Please upgrade your browser for the best possible experience.

Chrome Firefox Internet Explorer
×

Lethality vs. Engineering vs. MM for PvE Ops

STAR WARS: The Old Republic > English > Classes > Gunslinger / Sniper
Lethality vs. Engineering vs. MM for PvE Ops

Rischardo's Avatar


Rischardo
07.14.2012 , 10:55 PM | #11
Quote: Originally Posted by Hxxr View Post
Nice quadruple post there. ;>
Anyway as with other classes Snipers have gotten a great gift with the Field Respec. Discussions like this have to be done on a boss by boss basis and even then your role (and spec) in the fight can differ occasionally.
.
I refuse to buy Field Respec because I was so against it in the form that went live. I admit that combined with the reduced respec costs, it's not horrible, but I will hold out as long as possible; it's the principle of the matter!

That said, I don't think changing specs according to the boss fight is currently necessary, even for HM Denova. For instance, my guild DPSes Zorn and Toth down so fast that we have issues with the switch because the first tank still has fearful. Backing up a little early before the jump actually helps in some ways.

So why did I bother to include things like "burst DPS" and "AOE DPS" as categories? Well, in the future NM fights may not be so forgiving. For instance, maybe your particular Ops group is low on AOE DPS and suddenly you find yourself struggling with the Trandoshian adds phase during Kephess. Switching to Engineering for that one fight might make all the difference.

Poostabby's Avatar


Poostabby
07.14.2012 , 11:17 PM | #12
I have to agree with Rischardo on practically all points made. Nice write up!
Tibbel I certainly respect your hard work on all the simulationcraft for the sniper community, and greatly appreciate it, however I have yet to see ANY proof that MM ENG or any variation hybrid of the two can sustain lethality's numbers in an actual raid setting, only on paper. It may look good on paper but so many things are unaccounted for in regards to a sustained rotation in a live raid setting that make these claims somewhat silly. As far as precluding ourselves from proof, the parses from live raids speak for themselves. Also my brain hurts just thinking about calculating all the caveats for lethality dps such as energy refunds via dot crits, the increase damage to target sub 30%, etc...Way too many variables for me to even think of modeling. Currently no raid encounter pushes anything but lethality to the top of the food chain, or else I would be running it. All it would take is one parse to show it, but I have yet to see one, or produce one. And good god I have tried. I ran MM forever when I first starting playing. After testing and experimenting countless times over, Lethality is it. No contest. That being said thanks for the stat calc. You cant argue with a stat scaling model that is accurate.

Tibbel's Avatar


Tibbel
07.15.2012 , 01:30 AM | #13
Quote: Originally Posted by Rischardo View Post
Could you share your thoughts on how MM/Engineering hybrid stacks up against the 3 "pure" specs in the 5 categories?
Thoughts aren't really all that useful for the first 3 categories. (Who cares what spec some guy thinks is the highest DPS? We just wanna know which one is the highest. ) Just like everyone else, I don't have any empirical data that addresses inter-spec DPS balance. We can make some assumptions based on interpretations of mechanics and perform calculations based on models, but that's all I've seen so far. I've included the assumptions I have made in the spoiler tag below.

Spoiler


Quote: Originally Posted by Poostabby View Post
I have yet to see ANY proof that MM ENG or any variation hybrid of the two can sustain lethality's numbers in an actual raid setting, only on paper.
And I haven't seen any proof that Lethality is so hands-down ahead of every other spec, either. Fair enough?

Quote:
As far as precluding ourselves from proof, the parses from live raids speak for themselves.
My problem with what I've seen with raid combat log parses is that they have not been evaluated scientifically (compared apples-to-apples, so to speak). It's not convincing to me to say, "I always beat the other snipers in the raid who are X spec," or even, "I used to be X, but now I'm Y, and I do better now."

To satisfy that need... do you think that we, as a sniper community, could pull off a scientific study? I'd be happy to design and organize it, and even to compile the results, but we'd need probably at least a few dozen snipers to all agree to execute it and be diligent in recording results and conditions in a consistent way. Anyone down?

Quote:
Also my brain hurts just thinking about calculating all the caveats for lethality dps such as energy refunds via dot crits, the increase damage to target sub 30%, etc...Way too many variables for me to even think of modeling.
Yeah, that would be a nightmare to calculate in your head. Y'ought to try using the spreadsheet, which does all that for you.

marineopferman's Avatar


marineopferman
07.15.2012 , 10:25 AM | #14
Quote: Originally Posted by Rischardo View Post
You average 2100 DPS? Sorry, I have a tough time believing this. Most min-maxed Snipers claim/show parses of around 1800-1900 DPS on a single target for an Ops test dummy. Let's not even talk about how most boss fights have some sort of mechanic that lowers your DPS (movement, periods of immunity/down time with nothing to DPS etc.) I think the one time I had 2100+ DPS on a mostly single target fight was the last fight in HM LI, but that's because of the increased damage the boss takes in phase 3.

My DPS can spike past 2000 DPS in the opening moments of a fight with OS and adrenals, but I don't know anyone who can sustain that. I'd like to see a parse showing you actually do 2100 DPS over at least 3 minutes on a single target.
Guess I should have said the boss G4- B3 Heavy Fabricator the only one I have ever been able to do that on LOL kinda cheap if you think about the nonstop NO armor on him LOL but trust me man get group and just spam him with no armor if your getting 2k on other bosses you will beat me on him.

Poostabby's Avatar


Poostabby
07.15.2012 , 11:44 AM | #15
MM is the best sustained dps spec, here is a dummy parse and an ops parse reflecting that.

....oh wait.

Rischardo's Avatar


Rischardo
07.15.2012 , 04:29 PM | #16
Quote: Originally Posted by marineopferman View Post
Guess I should have said the boss G4- B3 Heavy Fabricator the only one I have ever been able to do that on LOL kinda cheap if you think about the nonstop NO armor on him LOL but trust me man get group and just spam him with no armor if your getting 2k on other bosses you will beat me on him.
hey man, I can tell you're sincere and I don't want you to take this the wrong way, but I've seen some of your other posts and I think some people may have misinformed you about some aspects of the game. I know you're trying to be helpful, but you're doing more harm than good if you don't have your facts straight. Heavy Fabricator is NOT a good indicator of your overall average DPS for the very reason you said, so I hope you don't go around claiming you do 2200 DPS. Now that I think about it, I honestly think I parsed once on Heavy Fabricator at 3300 DPS over the entire fight (although the people working the puzzle were very good and we used the "fire only the right terminal" strat).

Quote: Originally Posted by Tibbel View Post
MM/Eng Hybrid
Sustained DPS: Total energy regeneration is exactly the same as pure MM -- 20 energy from Imperial Methodology every 30 seconds is the same net value as the 30 energy from a well-executed Sniper Volley every 45 seconds, assuming neither allows energy to cap. MM/Eng does gain a high-DPE attack in IP, so it could balance out or even exceed pure MM's Imperial Assassin damage, but that could also depend on crit. I have not performed an analysis on this since patch 1.3, and as far as I have seen, MM/Eng's sustained DPS ranking is untested.

Burst DPS: Cluster Bombs adds residual damage to one of our primary burst abilities (EP) in addition to adding burst energy regen, which allows higher prioritization of high-DPCT abilities. It can't compete with a fully-loaded Cull, but it also does require less set-up time following a target switch. Which is preferable is probably subjective and situational.

AoE: Due to its option of including Engineer's Tool Belt and Experimental Explosives, MM/Eng has better AoE damage potential than MM, both in burst and sustain. No Plasma Probe means it can't compete with pure Eng. How it compares with pure Leth is untested, but my prediction is that MM/Eng is higher.

Mobility: MM/Eng has IP as an ability that is instant and does not require cover, which pure MM does not have access to. It also has a longer cooldown on SoS and lacks SV, so its required channeling time is lower than pure MM, and probably very comparable to pure Eng. However it still relies on Sniper's Nest for added energy regen, so it does lose some damage when movement is needed. I suppose I would rank its performance under heavy movement above that of pure MM, but lower than pure Eng or pure Leth. However, if there's literally no time to stop and channel SoS or Cull, then MM/Eng is probably the best of all specs due to its low dependence on channeled abilities.

Survivability/Raid Utility: MM/Eng does not have the option to spec into Siege Bunker at the top of the MM tree, nor can it spec into Augmented Shields or Deployed Shields at the top of the Eng tree. It does, however have access to both Ballistic Dampers in MM and either Vital Regulators or Vitality Serum in the 2nd tier of Eng. I haven't seen any mathematical analysis of sniper defensive options, so it's hard to say how it ranks in survivability compared to other specs.

Raid Utility is another subjective and situational topic. I can say that if you need to reduce an add's movement speed and accuracy, then there's no better spec for the job.
Thanks for this Tibbel, would you mind if I updated my original post with this (I'll give credit where credit is due )

As far as raid utility is concerned, I understand it is subjective and situational. The way I looked at it, though, was that Snipers really only have defensive raid utility. We don't have blood thirst or pulls that easily reposition mobs. We can only prevent damage, either to ourselves or the raid (hence why I lumped them together).

Quote: Originally Posted by Tibbel View Post
It's not convincing to me to say, "I always beat the other snipers in the raid who are X spec," or even, "I used to be X, but now I'm Y, and I do better now." To satisfy that need... do you think that we, as a sniper community, could pull off a scientific study?
I actually somewhat agree with this. There was a guy another thread saying "Our MM snipers switched to lethality, but did worse so they switched back" and claimed that as proof that MM was better. It doesn't prove that one spec is better than another; it could also prove that his snipers are taking the full advantage of MM, but not lethality.

That said, we have a ton of circumstantial evidence that suggests that lethality is our highest ST DPS. Many experienced Snipers have run countless tests between MM and Lethality and most, if not all, came to the conclusion that Lethality is superior. Coupled with the fact that virtually all of a MM's sniper damage is mitigated in some way, all of this circumstantial evidence points in favor of Lethality.

Before you knock on circumstantial evidence, keep in mind that it is sufficient to convict criminals, even if the potential sentence is death. And you use circumstantial evidence more in your every day life than you realize; for instance, every time you drive through an intersection, you just used circumstantial evidence. You saw that your light was green and "knew" that it was safe to go through. But did you actually know that it was safe? You can't actually see if the light is red for the cross traffic...it is entirely possible that the traffic light is malfunctioning, so you don't actually know that it's working as intended. Yet despite this possibility, I'm sure you drive through your green light without giving it a second thought.

The situation is analogous to the comparison between Lethality and MM. You're absolutely correct in saying that we don't actually know if Lethality is the best. But the experience for the vast majority of Snipers has shown that Lethality beats out MM. In the traffic light example, you could stop your car, hop out, check if the light was red and only then proceed through the intersection. In this case, we could run a big scientific study to compare the damage potential of the 3 specs and make sure the combat logs are working correctly. However, in the end, you'll end up finding out what you already knew from the circumstantial evidence: the light is green...just like the sludge from Lethality's corrosive grenade.

Now, I'm not saying that this means the community should just stop and accept that Lethality is the best. Given the fact that I see many MM Snipers "complaining" that they have energy to spare makes me wonder if a MM stacking alacrity could do more DPS by spending the excess energy faster. Tests like these should still be done. However, I feel that a test to ensure that the combat log is working correctly is excessive and unnecessary.

NotRonin's Avatar


NotRonin
07.15.2012 , 06:12 PM | #17
The 'perfect' rotation for MM and Engineering can only be done by a computer. Lethality is more lenient : anyone with the gear can do good dps on the dummy.

Since it's a game, just play the way you enjoy. Lethality is definitely the FoTM right now, that does not mean it'll stay that way forever. As it stands, most players already have enough DPS to kill every boss in the game, it's STAYING ALIVE which is the issue. With 1.4, we may see everyone stacking on Endurance rather than cunning!!

marineopferman's Avatar


marineopferman
07.15.2012 , 06:30 PM | #18
Rischardo not to be insulting in the least bit but you just asked where i did it and thats where I dont mean to be misleading but if their is any other post that I have put in please send me a mssg so i can improve myself or atleast put it in your sentence so i can learn instead of leaving me out to dry :P

Tibbel's Avatar


Tibbel
07.15.2012 , 09:28 PM | #19
Quote: Originally Posted by Rischardo View Post
...all of this circumstantial evidence points in favor of Lethality.
I'm fully in agreement that circumstantial evidence is used all the time and generally to good effect. My primary issue with it is when we call that "proof". Often when there is evidence supporting a certain conclusion but that evidence is not scientifically conclusive, then we say it "suggests" a certain conclusion. (See "Boson, Higgs" for an example of a conclusion that recently crossed that threshold. )

Basically what I'm saying is that, in my opinion, the best wording is something like:

Observations made by snipers of various specs in operations settings suggest that Lethality consistently produces the highest sustained DPS against single targets. However, conclusive inter-spec DPS data does has not yet been generated.

It's not my thread, though, so that call is up to you.

Somokon's Avatar


Somokon
07.15.2012 , 11:16 PM | #20
Tibbel is absolutely correct. You cannot take a bunch of anecdotes from forum posts and claim to prove anything about dps potentials of different specs.

Fortunately, we have access to a very good combat simulator that will (barring any programming bugs in the simulator) tell us the true dps potential of any spec. I spent a bit of time experimenting with the action lists for the 3 pure specs, trying to maximize dps. Here is what I came up with:

Marksman
Code:
actions=stim,type=exotech_skill
actions+=/coordination
actions+=/snapshot_stats
actions+=/take_cover,if=buff.take_cover.down
actions+=/shatter_shot,if=buff.shatter_shot.remains<1.5
actions+=/use_relics
actions+=/power_potion
actions+=/adrenaline_probe,if=energy<=76
actions+=/target_acquired,if=energy<=100
actions+=/laze_target
actions+=/sniper_volley,if=cooldown.series_of_shots.remains>5
actions+=/followthrough,if=buff.followthrough.up&energy>=72
actions+=/ambush,if=buff.reactive_shot.up&energy>=81
actions+=/series_of_shots,if=energy>86
actions+=/orbital_strike,if=energy>=96
actions+=/takedown,if=energy>=81
actions+=/snipe,if=energy>=86&cooldown.followthrough.remains<=1.5
actions+=/explosive_probe,if=energy>=86
actions+=/corrosive_dart,if=energy>=86&!dot.corrosive_dart.ticking
actions+=/snipe,if=energy>=86
actions+=/rifle_shot
Engineering
Code:
actions=stim,type=rakata_skill
actions+=/coordination
actions+=/snapshot_stats
actions+=/take_cover,if=buff.take_cover.down
actions+=/shatter_shot,if=buff.shatter_shot.remains<1.5
actions+=/use_relics
actions+=/power_potion
actions+=/adrenaline_probe,if=energy<=40
actions+=/target_acquired,if=energy<=100
actions+=/emp_discharge,if=dot.interrogation_probe.remains
actions+=/plasma_probe,if=energy>=82|buff.energy_overrides.up|cooldown.adrenaline_probe.remains<1.5
actions+=/explosive_probe,,if=energy>=82|buff.energy_overrides.up|cooldown.adrenaline_probe.remains<1.5
actions+=/orbital_strike,if=energy>=96|cooldown.adrenaline_probe.remains<1.5
actions+=/series_of_shots,if=energy>86|cooldown.adrenaline_probe.remains<1.5
actions+=/interrogation_probe,if=energy>=82|cooldown.adrenaline_probe.remains<1.5
actions+=/takedown,if=energy>=81|cooldown.adrenaline_probe.remains<1.5
actions+=/ambush,if=energy>=81|cooldown.adrenaline_probe.remains<1.5
actions+=/laze_target
actions+=/snipe,if=energy>=86|cooldown.adrenaline_probe.remains<1.5
actions+=/rifle_shot
Lethality
Code:
actions=stim,type=exotech_skill
actions+=/coordination
actions+=/snapshot_stats
actions+=/take_cover,if=buff.take_cover.down
actions+=/shatter_shot,if=buff.shatter_shot.remains<1.5
actions+=/use_relics
actions+=/power_potion
actions+=/adrenaline_probe,if=energy<=76
actions+=/target_acquired,if=energy<=100
actions+=/orbital_strike,if=energy>=96
actions+=/corrosive_grenade,if=!ticking&energy>=86
actions+=/corrosive_dart,if=!ticking&energy>=86
actions+=/weakening_blast
actions+=/cull,if=energy>=91&(dot.corrosive_dart.ticking|dot.corrosive_dart_weak.ticking)&(dot.corrosive_grenade.ticking|dot.corrosive_grenade_weak.ticking)
actions+=/series_of_shots,if=energy>86
actions+=/takedown,if=energy>=81
actions+=/explosive_probe,if=energy>=86
actions+=/ambush,if=energy>=81
actions+=/rifle_shot

The setting I used:
Fight length - 300s (with up to 20% variation on each trial)
Iterations - 10000
Fight style - Patchwerk (no movement)

The dps numbers (which are dependent on imported gear)
Marksman - 1827
Engineering - 1782
Lethality - 1655

The more people that start experimenting with the simulator the better. Try to come up with higher dps action lists, try different fight styles, player skill settings, etc.