo, as your passive will Crit for you. In a more realistic scenario, it will be more difficult to proc your passive and you will find that Crits happen less often. If you do the math, with 100% Crit rate and 200% Tempo (assuming 200% damage for every Crit), you'll do more damage than if you had 50% Crit and 300% Tempo. I currently run 103% Crit and ~200% Tempo on my Ninja.
^Basically that. Crit is broken when you can 1,3 combo kite some mobs and kill them nearly instantly.
I always kite and 3, 1. It's a lot easier and there are less chance to take damage, plus you end up where the loot is.
I've been testing this and in PvP tempo in my opinion is highly superior. As long as you have moderately good dodging skills Assassin + Tempo is crazy in PvP, just one stun and you win basically.
It depends on the natural speed of the weapon. As a fist Ninja I go moderate tempo and moderate crit since fists naturally have high attack speed. You basically don't do shit for damage with fists if you don't crit so you have to balance it out.
103% crit is waaay too much if you are relying on shuriken crits. Go 60% crit and the rest into tempo / stats.
Why is 103% "waaay" too much? It's a guaranteed critical and gives you significant more burst than 60% Crit and 80% more Tempo. Here is the math along with some statistics, since you didn't work it as I recommended you did in my previous post. If you compare item stats of legendary rarity (as far as I am aware), you can see that each crit point is worth two tempo points: http://imgur.com/USUBF09&hsMsg0I. A. My stats: 103% Crit, 200% Tempo B. Control stats: 80% Crit, 240% Tempo C. Scroats' recommended stats: 60% Crit, 280% Tempo Let's say I do "x" initial damage per attack. Assuming that Crits do 200% damage: A. (x)(200%)(100%)(200%) = 4x B. (x)(200%)(80%)(240%) = 3.84x C. (x)(200%)(60%)(280%) = 3.36x The first value is initial damage, followed by Crit multiplier, Crit chance, then Tempo. If you disagree with the math, please let me know so I can fix it in a way that is more accurate. Regardless, you can clearly see that Crit is invaluable and superior to Tempo. I rest my case.
The way I see it, get at least 100% in Crit, then you can go for Tempo. I did similar if not the same calculations as described in the above post, which led me to the same conclusion.
Your acting like every hit is going to happen 100% of the time. Those numbers may be the highest dps and in a perfect simulated world are great but that isn't realistic. I like balanced stats that not only allow for room for friendly fire but to take some hits. I'm also anti potion using so having some extra health / armor / res helps. Unholy cubes also give me life back pretty nicely. Built my warrior similarly in guild wars 2 which the meta game is very tank centric right now and requires you to have a balance instead of a glass cannon build.
1. If I miss 30% of my hits with a Crit build, I'm going to miss 30% of my hits on a Tempo build. Nothing changes. 2. How does building full Crit make you lose stats other than tempo? If you look in the screenshots, you'll see that the Crit weapons provided more health than the Tempo weapons. 3. Your third point is irrelevant and also not entirely true. Unholy spirits scale off of special "yellow" damage, meaning that Crit on shurikens will be more useful. 4. Building Crit does not make you a glass cannon; you get more health and more damage out of autoattacks than if you built tempo. Your logic fails you. If you really don't believe me, set up a time when we can both get on at the same time. Full health, only melee, no dodging/pots. Then we'll see.
DISCLAIMER: This technically isn't a double post sorta kinda, but I have good reason! It's just too long to fit with my previous post. Xera's math was stuck in my head for a few hours, and I finally realised what was wrong. I decided that I strongly disagree with the math used here, because it was just outright multiplication, i.e. not how you're supposed to do probability of a bonus: just for top lel: Proof of fail. (x)(200%)(20%)(360%) = 2.16x, when it clearly should be at least 3.6x the damage output. D. (x)(200%)(0%)(400%) = 4x (showing that optimal "damage" can only be gained at the extremes using your math) (well actually this would have turned out at 0x, which is CLEARLY not the case and makes your method of calculation completely void, but I figured that it's just 4x without any bonuses) So... I've done my own math. That'll show you. Ahem. Assume the following: base speed of 2 attacks per second base damage of 100 per attack a crit does 2x damage Example: "2 a/s" at "Tempo/100" speed, "crit chance"%, damage output is "2"*"Tempo/100"*"crit stuff" = # A: 2 attacks per second double speed, 100% crit, damage output is 2*2.0*(1+(100/100)) = 8, so 800 DPS B: 2 attacks per second at 2.4x speed, 80% crit, damage output is 2*2.4*(1+(80/100)) = 8.6, so 860 DPS C: 2 attacks per second at 2.8x speed, 60% crit, damage output is 2*2.8*(1+(60/100)) = 8.96, so 896 DPS Peak: 2 attacks per second at 3x speed, 50% crit, damage output is 2*3*(1+(50/100)) = 9, so 900 DPS D: 2 attacks per second at 4x speed, 0% crit, damage output is 2*4*(1+(0/100)) = 8, so 800 DPS As you can see, in terms of flat DPS and flat rates between Crit and Tempo, pouring into Tempo is more valuable than Crit 100%, up to a certain point. Overall damage increases as you balance more, decreases the more you bias towards either, as opposed to what your (incorrect) math showed. However, the overall increase in damage is trivial at 50% crit (1/8), AND the increase only counts for normal hits with weapons. And since in PvP the most used attack is a skill, it's much more beneficial to have that 100% chance of scoring a crit for it than 50%, for obvious reasons. The tradeoff of 100% crit to 1/8th extra damage is better in PvP situations, as fights don't last very long in terms of hits (what with your elitist no potion conventions) so there isn't much room to show off that extra 1/8 DPS, even if you do decide to go all out with weapon attacks (as you do, Scroats) Therefore, in the case of PvP, overall, having at least 100% Crit is more valuable than having as much Tempo as possible and not 100% crit. NOTE: this model does not account for damage fluctuation, hence there may be situations of chance where an optimum build does less damage than a non-optimum one. TL;DR I procrastinated from working on server stuff and proved you math-peasants wrong, also proving that the Buddhist way of balance is correct after all. P.S. "crit stuff" explained: "*(1+(%/100)" "*(1" The multiplier should never be below 1 as not having crit bonuses does not reduce your base damage. "+(%/100))" In an instance of 50%, 50/100, 5/10, if there were to be 10 attacks, then 5 would be doubled. The multiplier would look like this, "2, 2, 2, 2, 2, 1, 1, 1, 1, 1" which sums to 15 damage/10 attacks. The lowest is 10d/10a, done by "*(1", and any bonuses, i.e 50% or 5/10 are stacked on with "+(50/100))", making the sum also 15/10.
Yeah, something about the math was bothering me and I couldn't quite figure it out. With the 1:2 Crit to Tempo ratio, the end damage should have been different than what I found. I should have been thinking about what I was actually calculating. I am a dirty, filthy, scum-of-the-earth math peasant :c.
First of all, this is absolute and utter nonsense, as you can't do two attacks per second "but then double fast". Two attacks in a single second are two attacks in a single second and stay two attacks in a single second. You cannot just specify it's speed and then re-specify it's speed: this way you'd have to attack both 2 times per second and four times per second at once, which plainly is not possible. Let's just say the basic variant of a weapon does between d1 and d2 damage, with an damage of da raw damage per hit. And let's just say it does said amount of damage once every tb seconds. Now, without any chance of a critical strike whatsoever and a grand total of 100% tempo, which leaves the weapon's speed completely unmodified, you will be doing da / tb DPS on average. Nowever, this is still just a tad too generalized and all-in-all unspecific for our purposes. As you all know, DPS is determined by dividing the total amount of raw damage by the inverse of the frequency of the attack. A few basic variables you have to utilize in such a situation would be: the total average amount of raw damage dealt on each hit (in DMG), henceforth defined as dtotal the frequency of the attack (in Hz), henceforth defined as fatk the modifier of the speed of the attack (real number), henceforth defined as fmodif the probability of the attack to result in a so-called "critical strike" (real number), henceforth defined as dc First of all, we have to determine the total amount of damage dealt on average. While we already know that without any chance of a critical strike occurring our average damage output was da, this did not include the bonus damage given by critical strikes. On average, the amount of raw damage you will deal while including the possibility of a critical strike would be ((100 - dc) * da + dc * 2da) / 100 = dtotal, or (1 + dc / 100) * da for short. Now that we have the average damage of the attack, it is time to find out the speed at which said attack will take place, which is simply 1 / (fatk * (fmodif / 100)). All in all, that makes the amount of DPS you will averagely deal be equal to ((1 + dc / 100) * da) / (1 / (fatk * (fmodif / 100))). You can easily further simplify that, but right now it is 6 AM so I am just going to drop unconscious onto my bed now.
I said base speed of 2 attacks per second. And as far as I can remember, your total Tempo determines your final attack rate. So, if at 100% Tempo is your base rate, 200% Tempo would be double speed final. And anyway, by your standards, my calculation was (1 / fatk)*fmodif*(1 + dc / 100) *dtotal If that makes any more sense, you beautiful mathematician, you.
All I know is I dominate just about everyone with my set up. Math or not the leaderboards when they come will show.
and what I said it ultimately proven true by Ninths math in that balanced stats produce the highest dps at 900 dps compared to full one way or another.
That makes some sense. However the skills to use dodge/attacks properly out weighs the stats you have in my opinion. Practice makes perfect xD.
