test content
What is the Arc Client?
Install Arc

Release Notes: 3/20/14

2»

Comments

  • aiqaaiqa Posts: 2,620 Arc User
    edited March 2014
    quasimojo1 wrote: »
    Just to add $.02 (w/ some data behind it), I have a Fire Snake/Ice Form/Ice Blast toon and a 2GM/KM toon and a couple of LArc toons and regularly "bench-test" my toons with my log parser and Regenerating Test Dummies. My LArc toons do about 10% more DPS on average than the 2GM and Ice Blast, which are themselves pretty comparable. Your mileage my vary..

    From my LA/BallLightning/ElectricForm/LightningSheet build I get about 4300dps, TGM/Quarry/LockNLoad/Imbue about 5100dps, and IceBlast/FireSnake/IceForm/IceSheet build (if memory serves, this has been a while since I tested) about 5000dps.

    Of course those number vary from having different builds.
    Here are some test I did a while ago, using 1 build (dex/int/ego) to test multiple power combination.

    http://co-forum.perfectworld.com/showpost.php?p=3871051&postcount=27
    http://co-forum.perfectworld.com/showpost.php?p=3871061&postcount=28
    http://co-forum.perfectworld.com/showpost.php?p=3871071&postcount=29
  • kaiserin#0958 kaiserin Posts: 3,182 Cryptic Developer
    edited March 2014
    selphea wrote: »
    Would probably be more credible if you post a parse of Ice Blast Fire Snake vs Lightning Arc Ions vs High Severity TK Assault Mental Storm vs 50ft 2GM and... idk, Hex?. From what I've seen they're close enough that it seems fair, given the relative advantages of each passive.

    Stormbringer is tanky in every Rampage, Shadow Form is stealthy and Ice Form is... squishy except against Frosty. I used to run Ice Form and the high parses were very nice, but even then I gave that up for the advantages of the other passives.

    Of course there's also Fire Snakey Heat Waved Unstably Accelerated Conflag but it has so many disadvantages and needs so much setup I'm happy to let it have the crown.

    I'm about to (possibly) go off for a week otherwise I'd do a thorough test myself.

    I'd be hesitant to add Mental Storm into testing protocols as anyone can slap to on their build and have it add to their damage output, even if it fits with the attack.

    Currently my hardest hitting build (by a landslide) is a Quarry/Ice Blast/Firesnake/Mental Storm/Rimefire build. Strafing Run could probably be shoved in there for even more laughs but I use a controller so none of that for me. Recent parse against a regen dummie, running 2 pieces Justice 1 piece Legion:

    obsidian_dps.jpg
    (I had a nightmare Generator equipped for a device)

    From my testings Ice Blast does trump the other overpowered blast abilities in raw output. Not by much, but then you have to consider how well Ice Blast synergizes with other damaging abilities.

    I still find Quarry to be the superior damage passive choice, even on non-physical builds. The Int/Ego boost is silly.

    I need to try Conflag again, but considering how much time you're wasting keeping up debuffs I don't think it would even make the list.
  • jonsillsjonsills Posts: 6,318 Arc User
    edited March 2014
    I never even figure that kind of thing. My main interest is - is using this power fun? Does it help me get through missions?

    If I wanted to work spreadsheets in my spare time, I'd play EvE, ya know? :smile:
    "Science teaches us to expect -- demand -- more than just eerie mysteries. What use is a puzzle that can't be solved? Patience is fine, but I'm not going to stop asking the universe to make sense!"

    - David Brin, "Those Eyes"
    Get the Forums Enhancement Extension!
  • quasimojo1quasimojo1 Posts: 642 Arc User
    edited March 2014
    aiqa wrote: »
    From my LA/BallLightning/ElectricForm/LightningSheet build I get about 4300dps, TGM/Quarry/LockNLoad/Imbue about 5100dps, and IceBlast/FireSnake/IceForm/IceSheet build (if memory serves, this has been a while since I tested) about 5000dps.

    Those are all higher than the #'s I tend to get, but seem reasonable.

    Just as a general comment on testing methodology, the chargy/bursty type attacks can tend to show very high numbers on short timeframes compared with maintain type attacks. To account for this when testing my own toons, I generally record for a 30-second period and take the overall DPS at the end of the 30 seconds. I may do a few runs this way and take the best. Sorry if this seems overly pedantic/technical--data analysis is kinda my thing.
    LTS since 2009. Author of ACT parser module for CO. Founder of Rampagers. Resident curmudgeon.

    "Without data, you're just another person with an opinion." -- W. Edwards Deming
  • kaiserin#0958 kaiserin Posts: 3,182 Cryptic Developer
    edited March 2014
    quasimojo1 wrote: »
    Those are all higher than the #'s I tend to get, but seem reasonable.

    Just as a general comment on testing methodology, the chargy/bursty type attacks can tend to show very high numbers on short timeframes compared with maintain type attacks. To account for this when testing my own toons, I generally record for a 30-second period and take the overall DPS at the end of the 30 seconds. I may do a few runs this way and take the best. Sorry if this seems overly pedantic/technical--data analysis is kinda my thing.

    30 seconds is very low to get accurate results, your crit chance won't average out.



    jonsills wrote: »
    I never even figure that kind of thing. My main interest is - is using this power fun? Does it help me get through missions?

    If I wanted to work spreadsheets in my spare time, I'd play EvE, ya know? :smile:

    But analyzing data is fun ;p
  • quasimojo1quasimojo1 Posts: 642 Arc User
    edited March 2014
    kaizerin wrote: »
    30 seconds is very low to get accurate results, your crit chance won't average out.

    Longer is better. 30 seconds is usually about the limits of my attention span for clicking. :-)
    The important point is that you have a consistent methodology and make sure it's long enough to average out the variations.
    kaizerin wrote: »
    But analyzing data is fun ;p

    Agree!
    LTS since 2009. Author of ACT parser module for CO. Founder of Rampagers. Resident curmudgeon.

    "Without data, you're just another person with an opinion." -- W. Edwards Deming
  • aiqaaiqa Posts: 2,620 Arc User
    edited March 2014
    quasimojo1 wrote: »
    Those are all higher than the #'s I tend to get, but seem reasonable.

    Just as a general comment on testing methodology, the chargy/bursty type attacks can tend to show very high numbers on short timeframes compared with maintain type attacks. To account for this when testing my own toons, I generally record for a 30-second period and take the overall DPS at the end of the 30 seconds. I may do a few runs this way and take the best. Sorry if this seems overly pedantic/technical--data analysis is kinda my thing.

    I generally test 100 seconds. In my experience dps is stable at that point.
  • selpheaselphea Posts: 1,229 Arc User
    edited March 2014
    kaizerin wrote: »
    I'd be hesitant to add Mental Storm into testing protocols as anyone can slap to on their build and have it add to their damage output, even if it fits with the attack.

    It's not so much about the theme but without Mental Storm's 24% debuff TK Assault won't be at full potential :x
    Currently my hardest hitting build (by a landslide) is a Quarry/Ice Blast/Firesnake/Mental Storm/Rimefire build. Strafing Run could probably be shoved in there for even more laughs but I use a controller so none of that for me. Recent parse against a regen dummie, running 2 pieces Justice 1 piece Legion:

    obsidian_dps.jpg
    (I had a nightmare Generator equipped for a device)

    I think Vixy runs something similar but with Throwing Blades in there somewhere. Seems very squishy though!
    From my testings Ice Blast does trump the other overpowered blast abilities in raw output. Not by much, but then you have to consider how well Ice Blast synergizes with other damaging abilities.

    I still find Quarry to be the superior damage passive choice, even on non-physical builds. The Int/Ego boost is silly.

    I need to try Conflag again, but considering how much time you're wasting keeping up debuffs I don't think it would even make the list.

    Well Quarry for damage yes, but 60-70% Crushing Resist vs Gravi and same amount of Cold Resist vs Frosty is nothing to sneeze at. Same with +120 Perception Stealth. It really depends on your definition of "damage passive" though - does it mean the passive with the highest DPS parses or a passive that can be used in a damage role, but also has a use other than damage? And how should that passive affect damage of powers that have synergy with it?
  • selpheaselphea Posts: 1,229 Arc User
    edited March 2014
    aiqa wrote: »
    From my LA/BallLightning/ElectricForm/LightningSheet build I get about 4300dps, TGM/Quarry/LockNLoad/Imbue about 5100dps, and IceBlast/FireSnake/IceForm/IceSheet build (if memory serves, this has been a while since I tested) about 5000dps.

    Of course those number vary from having different builds.
    Here are some test I did a while ago, using 1 build (dex/int/ego) to test multiple power combination.

    http://co-forum.perfectworld.com/showpost.php?p=3871051&postcount=27
    http://co-forum.perfectworld.com/showpost.php?p=3871061&postcount=28
    http://co-forum.perfectworld.com/showpost.php?p=3871071&postcount=29

    What happens if you standardize the AO to a singular Ego Surge + NM (or Imbue if low Con), or an dual AO Ego Surge-Imbue rotation? Seems like the AO may be skewing numbers a bit as Electric Sheath is the only one out of the three that doesn't buff crits, while 2GM benefits from 2 AOs (and no secondary attack).
  • kaiserin#0958 kaiserin Posts: 3,182 Cryptic Developer
    edited March 2014
    I get why you'd want to use Mental Storm in testing, I'm just saying it can be added to any build and increase performance, so if you only use it for ego builds it would skew results a tad.

    I consider Quarry superior as it provides the best damage output of passives that are oriented to increasing your damage, petspam AoED aside. While I can't exactly tank Frosticus with that character, he's far from squishy in a PvE setting thanks to the massive amount of cd reduction and being able to keep MD up most of the time.

    I like Stormbringer, but I don't see much purpose to it at the moment when I'm not having survivability problems.

    I haven't played much with Night Warrior, though. I can't get past it cluttering my power bar.

    I'd have to think about how I think damage passives should function.
  • selpheaselphea Posts: 1,229 Arc User
    edited March 2014
    kaizerin wrote: »
    I get why you'd want to use Mental Storm in testing, I'm just saying it can be added to any build and increase performance, so if you only use it for ego builds it would skew results a tad.

    I consider Quarry superior as it provides the best damage output of passives that are oriented to increasing your damage, petspam AoED aside. While I can't exactly tank Frosticus with that character, he's far from squishy in a PvE setting thanks to the massive amount of cd reduction and being able to keep MD up most of the time.

    I like Stormbringer, but I don't see much purpose to it at the moment when I'm not having survivability problems.

    I haven't played much with Night Warrior, though. I can't get past it cluttering my power bar.

    I'd have to think about how I think damage passives should function.

    Was actually referring to Shadow Form re: 120 Perception Stealth :p

    I suppose the idea is that I think the number of variables need to be minimized to find out which power is actually overperforming, e.g. are we using one AO or two AOs? Should AOs be standardized?

    How many damage powers should there be in the rotation - should Ice benefit from a 3-power Ice Blast - Fire Snake - Rimefire while Lightning is limited to two powers? How about a 4-power Ice Blast - Fire Snake - Rimefire - Gas Pellets on Mega-D? And so on.
  • aiqaaiqa Posts: 2,620 Arc User
    edited March 2014
    selphea wrote: »
    What happens if you standardize the AO to a singular Ego Surge + NM (or Imbue if low Con), or an dual AO Ego Surge-Imbue rotation? Seems like the AO may be skewing numbers a bit as Electric Sheath is the only one out of the three that doesn't buff crits, while 2GM benefits from 2 AOs (and no secondary attack).

    I am not home right now, so I can't test. But the numbers I get on my normal builds are also quite different due to the builds being different. Statswise the TGM build uses dex/int/ego, the LA build uses ego/rec/end, and the ice build uses dex/int/rec.
    That is why I added those other tests that all use one build, LA+BallLightning can get pretty close to TGM alone, but TGM still does significantly more damage when you use quarry, and similar damage with kinetic manipulation, adding AO (rotations) should not make a difference in that.
  • kaiserin#0958 kaiserin Posts: 3,182 Cryptic Developer
    edited March 2014
    I would say initial testing should be done with a generic +damage boost passive that provides no other damage perks during combat, something like Stormbringer, as well as a form. And testing should be done without debuffs.

    This would just be there to give an idea where the powers on their own fall, compare their output based on tier, range and energy cost.

    From their start taking into consideration what sort of debuffs they have access to and evaluating whether or not certain damage types need access to more debuffs to match other sets.
  • aiqaaiqa Posts: 2,620 Arc User
    edited March 2014
    kaizerin wrote: »
    This would just be there to give an idea where the powers on their own fall, compare their output based on tier, range and energy cost.

    *stamps feet* Tiers are almost meaningless! :tongue:
    Also I'd add mobility to that.
  • kaiserin#0958 kaiserin Posts: 3,182 Cryptic Developer
    edited March 2014
    Tiers exist in the game, unless they do away with them I'm going to take them into consideration! Being able to have your op power at level 6 makes a difference in build construction! :p

    And yah, forgot about mobility.
Sign In or Register to comment.