Let's STOP the spread of misinformation about operative DPS (high OR low!)

[Ahebish]

Here's an example of an easy test to test class balance.

 

Nude characters, nothing but their weapons of same quality.

 

For example Marauder dual wield Rakata sabers vs a concealment Rakata blaster rifle and vibro.

 

Completely naked no other gear.

 

Now do a DPS parse on the target dummies with no additional buffs other than what your class is given. (Kinda weird with 1.2 legacy but you can remove non class buffs if necessary)

 

This truely shows the damage numbers difference between the 2 classes. Without factoring in to much gear or any gear differences.

 

As someone once told me... KISS

 

Keep it simple stupid.

[Ahebish]

All I'm saying is stop comparing max level gear for these parses...

 

Parse the base class first, before you start parsing gear.

Edited April 30, 2012 by Ahebish

[KyoMamoru]

I am practical about it. Your example is flawed, since it is based on someone looking at their DPS through the eyes of 100% perfect buffs, and 100% perfect debuffs. I'm looking at the game through only what the class can provide, without reliance on spike additives such as Adrenals/click relics. If I was to press my power Adrenal/relic at the exact same time, and have a lucky string of crits, I could dramatically skew my numbers. Throw in a blood thirst during that time, and the difference in my dps would be dramatic.

 

Yes, there are positional requirements in PVE encounters, but I do not want to look into those at the moment. I care about the 100% up time results currently. Once I've nailed that out completely, I can then distill various boss encounters based on mechanics that force Melee to stop attacking. During those phases, I can still dps for ~900 with my ranged attacks, while their damage is at 0.

 

You made yet another double [actually a triple now] post after I typed this, so I'll continue here. Your request is flawed. Classes scale with stats differently. Did you know Marauders don't have a stat-scaling talent like the Operative 9% extra cunning? What about classes with auto-crit abilities, but are then deprived completely of their surge [their most important stat]?

 

Technically, by the time the next tier of gear rolls around Marauders will be prenerfed due to their lack of stat-scaling talents.

Edited April 30, 2012 by KyoMamoru

[Ahebish]

I am practical about it. Your example is flawed, since it is based on someone looking at their DPS through the eyes of 100% perfect buffs, and 100% perfect debuffs. I'm looking at the game through only what the class can provide, without reliance on spike additives such as Adrenals/click relics. If I was to press my power Adrenal/relic at the exact same time, and have a lucky string of crits, I could dramatically skew my numbers. Throw in a blood thirst during that time, and the difference in my dps would be dramatic.

 

Yes, there are positional requirements in PVE encounters, but I do not want to look into those at the moment. I care about the 100% up time results currently. Once I've nailed that out completely, I can then distill various boss encounters based on mechanics that force Melee to stop attacking. During those phases, I can still dps for ~900 with my ranged attacks, while their damage is at 0.

 

You made yet another double [actually a triple now] post after I typed this, so I'll continue here. Your request is flawed. Classes scale with stats differently. Did you know Marauders don't have a stat-scaling talent like the Operative 9% extra cunning? What about classes with auto-crit abilities, but are then deprived completely of their surge [their most important stat]?

 

Technically, by the time the next tier of gear rolls around Marauders will be prenerfed due to their lack of stat-scaling talents.

 

Your comments are flawed as all of this can be tested under controlled circumstances

 

I'm not sure if you read my previous.

 

But keep it simple

 

Stop doing raid DPS parsing.

 

DPS parse the class.

 

We all know the DPS scales with stats.

 

So let's get a baseline without stats.

 

That is the only true way to keep gear differences out of the equation all together. The difference is just to vast.

 

First find out if the classes are on par with eachother without gear.

 

Everything after that is drivel because every person plays their toon differently, they have different builds, different specs, different mentalities for playing etc.

 

Eliminate all factors or as much as you can to keep it simple. Then base your DPS parses on that. So everyone has a starting point.

 

Are marauders truely out performing? You won't see that in a raid circumstance to a decided factor that can no longer be argued.

[brezman]

Parsing without gear completely ignores class scaling such as Kyo demonstrated above with the 9% cunning example. It's an even less valid way then what we are comparing now. Parsing PvP fights is no better as they skew burst situations and are highly susceptible to RNG effects. Averaging multiple 5 to 6 min parses is the best way currently to replicate boss dps.

 

This has been a really good thread thus far so we should keep this thread clean of whining (which is what these forums are known for) and continue to post parses with the gear level, buffs and rotation used, so that Bioware will recognize the disparity in our dps, if it happens to exist.

[KyoMamoru]

I'm actually about to change my set bonus to two pieces of Field Tech, to allow me to have an extra hit of OS for each cast. Assuming the exact same dps numbers as my latest parse, this would provide an additional 12,410 damage, or yield 1,459.92 dps. This isn't including the loss of dps from 3 hidden strikes + acid blades, which should yield 12000 extra damage [not counting the armor penetration they'd have], or 1501 dps. I'm missing ~80 cunning due to gear not being full Rakata [bM Implants, Bm Rifle, and my Chest isn't Rakata.] I'm going to work hard this weekend to improve my gear more.

 

I just wanted to follow up on this statement, since I haven't done any parsing this week. I have verified that yes, the Field Tech does produce one extra strike for the same damage as all the other strikes, despite not updating the tool tip to an increased amount of damage. I've yet to move the mods over in them, but I configured my gear to produce only a few points of damage difference in my test, and found that the individual strikes were producing the same damage. As mentioned before, this change is an increase of ~42 dps from simply using OS every cool down. This does not count the ease of popping OS + Adrenal/Relic + Blood Thirst and then doing your rotation which overtime would generate substantially more damage, or the fact that in actual combat, we're discussing the merits of an attack that has -no- cap for total targets with a very large radius.

 

Since we already know the Energy gain from the Enforcer set is crap, our best set is going to be two piece Back Stab critical increase, and two piece extra Orbital Strike. Don't forget, always make sure to have your target fully dotted [this includes Acid Blade's dot being fresh] before casting Orbital Strike.

 

In other news, Training Dummies have 35% DR against Kinetic and Energy. Their total armor is at 5814.

Operations training dummies have 5814 armor rating (not visible to you), which translates through our armor formula to about 35% DR.

 

Our armor penetration brings that down to 4,069.8 armor, or 27.36% DR.

 

The damage proc relics are:

184 Internal

246 Energy/Kinetic or 178.6944 after Acid Blade's Penetration.

 

With that said, during an Operation, the Energy/Kinetic will outpace the internal, as long as you're partied with a Juggernaut/Mercenary. Also, due to being a dot class, we are incredibly effective at using the proc relics so it is highly unlikely that we will find ourselves not having the effect applied on cool down.

Edited May 2, 2012 by KyoMamoru

[Ahebish]

Parsing without gear completely ignores class scaling such as Kyo demonstrated above with the 9% cunning example. It's an even less valid way then what we are comparing now. Parsing PvP fights is no better as they skew burst situations and are highly susceptible to RNG effects. Averaging multiple 5 to 6 min parses is the best way currently to replicate boss dps.

 

This has been a really good thread thus far so we should keep this thread clean of whining (which is what these forums are known for) and continue to post parses with the gear level, buffs and rotation used, so that Bioware will recognize the disparity in our dps, if it happens to exist.

 

Point taken.

 

But my point wasn't to help you PVE'ers understand that the DPS is lower. But to show it's even more considerable of a difference for PVP compared to all other classes. You won't get that from a PVE dummy.

 

Additionally... base stat comparisons ignores scaling as a factor... noone has the same gear setup each of have our own build ideas. Mine differ from yours. Rather PVE or PVP. Eliminating scaling as a factor allows people to get base numbers. With these base numbers we can theoretically apply scaling calculations. But to try to understand these calculations without a basis to start from is just a waste of time and will prove nothing to anyone as the arguement for situational issues is endless.

 

All I'm saying.

Edited May 3, 2012 by Ahebish

[Ryemfoh]

Additionally... base stat comparisons ignores scaling as a factor... noone has the same gear setup each of have our own build ideas. Mine differ from yours. Rather PVE or PVP. Eliminating scaling as a factor allows people to get base numbers. With these base numbers we can theoretically apply scaling calculations. But to try to understand these calculations without a basis to start from is just a waste of time and will prove nothing to anyone as the arguement for situational issues is endless.

 

What you're describing is only of value for theorycrafters trying to reverse-engineer the damage / scaling equations. The typical process would be: remove all stats > establish baseline > add (if possible) 1 point of stat x > establish new values > rince & repeat.

 

The purpose of dummy parses is to establish what it is possible to do with those abilities, not how much they hit for. If I wanted to know how much BS hit for, I'd stand and hit a dummy with just BS 1000 times. That type of analysis doesn't take energy, cooldowns, proc chances, temporary buffs, target debuffs, etc, etc into account which are required to determine your actual damage potential. They are there to tell us which abilities to prioritise, when to pop our trinkets for maximum effect, how long you can afford to delay an ability, whether it is worthwhile using something like Stim Boost (you are fond of claiming is worthless ... perhaps you can post your parses with and without stim to prove this point?), etc.

 

Simulators are even better at this because they allow you to run thousands of tests using varying rotations / priorities - and I believe the Operative Sim for ToR is nearly ready - but sim'd values still need to be verified via dummy tests otherwise they have no actual value.

[Bazzoong]

^^ The trouble with reverse engineering the abilities and stat effects and diminishing returns is that taking the RNG variation on the damage range into account while doing propers statistics with applying basic standard deviation 1000 hits would be nowhere near of a statistically valid sample, it would literally take about a day of spamming BS to get a sample of valid size, no sane person will do that, so the samples tend to be way to small which in turn heavily limits the validity of any sims created from that data.

 

So take sims a general guideline.

[Ryemfoh]

^^ The trouble with reverse engineering the abilities and stat effects and diminishing returns is that taking the RNG variation on the damage range into account while doing propers statistics with applying basic standard deviation 1000 hits would be nowhere near of a statistically valid sample, it would literally take about a day of spamming BS to get a sample of valid size, no sane person will do that, so the samples tend to be way to small which in turn heavily limits the validity of any sims created from that data.

 

So take sims a general guideline.

 

I suspect 1000 hits at each stat level would be more than sufficient to get statistically sound results. Regardless of whether or not that randomly selected number is valid, sims allow you to set the desired number of iterations (I believe the default in SimCraft is 10000)

[Jorojus]

Excuse me if I interrupt this nice discussion here but can anyone tell me how I can make these logs? Do I have to download something from somewhere, or have to activate them ingame ?

[Ryemfoh]

Excuse me if I interrupt this nice discussion here but can anyone tell me how I can make these logs? Do I have to download something from somewhere, or have to activate them ingame ?

 

There's an option to turn on combat logging to a file in the preferences. Log files are created in Documents>Star Wars - The Old Republic>CombatLogs

[Ryemfoh]

Bump because people keep asking questions that would be answered by this thread

[HollowVamp]

My mistake. However, I'd greatly appreciate constructive comments in this thread, rather than antagonistic ones. Passive agressive and derisive comments aren't helping anyone.

 

hes just calling out people who dont play maras, getting all this misguided false information

[Kaijan]

hes just calling out people who dont play maras, getting all this misguided false information

 

In a derisive manner. A statement such as "Actually, marauders do X not Y" would be far more beneficial than intentionally causing irritation.

[Hubhardt]

Hey Guys and Girls,

 

 

Here's a screenshot of my damage outpout

http://i45.tinypic.com/2ugjng3.jpg

 

I did this to the training dummy at my ship, which should be the same as the Ops-training-dummy.

Don't know if this is good or bad and I don't know what other classes are dealing :>

Just wanted to let you know, a 1400-1500 dps is possible! ^^

 

Notes concerning my toons gear and other circumstances:

- concealment built

- Full Rakata Gear

- Matrix Cube + Campaign Kinetic Tempest Relic

- changed all accuracy containing enhancements to Battle Enhancement and Adept Enhancement, some of them are T2 though

- I was fully buffed, thanks to my legacy

- used Rakata Skill Stim (I'm a Biochemist ó.ò)

- did not use Rakata Attack Adrenal

 

I spent some evenings reconfiguring my gear and I tested a lot, especially tested accuracy vs. surge/crit/power and somehow it turns out, that accuracy doesn't contribute that much to my dps than the other stats. atm I have 51 accuracy rating due to my offhand weapon.

I also checked my operation logs with accuracy at softcap vs. operation logs without accuracy at softcap and it turns out, I am better off ignoring accuracy and hoarding surge (cause I either had accuracy + crit/power or surge + crit/power enhancements).

Maybe in PvP you get other results, thinking of how useful extra accuracy can be against tanky enemies ô.ô but I'm more of an operations player.

So these are my findings! I would very much like to hear/read your opinions and get some advice to get better :>

Edited May 14, 2012 by Hubhardt
correcting fails

[Coplann]

1420ish DPS fully buffed, stimed and running optimized gear. Disapointing....

[Ich_Bin]

1420ish DPS fully buffed, stimed and running optimized gear. Disapointing....

 

I gotta agree. 1422 certainly is a very good result for an Operative... But other dps classes can do much more in that sort of gear + stim.

Edited May 14, 2012 by Ich_Bin

[Ryemfoh]

I gotta agree. 1422 certainly is a very good result for an Operative... But other dps classes can do much more in that sort of gear + stim.

 

We would need the ability and damage breakdown, and exact spec used to do any real analysis. KyoMamoru has already shown a few pages back what a difference these seemingly small changes can make (as you well know).

[Ich_Bin]

We would need the ability and damage breakdown, and exact spec used to do any real analysis. KyoMamoru has already shown a few pages back what a difference these seemingly small changes can make (as you well know).

 

That's certainly true, but even he didn't do nearly as much dps as most other dps classes with that amount of optimization can do.

 

Also we all know that Operative dps is probably the least forgiving class. A small error on our part or a knockback can screw up our rotation and dps real bad. So I think dps parses that have been achieved using a 100% optimal damage rotation are impressive, but they yield no results in comparing actual dps done during a raid.

Edited May 14, 2012 by Ich_Bin

[Hubhardt]

Here you go:

 

http://i50.tinypic.com/2zhjbzd.jpg

 

this is respective to my previously posted parse.

 

and the spec I used is this one:

http://db.darthhater.com/skill_calc/imperial_agent/operative/#::e6fefe2fe4fe3f19ef6ef8efef2

 

got the 9% healing received thingy to comfort my healers xD

Edited May 14, 2012 by Hubhardt
added spec

[Hubhardt]

Argh! ^^

 

Why do it in a simple way, if you can screenshot logs, cut out something, save this there, and put it up somewhere? ^^ it could have been sooooo easy:

 

http://www.torparse.com/a/35639

[Daiyukie]

Argh! ^^

 

Why do it in a simple way, if you can screenshot logs, cut out something, save this there, and put it up somewhere? ^^ it could have been sooooo easy:

 

http://www.torparse.com/a/35639

 

Maybe I'm missing something, but I don't understand what you're saying or even if you're complaining about the previous post or just trying to be helpful by reporting more of the info contained in the previous screenshot. I didn't see a problem with the screen personally, but couldn't you just edit the post? =p

Edited May 14, 2012 by Daiyukie

[Ryemfoh]

Argh! ^^

 

Why do it in a simple way, if you can screenshot logs, cut out something, save this there, and put it up somewhere? ^^ it could have been sooooo easy:

 

http://www.torparse.com/a/35639

 

Thanks - useful info. If I can make a few comments:

 

1) Try to parse at least 5 mins. This will ensure that any temporary buffs (e.g. Adrenals / Adrenaline Probe) don't skew the results (although you mentioned you didn't use any Adrenals), and is more representative of a real boss fight.

2) Check out my post on pg 4 (gives ability DPE breakdowns) and KyoMamoru posts a few pages back (starting around pg 11). There is some decent info in there which might help you on your rotation (e.g. you haven't used OrbS at all, and EP is quite far ahead of Overload Shot in damage and DPE)

3) In your opener you used Backstab without Acid Blade. This is a DPS loss over using BS to re-apply Acid Blade later in your opening rotation (from the logs it also looks like your TA was consumed by Stim Boost before BS actually hit which meant you lost 2% damage on BS and 2x AB ticks). There's some discussion on openers here, and again KyoMamoru's post on his rotation changes are really good at explaining his reasoning and methodology in raising his DPS.

4) You lost 2 TA's in the parse .. 1 to a superflous Stim Boost at the very end of the fight, and one somewhere else (this is assuming you maintain 1 TA at all times for the damage boost)

5) Adrenaline Probe was only used once (could have been used twice given the fight duration). That area of the logs (9:40:23 - 9:40:28) is actually a good indication of why you shouldn't refresh Stim Boost early.

6) Finally - it looks like there are about 5 seconds of inactivity at the end of the parse (this seems to be normal for the logs I've seen). This means your actual DPS was a bit higher while you were active (1479). Apparently if you activate Cloaking Screen right at the end of the fight you can force the logs to stop without this extra padding.

 

Thanks again for the extra info - I'll see whether I have time to do some more in-depth analysis after work.

[Hubhardt]

O.O

 

slowly it comes to mind, I never really thought about a good opener... and i never really thought about using EP when being in melee. ^^

seems like I have a lot to learn yet. and refreshing SB too early is a bad habit of mine I'm trying to get rid off. xD

 

Thanks for the advice, I'm looking forward to all further.

But for now, I gotta go praticing. xD