Jump to content

The Best View in SWTOR contest has returned! ×

Powertech PvE DPS Build Comparison


MorningMusume

Recommended Posts

Should I fish for rail shot crits, or am I better off building flame thrower stacks?

 

That's a hard question to answer without a simulation.

 

From the spreadsheet I can tell you to use RP and RB on CD, so that's the most rail shot crits fishing you can do. So you'll probably don't build that much stacks of FT, but you'll do more sustained dps. Seems that increasing the number of FB and reducing the number of RP doesn't increase the dps. And RB of course if out of question, because is the hardest hitting skill per cast time.

Link to comment
Share on other sites

For the 21/2/18, did you factor in flame shield procs refreshing rocket punch? I know it's hard to find a value for when rocket punch will refresh with flame shield present on average, but the synergy between flame shield and particle accelerator is a major component of the build (refreshing RP which in turn refreshes rail shot).

 

Maybe it's worth running two simulations for this build - one where flame shield procs at or near its internal cooldown every time (such as in a full pve tanking scenario) and one where flame shield is ignored (such as in pvp when you're being ignored).

Link to comment
Share on other sites

Nope, because that comparison are with dps gear on, and not tanking.

 

You can use the spreadsheet to get the values when tanking if you want, and I modeled that interaction via a few hours simulator that for each % shield and boss attack delay returns the average Rocket Punch and Rail Shots per minute.

 

You will never go near that dps value when tanking tought, unless you stay with dps gear, and you'll probably get crushed. I don't have the values here, but a 21/2/18 tanking was around 890 dps, around 4-5% higher than the typical 31/8/2.

Link to comment
Share on other sites

Small question that I hope it fits here more than any where else. I am the main tank of a very small guild (5-7 members), I am currently level 26 so early still in my progress, we are all leveling together. I am currently using the Tank/Pyro build (http://www.torhead.com/skill-calc#301RMurd0roZ0MZfhbrzh.1) but my question is should I go all the way up in the tank tree before switching over to the Pyro? Or is there a time period that would help me.

 

I find that there are cases where the DPS (Operative and Assassin) over aggro me that it takes some getting my taunting back on me.

 

Is there a "level by level" guide as to when we should be allocating skill points? I was putting my points in to Shield until level 23 then started putting points into pyro.

 

Thanks in advance!

Link to comment
Share on other sites

Can't help you, as I leveled full Shield Spec, but maybe for lower level Flashpoints you don't need that much defensive talents, so maybe going 18 Pyro first for the heat control can help you greatly with threat/leveling. Then you can fill the Shield Spec, and leave the last 2 points on AP for the end.
Link to comment
Share on other sites

For the 21/2/18, did you factor in flame shield procs refreshing rocket punch? I know it's hard to find a value for when rocket punch will refresh with flame shield present on average, but the synergy between flame shield and particle accelerator is a major component of the build (refreshing RP which in turn refreshes rail shot).

 

Maybe it's worth running two simulations for this build - one where flame shield procs at or near its internal cooldown every time (such as in a full pve tanking scenario) and one where flame shield is ignored (such as in pvp when you're being ignored).

 

Assuming you have the base 5% shield rate, and assuming you were even being attacked to shield something, and assuming you have a shield generator instead of a damage generator, we are looking at a 45% chance of a 50% chance of a 5% chance on every shieldable attack of resetting the cooldown on RS.

 

You would likely be seeing a very small increase in the availability of Rail Shot at the cost of increased damage taken, and the loss of stats from using a shield generator - which I will hazard to say will be a net loss in damage done.

Link to comment
Share on other sites

Assuming you have the base 5% shield rate, and assuming you were even being attacked to shield something, and assuming you have a shield generator instead of a damage generator, we are looking at a 45% chance of a 50% chance of a 5% chance on every shieldable attack of resetting the cooldown on RS.

 

You would likely be seeing a very small increase in the availability of Rail Shot at the cost of increased damage taken, and the loss of stats from using a shield generator - which I will hazard to say will be a net loss in damage done.

 

Err... what? How can it be a net loss if it's just accounting for something that's already in the build?

 

Every shieldable hit gives you a shield%/2 chance to cut the cooldown time of rocket punch in half. Considering that most hits in pve at least are multi-hit, the odds of lowering the cooldown on RP are very high. This is fairly obvious to most shieldtech players - we do a lot of rocket punching. It's much more than a 5% dps increase even without particle accelerator.

 

@OP - thanks for the response. I'm not sure how I would modify the numbers to account for it - you'd have to make a lot of assumptions about how often you're being attacked and by what type of attack.

Edited by Valkenheineken
Link to comment
Share on other sites

Err... what? How can it be a net loss if it's just accounting for something that's already in the build?

 

Every shieldable hit gives you a shield%/2 chance to cut the cooldown time of rocket punch in half. Considering that most hits in pve at least are multi-hit, the odds of lowering the cooldown on RP are very high. This is fairly obvious to most shieldtech players - we do a lot of rocket punching. It's much more than a 5% dps increase even without particle accelerator.

 

@OP - thanks for the response. I'm not sure how I would modify the numbers to account for it - you'd have to make a lot of assumptions about how often you're being attacked and by what type of attack.

 

That's exactly the problem. We are discussing PvE dps, which implies that the player is not actively taking damage from attacks. In order for Flame Shield to even apply we have to make several assumptions about the setup of the player:

 

- They are being attacked, or at least taking shieldable damage.

- They have a shield generator instead of a power generator.

- They have CGC active, and are not benefiting from the additional shield chance of IGC.

 

Assuming everything above, and assuming the player is taking 1 shieldable attack every second - which I believe is a very generous estimate - you are looking at the following:

 

- A 5% chance of shielding the attack, which provides an average of 1 shielded attack every 20 seconds.

 

- Of those shielded attacks there is a 45% chance of resetting the cooldown on Rocket Punch, which averages out to an extra rocket punch approximately every 42 seconds.

 

- Of those Rocket Punches, there is a 50% chance of resetting the cooldown on Rail Shot, which averages to a free Rail Shot as a result of the prior Rocket Punch every 84 seconds.

 

It seems clear that the benefit of Flame Shield in this scenario is one Rocket Punch 40 seconds, or maybe 7 over the course of a 5 minute boss encounter, with maybe 3 extra Rail Shots as a result (please keep in mind this assumes a shieldable attack every second for the duration).

 

Since the cost of gaining these extra attacks is using a shield generator, every attack additionally suffers from the loss of offensive stats - which I will hazard to say makes Flame Shield, and the primary power of this spec, ineffectual.

 

I suppose that would at least be interesting to simulate out - at what rate of shieldable attacks does using a Shield Generator over using a Power Generator with this spec become feasible.

Edited by Lezale
Link to comment
Share on other sites

Anyway, as I said, this thread is only for DPS, you are not tanking here. I already did another thread about Tanking Builds, that is probably lost in the page count.

 

In no tanking situations, never ever having a shield generator will do better dps. The Tech Power lost is a great chunk of dps, for a minimal chance of an extra Rocket Punch. Not only even with 1 attack per second, you'll get like 1 RP per 40 seconds, that RP can be an almost wasted one, for example, with the normal RP cooldown going off in 1 second.

 

But you can already estimate that using the spreadsheet, just put 5% shield chance and a boss attack time. Shame is I only coded in 1 second and 1.5 seconds, and are probably not the best values for a dps situation.

 

Edit: also probably AOE damage will not be Weapon Damage damage, so you can't shield it.

 

Edit2: btw, for example, with 25% shielded attacks, and 1 second boss attack, my simulations are ~2 extra RP per minute, so with 5% shielded attacks you can be in the range of 0.4 extra RP per minute.

Edited by MorningMusume
Link to comment
Share on other sites

Can't help you, as I leveled full Shield Spec, but maybe for lower level Flashpoints you don't need that much defensive talents, so maybe going 18 Pyro first for the heat control can help you greatly with threat/leveling. Then you can fill the Shield Spec, and leave the last 2 points on AP for the end.

 

Thanks....maybe I will play around with the respec to get what I need to fit...I like having the base level of Shield, so maybe keeping that. What I did forget to mention is that we occasionally (well the healer (my wife) and I) do some PvP, so I need the DPS in addition to some survivability, and put the rest in Pyro.

 

Thanks again...if others do have suggestions, I would appreciate them.

Link to comment
Share on other sites

I understand what you're saying now. I don't think this spreadsheet can answer my question (which is, how much dps is added to a 21/2/18 tank rather than going 31/8/2). As an aside, I find that flame shield procs way more often, even in pvp, than 2 times per minute. I think that 1 attack/second assumption is off. Many bosses attack more rapidly, and certainly against multiple enemies you'll be taking more than 1 attack/second.

 

Oh well. It's really a discussion for another thread.

Link to comment
Share on other sites

I understand what you're saying now. I don't think this spreadsheet can answer my question (which is, how much dps is added to a 21/2/18 tank rather than going 31/8/2). As an aside, I find that flame shield procs way more often, even in pvp, than 2 times per minute. I think that 1 attack/second assumption is off. Many bosses attack more rapidly, and certainly against multiple enemies you'll be taking more than 1 attack/second.

 

Oh well. It's really a discussion for another thread.

 

Sorry if I'm not clear enough, English is not my main language.

 

First: my spreadsheet can help you with those kind of doubts, because is designed for that, with tanking in mind as a first priority. You can compare 21/2/18 to 31/8/2 on tanking scenarios with a -I hope- good approximation.

 

Second: extra Rocket Punch per minute doesn't mean Rocket Punch procs. Mean that if you usually do 60/9 = 6.67 RP per minute, you do 6.67 + X, where X is the extra Rocket Punch per minute.

 

And no, X is not the number of procs you get. Because when you get a proc, Rocket Punch reset his normal cooldown, so you won't do 6.67 "normal" RP per minute now, but less.

 

A simple example:

 

Let's say you have 25% shielded attacks, and the boss attacks every 1 second (not many raid situations have higher attacks than that). On average you'll get 12.5% procs (as the talent has a 50% chance on shield). That's 1 proc every 8 seconds on average.

 

So over 1 minute, you'll have 7.5 procs. But you can't have procs all the time, as the talent has 4.5 seconds of internal cooldown. So if you have 7.5 procs per minute that's 33.7 seconds you CANT have procs. Hmmm, but those 7.5 procs where calculated over 60 seconds, doesn't make sense. That's because you'll never have 7.5 procs, you'll have less, probably around 5 per minute (being generous).

 

But then, those 5 procs per minute can land on a good spot or not. A good spot helps greatly to the number of Rocket Punch / minute, a bad spot does not.

 

Good Spot:

 

0s -> Rocket Punch (normal)

1.5s -> Rocket Punch (proc)

10.5s -> Rocket Punch (normal)

 

With just 1 proc, you got 3 RP in 10.5 seconds.

 

Bad Spot:

 

0s -> Rocket Punch (normal)

7.5s -> Rocket Punch (extra)

16.5s -> Rocket Punch (normal)

 

With 1 proc, you got 3 RP in 16.5 seconds, just because you got the proc 7 seconds later.

 

Very Bad Spot:

 

0s -> Rocket Punch (normal)

8.5s -> Proc Rocket Punch

9s -> Rocket Punch (normal or extra, doesn't matter :p)

 

With 1 proc, you got 2 RP, exactly the same if you didn't get any proc.

 

That's why you need to simulate real rotations (as I did) and calculate averages. You can have 5 procs per minute and never do an extra Rocket Punch with bad luck. Or you can have 5 lucky procs and do 10.5 RP per minute, but you will never ever get 6.67 + 5 Rocket Punch from 5 procs.

 

The average value of Rocket Punch per minute with those values is 8.77, or 2 extra.

Edited by MorningMusume
Link to comment
Share on other sites

  • 3 weeks later...

*Bumps*

 

Hey MM, does your spreadsheet already use or are you planning on adding in the surge calculation implemented in 1.1.3?

Surge rating has been re-balanced. It now reaches diminishing returns the same way as other damage ratings, and its per point damage contribution has been reduced by approximately 10%.

Link to comment
Share on other sites

  • 3 weeks later...

Hi MorningMusume,

 

very useful info here - thanks a lot. I stumbled on this when I was looking to get more DPS as PT in a offtank role, and move away from my full tanking shield built, so I can be more useful in single boss ops fight for DPS.

 

I really like 21-2-18 1074.03 DPS, but I'm having problems figuring out the rotation. Could you help me out with that?

 

Regards,

 

Dr. Scarlett

Link to comment
Share on other sites

×
×
  • Create New...