test content
What is the Arc Client?
Install Arc

DPS Differences PUG/high DPS groups

yreodredyreodred Member Posts: 3,527 Arc User
Hello, everyone!
I haven't been on this forum for quite some time, trying to avoid getting headaches lol.

When i was researching for new build inspirations for command BCs and Flagship Star Cruisers (haven't really cared about FOTM builds since Delta release), i stumbled over some extremely high DPS builds with values over 100k DPS (there are even some 150k+ ISA runs on YouTube). The funny thing is that some of my ship builds where already quite similar to those (especially some A2B or AHOD/Reciprocity builds). So i did a few ISA runs (PUGs) out of curiosity and analysed the logs afterwards (using Combat log Meter).
With 20-30k the results where more than disappointing and puzzling. Of course, i wasn't expecting values above 70k, but when i saw some rather mediocre (looking) builds that where used to archive 80k and more, i thought that archiving similar (or slightly smaller) values shouldn't be a big thing tbh.
I never really used any third party DPS meter in STO, i thought that @tooltip was good enough for me, as far as i can remember. (i made a long break playing STO and after that i didn't care much about that matter yet.)

Maybe i need to say that i'm aware of most Traits and most DOFF/BOFF power combinations that where introduced with and after the Delta Quadrant add-on, i'm wondering what causes such a huge difference in DPS numbers?
Can a jump from ultra Rare to Epic weapons/consoles be the cause of such a insane DPS increase?
Or can results vary THAT much depending if someone flies in random groups or high DPS groups?

Please don't get me wrong, i'm not asking for specific builds or "how to fly", i'm in general not that interested in min/maxing in STO or spamming FAW like crazy just to raise some numbers.
I'm much more interested in understanding how things work and how i can use that knowledge for my own type of builds, It just feels like i have missed something important which no one talks about (lol).
(But i would appreciate a good source for ship builds or explanations, of course. Finding or creating a decent ship build has become quite a bit confusing compared to pre DA, imo. :) )


I have also noticed that once in a while there are some players who seem to easily steam roll ISA in mere seconds (and humiliating everyone else BTW, ;) ). So my guess is, its all about equipment rather than how much DPS other people do, or is there more behind all that?

Thanks in advance!


@Mod:
I hope this is the right place for a topic like this, if not please move it to a more appropriate sub forum.

"...'With the first link, the chain is forged. The first speech censured...the first thought forbidden...the first freedom denied--chains us all irrevocably.' ... The first time any man's freedom is trodden on, we're all damaged. I fear that today--" - (TNG) Picard, quoting Judge Aaron Satie

A tale of two Picards
(also applies to Star Trek in general)

Comments

  • postinggumpostinggum Member Posts: 1,117 Arc User
    DPS groups stack debuffs waaaay more than pugs, including the insanely expensive kemo skills. Additionally time buffs for the team, so for example they can run with tac fleet up all the time, similar with the iconian 4 piece boost, which I think works out as 1.374*1.33 +83% for the entire run, similar can occur for the intel team penetration ability, what with everyone having AHOD.

    they practise exact positioning and timing of powers so some things like pets and fleet support aren't used on the first group but held back for the transformer and spheres. People will do 100+ runs just to get that down, if using the intel spec thay learn where the flanking facings are on ISA and attack those. Also they make sure to be shooting at close range often using the point blank(?) trait. They say piloting is the biggest issue.

    A dps run does not start with people just flying to the cube and shooting when they see fit, instead they pre-launch pets and recall them, position and then on a signal attack as a group; how many pugs do see someone zoom in ahead and get all the aggro? the parse does not record combat until combat starts, so the dps is helped.

    use epicced gear and consumeables that pug players probably won't bother with; drop armour for dps increasing consoles and threat reducing consoles and as a group kill stuff so quickly they generally survive.

    Spend for traits, doffs, boffs, gear anything that adds that tinniest boost. For example a lobi store console costs 300 lobi, 5.5 crystals per box 4.5m a box, if the stuff from the boxes is sold, a lobi console will cost about 100m or a couple of mil dil, just for the tachykinetic console that adds .76CH and 7.6CD

    Report the good runs. At the weekend someone was reporting repeated 200k+ runs
  • tobiashirttobiashirt Member Posts: 630 Arc User
    The above is essentially true...epic weapons, for example, have a combined [Acc/Dm] mod that turns out to be a 6% final-category modifier after all other buffs. The biggest part of the dps increase, though, is the coordination and the extremely short runs, plus the reporting only of the best runs. As a rule of thumb, take whatever dps you're doing and approximately double it to get an idea about running in one of the well-oiled premades. A 1-minute run, which is the high end, means that any dps will look higher than it otherwise would and the debuff stacking means that even an average-dps person will perform well.

    From my personal experience, I get a spread of between 50k and 70k, with a fleetmate who gets around 80-90k in most runs. My fleetie then ran with a pre-arranged group coordinated by voice chat and got his personal record of 177k.
  • yreodredyreodred Member Posts: 3,527 Arc User
    So when doing mostly (or better said exclusively) PUGs what is considered a good value to aim at?
    Or is there a good benchmark for solo missions?
    "...'With the first link, the chain is forged. The first speech censured...the first thought forbidden...the first freedom denied--chains us all irrevocably.' ... The first time any man's freedom is trodden on, we're all damaged. I fear that today--" - (TNG) Picard, quoting Judge Aaron Satie

    A tale of two Picards
    (also applies to Star Trek in general)
  • edited April 2016
    This content has been removed.
  • postinggumpostinggum Member Posts: 1,117 Arc User
    In my experience with pugs, the group can easily change the parse by +/- 30%.
  • yreodredyreodred Member Posts: 3,527 Arc User
    postinggum wrote: »
    DPS groups stack debuffs waaaay more than pugs, including the insanely expensive kemo skills. Additionally time buffs for the team, so for example they can run with tac fleet up all the time, similar with the iconian 4 piece boost, which I think works out as 1.374*1.33 +83% for the entire run, similar can occur for the intel team penetration ability, what with everyone having AHOD.

    they practise exact positioning and timing of powers so some things like pets and fleet support aren't used on the first group but held back for the transformer and spheres. People will do 100+ runs just to get that down, if using the intel spec thay learn where the flanking facings are on ISA and attack those. Also they make sure to be shooting at close range often using the point blank(?) trait. They say piloting is the biggest issue.
    So if i understand this right, a 1xxK DPS run does not only say something about the quality and effectiveness of the build itself, but but also about the team and each players piloting skills?
    If so, what's the point of presenting a build with x DPS made in a mission where so many other factors can distort the result that much? It doesn't really give a clear information about the ship builds individual effectiveness or even pure firepower, it just says that 4 people where especially successful in pushing one team members DPS to ridiculous high numbers. (or am i wrong?)

    In order to get a clear result, shouldn't outside factors (like other players ship builds/piloting skills and Team play in general) be eliminated to make a DPS value more meaningful regarding the performance or effectiveness of a individual build?
    Wouldn't it be better, if DPS was measured in a appropriate designed solo mission?

    Don't get me wrong i'm not jealous or something, i couldn't care less about how good some players can work together.
    My point is when i see someone posting a ship build capable of doing 150k DPS for example, it doesn't really testify to me how much firepower that build can produce or how "effective" that build is in general (its general survivability, versatility in other missions, or even solo missions), it only measures how much DPS it can do in a certain (very narrow IMO) context.

    So are most published "ISA high DPS" builds are just a worthless pile of junk? Or are they in fact valuable numbers to work with, regardless of your play style or if you frequently play in a fixed team or not at all?


    BTW: can the games tooltip (when you hover over the ships weapons in space) be trusted?

    Sorry because of my bad English.
    I'm not used to write about technical stuff, i hope it's understandable anyway.
    "...'With the first link, the chain is forged. The first speech censured...the first thought forbidden...the first freedom denied--chains us all irrevocably.' ... The first time any man's freedom is trodden on, we're all damaged. I fear that today--" - (TNG) Picard, quoting Judge Aaron Satie

    A tale of two Picards
    (also applies to Star Trek in general)
  • tom61stotom61sto Member Posts: 3,693 Arc User
    yreodred wrote: »
    So if i understand this right, a 1xxK DPS run does not only say something about the quality and effectiveness of the build itself, but but also about the team and each players piloting skills?
    If so, what's the point of presenting a build with x DPS made in a mission where so many other factors can distort the result that much? It doesn't really give a clear information about the ship builds individual effectiveness or even pure firepower, it just says that 4 people where especially successful in pushing one team members DPS to ridiculous high numbers. (or am i wrong?)

    Not all builds, even in that exact team, would be able to hit those same numbers, so, yes and no.
    yreodred wrote: »
    n order to get a clear result, shouldn't outside factors (like other players ship builds/piloting skills and Team play in general) be eliminated to make a DPS value more meaningful regarding the performance or effectiveness of a individual build?
    Wouldn't it be better, if DPS was measured in a appropriate designed solo mission?

    Don't get me wrong i'm not jealous or something, i couldn't care less about how good some players can work together.
    My point is when i see someone posting a ship build capable of doing 150k DPS for example, it doesn't really testify to me how much firepower that build can produce or how "effective" that build is in general (its general survivability, versatility in other missions, or even solo missions), it only measures how much DPS it can do in a certain (very narrow IMO) context.

    Most people play in groups outside the mission-of-the-month, though, so a group score is somewhat indicative of what others playing like themselves would get out of it. Also,there's no official DPS check map, so most just use ISA. There's running Starbase 24 solo if you're Fed, but that leaves out KDF builds.
    yreodred wrote: »
    So are most published "ISA high DPS" builds are just a worthless pile of junk? Or are they in fact valuable numbers to work with, regardless of your play style or if you frequently play in a fixed team or not at all?

    They are regarded as more of 'starting point' from what I can tell , as I read more about DPS chasers than I actually participate. When listing recommended DPS for different Queues, people will list the approximate ISA DPS number to shoot for, though you'll likely need to alter the build to suit the new queue, like slotting a lot more durability for Hive Elite than a pure-DPS-numbers ISA build.
    yreodred wrote: »
    BTW: can the games tooltip (when you hover over the ships weapons in space) be trusted?

    Sitting idle in non-Sector Space, it's only a starting point for figuring out things. What your tooltip says won't be even close to what a parse will say your DPS is, there's tons of buffs and debuffs that you should be using in actual combat that the number you see is not showing (and often cannot show).

    yreodred wrote: »
    Sorry because of my bad English.
    I'm not used to write about technical stuff, i hope it's understandable anyway.

    Your English is fine, very understandable.

  • yreodredyreodred Member Posts: 3,527 Arc User
    Thank you for your explainations, very appreciated!
    Although using a team mission to benchmark a individual ship build still doesn't make sense to me.
    But i assume it is most likely another odd habit of MMORPG players, which i don't understand, lol.
    "...'With the first link, the chain is forged. The first speech censured...the first thought forbidden...the first freedom denied--chains us all irrevocably.' ... The first time any man's freedom is trodden on, we're all damaged. I fear that today--" - (TNG) Picard, quoting Judge Aaron Satie

    A tale of two Picards
    (also applies to Star Trek in general)
  • e30erneste30ernest Member Posts: 1,794 Arc User
    It's not just due to well-coordinated teams now.

    I've run with people that do 90-100k+ DPS in a PUG pretty consistently. And they are not glass cannons either (able to take 80-90% of all the NPC attacks in an ISA without dying). That's even without any coordinated team buffs. It's all good map knowledge, ability timing, build and positioning.
  • postinggumpostinggum Member Posts: 1,117 Arc User
    The 'meaning' and methods of the high dps runs are very contentious, some have done solo runs, others stack the group with Tholian Recluses as debuffing 'nannies' just to get 1 player that killer run, every methodology has supporters and detractors.

    I think there are a few basic truths:

    High dps players, however measured, do lots and lots of damage and have done for years, it went crazy with the romulans expansion.
    Give a high dpser the same ship as another player and they will do more damage with it.
    Using ISA as the single metric produces highly stylized results, as does reporting only the best runs.
    The dps builds are useful guides but have to adapted for your needs, including budget.
    Playing the same level thousands of times just to get a better rate of damage is.....
  • yreodredyreodred Member Posts: 3,527 Arc User
    e30ernest wrote: »
    It's not just due to well-coordinated teams now.

    I've run with people that do 90-100k+ DPS in a PUG pretty consistently. And they are not glass cannons either (able to take 80-90% of all the NPC attacks in an ISA without dying). That's even without any coordinated team buffs. It's all good map knowledge, ability timing, build and positioning.
    My point is to distinguish a good build from a bad one, things like map knowledge, team ability and positioning shoudln't be in that equation in the first place.
    It's like saying my car has "more power", simply because i can drive very fast from point A to point B at rush hour (without mentioning the route, amount of traffic, traffic lights and so on). That's all very vague and not especially helpful imo.
    Maybe for that player and his friends it can give them a vague hint on how to improve their results more, but to everyone else... :/


    postinggum wrote: »
    The 'meaning' and methods of the high dps runs are very contentious, some have done solo runs, others stack the group with Tholian Recluses as debuffing 'nannies' just to get 1 player that killer run, every methodology has supporters and detractors.

    I think there are a few basic truths:

    High dps players, however measured, do lots and lots of damage and have done for years, it went crazy with the romulans expansion.
    Give a high dpser the same ship as another player and they will do more damage with it.
    Using ISA as the single metric produces highly stylized results, as does reporting only the best runs.
    The dps builds are useful guides but have to adapted for your needs, including budget.
    Playing the same level thousands of times just to get a better rate of damage is.....
    So are there other missions or methods to measure a builds effectivity (which would make more sense)?
    "...'With the first link, the chain is forged. The first speech censured...the first thought forbidden...the first freedom denied--chains us all irrevocably.' ... The first time any man's freedom is trodden on, we're all damaged. I fear that today--" - (TNG) Picard, quoting Judge Aaron Satie

    A tale of two Picards
    (also applies to Star Trek in general)
  • dareaudareau Member Posts: 2,390 Arc User
    yreodred wrote: »
    e30ernest wrote: »
    It's not just due to well-coordinated teams now.

    I've run with people that do 90-100k+ DPS in a PUG pretty consistently. And they are not glass cannons either (able to take 80-90% of all the NPC attacks in an ISA without dying). That's even without any coordinated team buffs. It's all good map knowledge, ability timing, build and positioning.
    My point is to distinguish a good build from a bad one, things like map knowledge, team ability and positioning shoudln't be in that equation in the first place.
    It's like saying my car has "more power", simply because i can drive very fast from point A to point B at rush hour (without mentioning the route, amount of traffic, traffic lights and so on). That's all very vague and not especially helpful imo.
    Maybe for that player and his friends it can give them a vague hint on how to improve their results more, but to everyone else... :/

    I've seen it stated that the difference between a 30k build and a 100k build, sadly, is things like map knowledge, ability timing, positioning.

    Map knowledge aids DPS because while the enemy HP pool (and therefore damage dealable) pool is the same from run to run, killing a mission in 1 minute produces DPS numbers twice as high as a 2 minute run (gobs / 60 > gobs / 120).

    Positioning is important for a number of factors, mostly making sure to get flanking bonuses and to minimize weapons dropoff. While my math's hazy at the moment, taking like a 500 DPS beam, trebling the damage because of 125 weapons power = 1500 DPS per array - watching even 75 or 150 (5%-10%) of this "evaporate" due to range adds up over the course of a run - exaggerated because most people find ways to overcap weapons at 130 (more DPS per shot) and get flanking bonuses - what "seems" to be an innocent 75's now inching closer to 200 lost per shot, per weapon. 5x beam Scimitar? ~1000 DPS 'lost' just because someone was off by .1km... Did I mention crits yet?
    yreodred wrote: »
    So are there other missions or methods to measure a builds effectivity (which would make more sense)?

    Sadly, no. Because the main "semi reliable" static testing ground (the immortal borg cube in that mission where you were about to enter fluidic space) has been replaced, and of course a static testing ground does not allow the screams of "we've perfected our flying times and positions" - I've "pushed" 12k-15k against the Elite Tac Cube and Gateway in ISA, yet "scored" a 6k overall, because I'd do silly things like slap a GW on the sphere train, not paying attention to Mr. "I'm gonna vape the transformer while you're not shooting at it", or die because I was in range of said sphere train when vapey causes them to all turn on their weapons...
    Detecting big-time "anti-old-school" bias here. NX? Lobi. TOS/TMP Connie? Super-promotion-box. (aka the two hardest ways to get ships) Excelsior & all 3 TNG "big hero" ships? C-Store. Please Equalize...

    To rob a line: [quote: Mariemaia Kushrenada] Forum Posting is much like an endless waltz. The three beats of war, peace and revolution continue on forever. However, opinions will change upon the reading of my post.[/quote]
  • e30erneste30ernest Member Posts: 1,794 Arc User
    yreodred wrote: »
    My point is to distinguish a good build from a bad one, things like map knowledge, team ability and positioning shoudln't be in that equation in the first place.

    Then actual runs in maps like ISA shouldn't be your metric. You can also do the math for a theoretical damage output of a ship. But that is only if you know the formulas and have the patience to compute for weapon damage.
    yreodred wrote: »
    It's like saying my car has "more power", simply because i can drive very fast from point A to point B at rush hour (without mentioning the route, amount of traffic, traffic lights and so on). That's all very vague and not especially helpful imo.
    Maybe for that player and his friends it can give them a vague hint on how to improve their results more, but to everyone else... :/

    Using your analogy, a ship's theoretical DPS (one you can compute for using your item stats) would count as your "power". DPS runs on the other hand are more like racing qualifying runs where you try to maximize what you get from that ship's potential.

    Think of it this way, you and I can both get into a modern Formula 1 car and drive a certain lap time, but we won't be able to match the lap times of someone like Lewis Hamilton or Sebastian Vettel on the same car. You have the car's power/downforce (theoretical performance) vs our actual driving (DPS output from a run).

    Like any run, you can get a good idea on your average performance once you have a big enough sample size. That's why some of us mix PUG and channel-organized runs. Those are 2 very different environments where channel runs are more predictable, following a script provided the rest of the team knows what they are doing and PUG runs give a degree of unpredictability.

    That's why DPS in a channel organized run is more consistent. You'll usually fall into a predictable range because you are essentially following a script. Fly to point A activate XYZ, move to point B activate TUV and so on. And that is the real secret to those high DPS runs. They are following a script, a fixed flight plan all made to maximize their weapons coverage, buff uptimes and minimize time to the next targets.

    PUG runs on the other hand have a lot of unpredictable variables, like having to chase spheres because a guy was pushing stuff away with Tractor Beam Repulsors or HY torps with Overwhelming Force for example. Or groups that don't have enough DPS to actually complete the objectives so you'll have to fly more defensively due to all the aggro you are getting while trying to contain the situation. Sometimes you'll run into a PUG with very high DPS players, so you'll have to fly more aggressively just to keep up. That unpredictability makes PUGs fun for me.
  • yreodredyreodred Member Posts: 3,527 Arc User
    Thank you guys your all your explainations and patience with me. :)
    e30ernest wrote: »
    Using your analogy, a ship's theoretical DPS (one you can compute for using your item stats) would count as your "power". DPS runs on the other hand are more like racing qualifying runs where you try to maximize what you get from that ship's potential.

    Think of it this way, you and I can both get into a modern Formula 1 car and drive a certain lap time, but we won't be able to match the lap times of someone like Lewis Hamilton or Sebastian Vettel on the same car. You have the car's power/downforce (theoretical performance) vs our actual driving (DPS output from a run).
    This makes sense to me.

    The problem i have is, how am i supposed to know which combinations of equipment (sets, weapons, consoles) are more effective than others?
    If high dps builds are more made of teamwork, positioning etc. how am i supposed to filter out useful information for my own ships?
    (As much as i understand, many consoles, sets and so on work differntly than stated on its tooltip (like x % dmg bonus which is in reality much less for example))

    e30ernest wrote: »
    Like any run, you can get a good idea on your average performance once you have a big enough sample size. That's why some of us mix PUG and channel-organized runs. Those are 2 very different environments where channel runs are more predictable, following a script provided the rest of the team knows what they are doing and PUG runs give a degree of unpredictability.

    That's why DPS in a channel organized run is more consistent. You'll usually fall into a predictable range because you are essentially following a script. Fly to point A activate XYZ, move to point B activate TUV and so on. And that is the real secret to those high DPS runs. They are following a script, a fixed flight plan all made to maximize their weapons coverage, buff uptimes and minimize time to the next targets.

    PUG runs on the other hand have a lot of unpredictable variables, like having to chase spheres because a guy was pushing stuff away with Tractor Beam Repulsors or HY torps with Overwhelming Force for example. Or groups that don't have enough DPS to actually complete the objectives so you'll have to fly more defensively due to all the aggro you are getting while trying to contain the situation. Sometimes you'll run into a PUG with very high DPS players, so you'll have to fly more aggressively just to keep up. That unpredictability makes PUGs fun for me.
    I too prefer the unpredictability of a PUG over a scripted high DPS run any day. It's much more fun, ESPECIALLY when things do NOT go smooth. :)
    It's just sad that so many queues are empty nowadays.



    "...'With the first link, the chain is forged. The first speech censured...the first thought forbidden...the first freedom denied--chains us all irrevocably.' ... The first time any man's freedom is trodden on, we're all damaged. I fear that today--" - (TNG) Picard, quoting Judge Aaron Satie

    A tale of two Picards
    (also applies to Star Trek in general)
  • e30erneste30ernest Member Posts: 1,794 Arc User
    Your best bet is to do the math and/or test and test and retest. That's how the real ship builders in this game do it. They do a lot of research, testing (both in STFs or PVP zones for a more controlled environment) and practice.
  • postinggumpostinggum Member Posts: 1,117 Arc User
    The essential differences between dps channel builds and other builds are: budget, survivability and peak speed.

    Budget: they use the best of everything they can get their hands on, which is why the plasmonic leech is 70m, inspirational leader 250m etc, whatever the build there are expensive toys to make it work better. For example TBR does very nice damage on a part gen build, but as noted above, spreads the targets about and slows down the run, the 50m+ doff graga mal(?) turns the repulse into an attract so the spheres clump together and mr part gen gets his biggest hit ability to work for the team. Best rom boff in game has the superior crit boost AND takes the ambush damage to 45%, only available from 15k zen cstore delta alliance pack.

    The lessons players on a budget can get include:
    what things to upgrade first - weapons, which mods should those weapons have, how to arrange them on a ship etc.
    How come they use leech - power is a key to power, if in doubt go for the power boosting option.
    EPS is DPS, don't be afarid of using a pure eps console or plasma-integrated core/singularity
    CritH and critD - but how to balance these, build CH and then start adding CD around I think 20, do not just stack CH with low CD.
    DPSers use wildly expensive doffs, what to do? see how much the blue or green version are and how much of a dropoff there is, sometimes 80%+ price reduction for 20% performance loss.
    Should I invest in space set X? well the dps guys will test and report, and the word for energy builds is go Iconian.
    Starship traits - at 3k zen a pop cstore ship with potentially nice trait are too expensive to buy and test, fortunately the dps channel guys will do it for us.



    They use consumeable devices/batteries to up that peak a bit more , +20% damage for 30sec or dueterium surplus to fly from one side to another. No one is going to burn through consumeables hopping into a red alert or on a featured episode, but when you want to blow stuff up quickly eat those batteries.

    Surviveability, simply put in a run lasting a minute the borg die you probably won't, out dps others in a pug and you'll get 50%+ of the total enemy damage directed at you. So a dps channel build with 1 heal and no armour is not going to work, however their use of speed and cloaking/de-cloaking does help keep them alive.

    Peak speed, you can't kill stuff from 30k away, maybe dueterium surplus is too much of effort, how about: aux to damp, with that nice doff helping damage resistance with high up-time.
    EPE
    Upgrading engines to mk 14
    Taking speed boosting traits


    In terms of measurement I'd rather they had a decathlon type approach instead of a 100m sprint, getting a bunch of gamers to agree what that would be is more or less impossible - there are two leagues because they couldn't play nice over what their sprint should be like.
  • snoggymack22snoggymack22 Member Posts: 7,084 Arc User
    yreodred wrote: »
    So are most published "ISA high DPS" builds are just a worthless pile of junk? Or are they in fact valuable numbers to work with, regardless of your play style or if you frequently play in a fixed team or not at all?

    No. The information is valuable. The context is still important. Remember at its core DPS parsing is just a test. And the exercise is really about testing and consistency. I started using parses back when I played Everquest. There were specific mobs and such where you could essentially be in combat for very very long stretches of time. The target's regen was too high to actually kill it, but it's damage wasn't high enough to overcome you're own regen. These targets were great for testing.

    The goal back then was to work on your gear, and your ability rotations to maximize your own damage output in real encounters. The target's had some specific issues (insanely high resists, etc) so that the damage output did not translate directly to what you would see when you went on your next dragon raid. But with everyone using the same targets, the baseline was consistent. For melee DPS mostly. And when you could run a 20 minute parse, getting millions of data points, it had value because it was a baseline to compare adjustments you were making.

    So it was great for deciding, Weapon X or Weapon Y? New Helmet improves DPS stat, is it going to help? My kick ability is adding how much damage to my rotation?

    The competitive nature eventually came out. We all want to be the best. And so parses on raid encounters, parses to prove you had the best weapons, etc, etc, etc.

    So what's my point? ISA is consistent. So it's a great baseline. The data isn't garbage. It is tuned specifically to that encounter. And it's one of the oldest encounters in STO's "endgame." So your 100k DPS won't translate to the new Featured Episode. But your 100k BUILD will be able to handle the new Featured Episode just fine.

    Parsing is about testing and finding a measuring stick. That's all. Treat it as such and you can find value in it.

    [SIGPIC][/SIGPIC]
  • yreodredyreodred Member Posts: 3,527 Arc User
    yreodred wrote: »
    So are most published "ISA high DPS" builds are just a worthless pile of junk? Or are they in fact valuable numbers to work with, regardless of your play style or if you frequently play in a fixed team or not at all?

    No. The information is valuable. The context is still important. Remember at its core DPS parsing is just a test. And the exercise is really about testing and consistency. I started using parses back when I played Everquest. There were specific mobs and such where you could essentially be in combat for very very long stretches of time. The target's regen was too high to actually kill it, but it's damage wasn't high enough to overcome you're own regen. These targets were great for testing.

    The goal back then was to work on your gear, and your ability rotations to maximize your own damage output in real encounters. The target's had some specific issues (insanely high resists, etc) so that the damage output did not translate directly to what you would see when you went on your next dragon raid. But with everyone using the same targets, the baseline was consistent. For melee DPS mostly. And when you could run a 20 minute parse, getting millions of data points, it had value because it was a baseline to compare adjustments you were making.

    So it was great for deciding, Weapon X or Weapon Y? New Helmet improves DPS stat, is it going to help? My kick ability is adding how much damage to my rotation?

    The competitive nature eventually came out. We all want to be the best. And so parses on raid encounters, parses to prove you had the best weapons, etc, etc, etc.

    So what's my point? ISA is consistent. So it's a great baseline. The data isn't garbage. It is tuned specifically to that encounter. And it's one of the oldest encounters in STO's "endgame." So your 100k DPS won't translate to the new Featured Episode. But your 100k BUILD will be able to handle the new Featured Episode just fine.

    Parsing is about testing and finding a measuring stick. That's all. Treat it as such and you can find value in it.
    I think it depends on how you want to play.
    Do you do STF mostly with other high DPS players, who exactly know where to position and what powers to activate at the right time? Or do you like to play PUGs where everything is much more unpredictable?
    I think High DPS builds and especially their excessive high numbers work in a very narrow and optimized context (ISA and team), but outside of that they are (as you said) a questionable (at least to me) "measuring stick".

    What kind of use can someone get out of High DPS builds, even if that person isn't part of the High DPS club (or doesn't want to join)?
    Well i think now, equipment and what kind of BOFFs are obvious and DOFF powers and trait combinations too. But what i miss is often a certain versatility in these builds (how tanky is it, crowd control etc.).
    To me it would often be much more interesting to see when and first and foremost why the creator chose a certain BO power instead of another one and what the "heart" of a build is (does it only work with a certain DOFF or Trait for example?).


    Sure STO favours firepower Plus firepower with even more firepower on top of it.
    It's even worse, since so many people prefer to do "boring" STF runs with other high DPS aimed people where nothing unpredictable can happen, just to "artificially" squeeze out even more DPS. (Which is especially absurd and rediculus when the whole team is explicitly buffing and supporting just that one person for more DPS) Don't get me wrong if it makes them happy i don't have a problem with it.

    But that's not why I play STO.
    To be honest, i am not a friend of min/maxing. And i don't like to copy other peoples builds or playstyle just to get a higher DPS count. I know that i will never break records, since i do only PUGs, zones and solo missions but that's not my goal in the first place. In my opinion doing such missions require a very differnt kind of build, least of all a cruiser glass cannon type of build.
    Don't get me wrong i'm not a purist who says a Starfleet ship should be equipped only with phasers and quantum/photon torpedoes. I like to use and experiment with differend kinds of equipment and BO powers.

    Regading to be fit for the next FE, i never had any problems playing FEs with the kind of builds i prefer (on "Advanced" difficulty). Sure space combat isn't over within seconds, but to be honest i would find such short fights rather disappointing.

    My point was not necessarily to increase the firepower of my builds at any cost, but to know how to filter out useful information from other peoples builds. For example, what Deflector + Impulse + Shield are best for a certain situation and why? (do they really work as the tooltip states?)
    Why and in what context use a certain Weapon + Console combination or BO power X in favour of BO power Y?
    Is using some Krenim BOFFs in the new Skill system a viable alternative to A2B type of build?
    Maybe things have like the amount of equipment, Traits DOFFs and BOFF powers have become a bit too confusing and overwhelming. Even worse if things don't work as advertised, like wrong tooltips and game mechanics in general.
    Like "What? That one console does NOT add 20% damage, only "base" damage?" That kind of stuff, lol.


    Sorry for that wall of text, i hope someone found the patience and endurance to read it, lol.
    Have a great day!
    (and please ignore typos :/ )
    "...'With the first link, the chain is forged. The first speech censured...the first thought forbidden...the first freedom denied--chains us all irrevocably.' ... The first time any man's freedom is trodden on, we're all damaged. I fear that today--" - (TNG) Picard, quoting Judge Aaron Satie

    A tale of two Picards
    (also applies to Star Trek in general)
  • berytakberytak Member Posts: 24 Arc User
    I think there are only three general types of players. Let me explain it a bit.

    When you start a game, as complex in mechanics, as STO, you'll be at first overwhelmed (I know a lot ok ppl that even quit the game due to complexity). You choose some skills, add some boffs and don't even mind any doffs. In the beginning it works well and you can do the missions. But I guess at some point in this game, you'll get you a** handed by some enemy ... over and over and over again. So, what will you do?

    Type 1: Doesn't bother and plays as he played before, of even quits the game because of frustration.

    The other Players start to read and posts the forum and/or reddit or speaks with other players and tries to improve.
    Those players reaching this point seperate into the other two types.

    Type 2: Tries to min/max his char for max dps. This player goes for the current meta in game.
    Type 3: The type I'm counting myself, too. You get all this information from the different sources and try to build your own thing. Your own idea, that's not the current meta and you're going to optimize it. Maybe that's the reason why I'm flying all the ships none likes :blush:

    Either way if you belong to type 2 or 3, the essence to get the optimal out of you build is consistent data. And that's the point at which a parser comes into play.
    Yeah, it's nice to see, that your unusual build seem to work better than some other builds, but the essence to this is testing diffenten combinations of equipment, boff traits ans skills and doffs so see how they will affect your goal. I'm using my parser just to optimize my own idea I had and not as competative measuring stick.
    yreodred wrote: »
    (...) But that's not why I play STO.
    To be honest, i am not a friend of min/maxing. And i don't like to copy other peoples builds or playstyle just to get a higher DPS count. I know that i will never break records, since i do only PUGs, zones and solo missions but that's not my goal in the first place. In my opinion doing such missions require a very differnt kind of build, least of all a cruiser glass cannon type of build.
    Don't get me wrong i'm not a purist who says a Starfleet ship should be equipped only with phasers and quantum/photon torpedoes. I like to use and experiment with differend kinds of equipment and BO powers.

    Regading to be fit for the next FE, i never had any problems playing FEs with the kind of builds i prefer (on "Advanced" difficulty). Sure space combat isn't over within seconds, but to be honest i would find such short fights rather disappointing.

    My point was not necessarily to increase the firepower of my builds at any cost, but to know how to filter out useful information from other peoples builds. For example, what Deflector + Impulse + Shield are best for a certain situation and why? (do they really work as the tooltip states?)
    (...)

    So, to get straight to the point ... it depends :wink:
    There are a lot of ppl out there, that don't wanna min/max their builds and they have a special idea. Mainly those ppl explain what they intended to do with their build. See if your idea and theirs are similar and look carefully why they choose something over another. Or ask them, why they did it this way and not another. So the point is, that there are alot of ppl who will help you to optimize your build regarding your wishes and ideas and not in favour for max dps.
    I hope this will help you a bit.
    Cruiser Captain since 87706.01
  • yreodredyreodred Member Posts: 3,527 Arc User
    Thanks for all your answers!
    It really helped me to get a better view at this matter.
    "...'With the first link, the chain is forged. The first speech censured...the first thought forbidden...the first freedom denied--chains us all irrevocably.' ... The first time any man's freedom is trodden on, we're all damaged. I fear that today--" - (TNG) Picard, quoting Judge Aaron Satie

    A tale of two Picards
    (also applies to Star Trek in general)
Sign In or Register to comment.