Maintenance for the week of September 22:
• NA megaservers for maintenance – September 22, 4:00AM EDT (8:00 UTC) - 10:00AM EDT (14:00 UTC)
• EU megaservers for maintenance – September 22, 8:00 UTC (4:00AM EDT) - 14:00 UTC (10:00AM EDT)

Numbers Don't Mean Anything? What was I missing from ESO PvP

Zski
Zski
✭✭✭
Short story - [snip] looking forward to a new world of PvP experiences to get dunked on in. But this isn't a goodbye post and that's not why we're here.

We're here because I have to know. Like one solid, intelligible, actual answer as to why the PvP in ESO never made sense to me.

See a few days ago, I decided I'd come back and try to find my answers to the Ninja Havel Problem https://forums.elderscrollsonline.com/en/discussion/499667/ninja-havel-monster-pvp-perspective-and-questions#latest, or, for those that don't want to Read Stuff, "whycome there be mans what deal billionty but take notbillionty". The answers there were helpful in the vague, not-actually-particularly-helpful-or-real-answers way that comments and youtube videos and twitch streams tend to be. I.E. it's not bad advice, but the gap here isn't covered in those pages. "Practice your combos" isn't an answer when "my combos are executing but they don't matter" is the question. Likewise defensively. They feel good to say and feel wise to think and are probably *true* but they aren't an *answer*.

Few excerpts from my recent misadventures -

CP: enough to have all passives and 4 slotted in all trees.

Before my break (postnoProc): Imperial StamDK, 2h/DW, Deadly/Pariah/Malacath/1pc Balorgh. 4M3H. Deadly DW Nirn/sharp swords frontbar w/ Doubledot, Nirnhoned Vateshran 2h back. Dump my 3 dots into an enemy stamdk, fossilize, wind up heavy DW + Leap combo while he's breaking fossilize, nail it with all buffs up - and my guy shrugs it off, stays 85% or above while I hit my Executioner like a dummy having expected to do something. He proceeds to dot (i vigor), dot, I roll back, he catches me with Dizzy Dizzy, Leap...and I'm dead.

Yesterday: Imperial Stamcro in small group play (experienced and semiexperienced, regular PvPers. Not potat, been doing this on and off for years). NMA, Clever Alch back, Balorghs monster, Master's Sharpened 2h front. 5M2H. We chase into a tower to play nascar with an AD smallscale, scooby doo it to chase some around, and I find myself vs. a heavy looking stamboi. I've buffed up on my way up the stairs, throw some trops around him, pop Clever on my backbar in combat, summon BB, and opt to make him move to let BB connect by popping a 500 Ult Balorghs Colossus on him. Now he can move quickly out of the AoE and take BB, or block up and absorb BB but get nailed by the Colossus. So BB takes off as I'm moving in and Colossus is slamming and I'm throwing an LA+Cleave just to add *something* when I notice....he's just standing there. In everything. His healthbar didn't even move. I'm chugging moves into block but he can hold it forever it seems, then he swaps and dizzies me for like....more than all of that. One dizzy just evened out Caltrops/BB/Colossus/Cleave. We play footsy for awhile and he eventually falls when my roamers finish their 1v1s and climb up the tower but....what?

Just today: Breton Magblade. Malubeth, Stuhns, Pariah, vamp Stage 3. Sharpened inferno, Maelstrom resto back, 4H3L. There's a DC smallball outside bleaks, 3-5 players. There's a necro, rest of his group splits for a bit to pursue stragglers. My guy has blocks for actual days. At one point he drops it, and with Major Sorc and Mark up, I land the Concealed for Stuhns, and slip a Soul Harvest in...for a whopping 1.7k crit. Not blocking or anything, just...eats it. The whole 1.7 crit. My follow up Focus similarly tickles. Now, I'm in the high 20ks (25-29ish i think?) on pen when Stuhns is up and Breach is applied, and he was spiffing at the top of his healthbar so Undeath and Pariah aren't gonna do much at that point, so....what? Later that same man would smash one of my (medium heavy) groupmates with a 17k Blastbones, and later than that HIS colossus would take me from full with NB Resolve up to about 25-30% (thanks low health defenses, i guess). Wonder if he was using Clever/Balorgh?

There's an infinity more than this but....I mean you get the idea.

So, in a last ditch attempt....what did I miss when I played ESO? What about this game never made sense to me? What dead simple, perfectly smooth brain nonsense mistake was I making? Why when I used builds I found from content creators did their Surprise deal 10k and mine dealt 2k? Why was I topping participation/kills/dealing 1m+ dmg in a majority of my BGs and never mattered in CP Cyro? I have to know.

I have to know why my numbers never seemed to matter. I have to know why all those stupid hundreds of hours speccing builds and learning benchmarks and practicing combos never mattered in Cyrodiil. What'd I miss? What was the gap I couldn't close? I have to know.

And judging by a cursory google search, so do a lot of other people. Maybe help them while we're here. Or don't, I guess.

Happy hunting, may the numbers even once be in your favor.

[edited for goodbye/quitting ref]
Edited by ZOS_Icy on 1 August 2021 12:48
Dead account. Y'all deserve each other.

GLHF.
  • geonsocal
    geonsocal
    ✭✭✭✭✭
    ✭✭✭✭✭
    find easier pvp prey if winning is that important...there's plenty of us not so hard to kill players out there roaming around cyrodiil...
    PVP Campaigns Section: Playstation NA and EU (Gray Host) - This Must be the Place
  • Sluggy
    Sluggy
    ✭✭✭✭✭
    Numbers don't matter if your skills don't work. A lot of people don't want to admit it but all other things being equal, transit time to the server becomes the single deciding factor. If you opponent is always ten feet ahead of you, always getting their skills to go off first, and always able to react to your counters while you seem to be always out of range, and never able to block, dodge, or kite away from their sources of damage even when it's clear as day what is about to happen then you're probably just at a network communication disadvantage.

    There's basically only two ways to solve this. Get a better internet to ZoS's servers or make a build and develop a play style that can work around these issues. The real reason most players have been tanking up for the past two years straight is mostly just due to the fact that it's the only reliable way they can do anything. You don't have to worry about network transmission times if everything your build does is just sitting on the server waiting to be passively calculated.

    Note: I'm not talking about server lag here. I'm talking about a purely you-versus-them when it comes to communications speed to and from the server.
  • Waffennacht
    Waffennacht
    ✭✭✭✭✭
    ✭✭✭✭✭
    Desync lag etc is way easier to deal with on a tanky build. Also gives an advantage when facing less tanky specs.
    Gamer tag: DasPanzerKat NA Xbox One
    1300+ CP
    Battleground PvP'er

    Waffennacht' Builds
  • Zski
    Zski
    ✭✭✭
    Today: BG???

    Stamplar warping around at Mach 9 (clearly, significantly faster than sprinting Major Expo while he's casting), walking through DBs, eating Cutting Dive/Shalks/DB/Whirling Blades while also pummeling through Vigor/Artic Blast/Pariah/Evasion with an ult and 2 jabs. That's it. Nobody else around, nobody else on me. Dude tanks an entire rotation to the face, ignores damage entirely, doesn't fall below 95% HP and then 3 shots me through 30k hp, 25-28k armor, pariah, Evasion and 2 hots.

    Mods feel free to delete the thread. I no longer care what I was missing, as for at least some of these events there's simply no way this is legitimate. Either someone is still cheating or the game is so broken that attempting to navigate it is worthless.
    Dead account. Y'all deserve each other.

    GLHF.
  • divnyi
    divnyi
    ✭✭✭✭✭
    @Zski if you are this confident this is a game error, record it with win+alt+G and upload to Youtube.
  • Larcomar
    Larcomar
    ✭✭✭✭✭
    Not quite sure what you're wanting to hear. Is there a rational explanation for some of the things we see in Cyro / BGs? Sometimes, sure. Lag, desynchs, this set, that set etc. And above all ping. You can usually concoct some sort of explanation for just about anything. Yeah, there's a point where it just stretches credibility to breaking point. Players who take no damage, have unlimited sustain, run at cap, yet have massive damage.... But, realistically, have you ever played an online game where p[eople don't cheat / exploit? I mean, it's just par for the course. Zeni arn't going to do anything about it - it's not worth their while to tackle stuff like that, it's far more efficient to sweep it under the rug. If you come across someone who's too good to be true, just move on and go elsewhere. I get that's easier said than done in BGs, but Cyro's pretty good for it. There's always another fight you can go join. And if there isn't start your own...
  • Marcus_Aurelius
    Marcus_Aurelius
    ✭✭✭✭✭
    If you are on PC and still here I would advice you to install Combat Metrics and have it stream the damage you do/take on a chat window.
    It will show with a timestamp the skill used, the damage and if you hit someone who was blocking.

    I don't think cheating is a big thing here, analizyng those logs will help you understand how things went when something seemed strange.

  • Tatdad
    Tatdad
    ✭✭✭
    This game is like the matrix. Here’s what’s really going to noggle your noogin sort of topic. This thread right here.

    The real truth is, nobody wants to admit it or further investigate it but. It’s CHEATING. Yes. That’s right. There’s cheaters in the game.

    When you bring it up on Reddit you get downvoted. I’ll probably be the same here. But. OP. This is why your brain is having trouble logically accepting what’s happening. You have done everything right. No. You aren’t crazy regardless of the community gas lighting you with responses like practice your combos… I get the same responses.

    Just the other day something organic happened though which confirmed it for me and sealed the deal- playing on consoles with that audio bug where you can hear everyone ? I heard 2 guys talking about the cheats they were doing , heard them say “do you think we will get in trouble” and “man I almost feel bad” as he spoke with his group mate.

    I’ve always known it deep down inside whenever you get those moments that defy absolute all logic what so ever. So I just let it go because it’s never going to be fixed and it certainly won’t ever be fixed while the community doesn’t believe it
  • divnyi
    divnyi
    ✭✭✭✭✭
    Tatdad wrote: »
    This game is like the matrix. Here’s what’s really going to noggle your noogin sort of topic. This thread right here.

    The real truth is, nobody wants to admit it or further investigate it but. It’s CHEATING. Yes. That’s right. There’s cheaters in the game.

    Why nobody shows us proofs then? Any video of the person cheating?

    Nevermind that all calculations now happen on server (that's why we have lags), thus it should be impossible to cheat, because whatever you do on your machine is just an input for the server.
  • Daffen
    Daffen
    ✭✭✭✭
    Most likely you dont know how to min-max builds. All youtube builds are trash. The good players you fight have stats similar to other mix-maxers. I have 30k hp 40k resists at 100%, 13k pen and 6.3k wpn damage on a stamdk. 13.2k dizzy tooltip and 32k vigor fully buffed non combat frenzy or next to keep.

    Most important part of beeing unkillable is keeping buffs and hots up. Always use vigor before offensive combo and learn to adapt combos. Against squishy players i just dizzy into leap with vate 2h proc and they die after execute. Some tanky players need 3 dizzies before stun with leap to kill.

    Min-maxers or good theorycrafters dont give builds because they have spent weeks, maybe months perfecting it.
  • Zski
    Zski
    ✭✭✭
    Daffen wrote: »
    Most likely you dont know how to min-max builds. All youtube builds are trash. The good players you fight have stats similar to other mix-maxers. I have 30k hp 40k resists at 100%, 13k pen and 6.3k wpn damage on a stamdk. 13.2k dizzy tooltip and 32k vigor fully buffed non combat frenzy or next to keep.

    Most important part of beeing unkillable is keeping buffs and hots up. Always use vigor before offensive combo and learn to adapt combos. Against squishy players i just dizzy into leap with vate 2h proc and they die after execute. Some tanky players need 3 dizzies before stun with leap to kill.

    Min-maxers or good theorycrafters dont give builds because they have spent weeks, maybe months perfecting it.

    1. That's not even minmaxed Stats. That's not even really A Lot. That's kind of par for the course statblock and +/- where I'm usually at. Vigor's kind of high I guess?

    2. Missed the entire first line. "Practice and adapt your combos" doesn't do anything when your "combos" don't do anything. I'm executing them. They don't matter. Nothing I do matters. Nothing anyone does matters.

    3. Multiple examples of me having hots and defensive buffs up in the OP. You either didn't read or simply don't know so fall back on the exact kind of "i guess true but not particularly applicable or relevant" truisms I mention in the OP. "You need to learn how to anicancel and line up burst" isn't a good answer to "I dumped a billion tooltip burst damage into an enemy that didn't lose more than 5% hp".


    Honestly I'm not sure if it's cheating or just the game not working, but after ~6 years and countless hundreds or thousands of hours there's enough in the PvP that simply does not or cannot make sense. Enough of these threads with nonsense answers because when grappling with "someone had 8k WD and a ton of pen and they full dumped into a man that didn't care", there isn't a good answer for why that's possible, so it gets immediately rejected as "well that cant be true" instead of "if that is true there is something terribly wrong."

    Hint: Something is terribly wrong.

    I'm just no longer interested in finding out what that is. FWIW it probably isn't cheating, at least not regularly.

    Good luck PvP.
    Dead account. Y'all deserve each other.

    GLHF.
  • Zski
    Zski
    ✭✭✭
    If you are on PC and still here I would advice you to install Combat Metrics and have it stream the damage you do/take on a chat window.
    It will show with a timestamp the skill used, the damage and if you hit someone who was blocking.

    I don't think cheating is a big thing here, analizyng those logs will help you understand how things went when something seemed strange.

    Did that. Unfortunately I don't need /cmx to tell me I did 1k damage with my ulti and then took 15k from blastbones. I can see that on my bar. I can tell if they're blocking because the game tells me they're blocking when they do it.

    I appreciate the honest attempt at helping someone in the forums though, and it is good advice for anyone that wants to try and get better from the ground floor.

    Good luck bud. You rock.
    Dead account. Y'all deserve each other.

    GLHF.
  • Daffen
    Daffen
    ✭✭✭✭
    You say numbers doesnt mean anything, but dont even post stats of your character or screenshots/videos. Cp's matter alot and so does skill setup. Show some stuff and i might be able to answer your questions.
  • Zski
    Zski
    ✭✭✭
    Daffen wrote: »
    You say numbers doesnt mean anything, but dont even post stats of your character or screenshots/videos. Cp's matter alot and so does skill setup. Show some stuff and i might be able to answer your questions.

    Hundreds of builds over the years. Few examples in OP. Even named skills. CP is mix and match any four, with one almost always being defensively dedicated. AoE DD characters have the AoE and DD CP. Nightblades had crit and single target. CP 2.0 isn't rocket surgery.

    You know what doesn't matter a lot? 8k WD and a ton of pen. 40k Resists with undeath and evasion active vs. AoEs in a BG. Major Evasion and sprinting. like +60% healing. None of it matters until it does, totally arbitrarily. Nothing feels worthwhile, nothing is intuitive, nothing is relevant. It's made up. Stamblades were in Fortified Brass and Beekeeper twoshotting kids also in tank sets. The numbers don't matter.

    I'm not interested in screenshots A. because it no longer matters and B. because you're looking to nitpick like, "oh well you're missing 5% optimization here and maybe 5% more healing would be better than 3% less damage tak-" stop. Please. Stop yourself. Thank you but stop.

    2% min/max isn't why I can dump 200k tooltip into someone and they don't register more than 5% hp loss. It isn't why I had 25-30k pen and crit soul harvest someone (that just blasted another character for 17k with a blastbones) for 1.7k and followed it up with a Focus shot for like, ~2k. You're looking for molehills to explain mountains. I appreciate the effort for what it is here, but if your intent is to actually help anyone else that feels the way I do going forward, this isn't a helpful exercise. It just isn't. I'm sorry. You're going to suggest one 2% change and then another and then another and then another and they're going to do all of them and at the end literally nothing will change and they're going to 1.7k crit someone that twoshots them.

    Start with "why a single player can tank 3-4 veteran PvPers with playable builds for an indefinitely long period of time while also dealing 17k with a blastbones at will" and go from there. Because that's what the question is. We're not even going to get to the guys that can endlessly kite/tank like 10 people because no, not all 10 people are potatoes. Thats a fiction you use to duct tape your game-world view together and it's a lie and you know it. We'll settle for 3-4 known regularly decent performing quantities. Start there.

    I will no longer be monitoring this thread. GLHF dealing with the Ninja Havel Monsters.



    If you wound up here from a google search late one night because you're neck deep in frustration, random ESO PvPer, just know you're not alone. But also probably won't find your answers here.
    Dead account. Y'all deserve each other.

    GLHF.
  • Jameson18
    Jameson18
    ✭✭✭✭✭
    Daffen wrote: »
    You say numbers doesnt mean anything, but dont even post stats of your character or screenshots/videos. Cp's matter alot and so does skill setup. Show some stuff and i might be able to answer your questions.

    He gave stat rundowns in the OP.
  • fred4
    fred4
    ✭✭✭✭✭
    ✭✭✭
    To the OP: Your builds sound pretty OK, but that's not to say they are. The devil is in the details. Damage, this patch, can be pretty insane. I run a Pariah build on my stam DK, but just saying that doesn't really cover much. If I told you I had 35K of both resistances on my back bar, before Pariah kicks in, that's saying a bit more. Note: both, which is not that easy. Resistances work best when you stack them and this patch being so insane, you have to stack them that high, if you go that route, and you can still feel pretty squishy. CP also come into play. Going all damage on the blue CP is not necessarily the thing to do.

    If you are on PC, you have to get a combat log addon, such as the one included in FTC, and analyze your fights on a case by case basis. I do this regularly when something feels amiss and I could not find a case of cheating in years. What comes to mind when someone ignores or minimises all your damage is that they have a temporary buff, such as:
    • A DK running Corrosive Armor.
    • The 1 Hand + Shield ult.
    • +20K temporary resistances from the Juggernaut set.
    • Meridia's Blessed Armor, dodging all your attacks entirely for 5 seconds.
    Some people are using the latter options, because there is so much damage, this patch. They also tend to lure you into situations where they stump you, but they cannot do it all the time. Attacking their stamina sustain (regular CC) is still a good idea.

    I don't tend to play tanky builds myself. This patch is an exception ... I don't think Pariah is even all that great, but I find I need it on some classes, like my stam DK. I'm more into sustain and movement-based builds. I wouldn't run Pariah on a nightblade. I actually back bar Juggernaut on my magblade, because her time to kill is a bit too long and I need a defense option. It's (still) a Caluurion build and that is not all that great. Bombing, I know, is better, even 1v1, albeit more strategic. A straight up heavy attack (opener) damage magblade may actually be better ... but not with Pariah. On the stamblade I also just have more damage. Surviving by pressuring people seems eminently viable on that class.

    Your main problem may be a mental block. When you notice someone doesn't take any damage, that can often be a temporary combination of buffs. Don't fall on your sword, because you think victory is owed to you or something. I'm somewhat prone to doing that. If they only take 1.7K from Soul Harvest / Incap, that sounds awfully like Corrosive Armor. Whatever the reason, you have to back off. The difference between a fully-buffed PvPer and the noob who stands around and eats 18K Incaps (they do) is huge. That's pretty normal and there are reasons. Those reasons are impossible to figure out without studying in-depth combat logs on a case-by-case basis. I don't mean the useless ZOS death recap. You need to look at an actual second-by-second log file.

    Aside from that, if you play in Cyro at prime time, then lag throws everything out the window. Positional desyncs mean you're not seeing where the other player actually is and sometimes there are health desyncs, which means you go from 100 to 0. The combat logs usually show you that did not just happen in one second, it just meant the UI was laggy and you didn't see you were in danger.
  • Daffen
    Daffen
    ✭✭✭✭
    Jameson18 wrote: »
    Daffen wrote: »
    You say numbers doesnt mean anything, but dont even post stats of your character or screenshots/videos. Cp's matter alot and so does skill setup. Show some stuff and i might be able to answer your questions.

    He gave stat rundowns in the OP.

    He gave almost no stats, ye he gave his sets but thats only like 1/4th of the build. He tries to play dot dk when it sucks and asks why he does no damage, doesnt understand that going full damage on a stamcro will still die to 2 dizzy + leap. Magblade is just bad in general.

    There is alot of information missing, no proof of anything, no stats, no videos so we can review to help his situation. Hes just like every PvP'er that thinks they are good, but doesnt know that they miss half of game mechanics.
  • Zski
    Zski
    ✭✭✭
    Daffen wrote: »
    Jameson18 wrote: »
    Daffen wrote: »
    You say numbers doesnt mean anything, but dont even post stats of your character or screenshots/videos. Cp's matter alot and so does skill setup. Show some stuff and i might be able to answer your questions.

    He gave stat rundowns in the OP.

    He gave almost no stats, ye he gave his sets but thats only like 1/4th of the build. He tries to play dot dk when it sucks and asks why he does no damage, doesnt understand that going full damage on a stamcro will still die to 2 dizzy + leap. Magblade is just bad in general.

    There is alot of information missing, no proof of anything, no stats, no videos so we can review to help his situation. Hes just like every PvP'er that thinks they are good, but doesnt know that they miss half of game mechanics.

    Should've taken my own advice and stopped checking.

    Relevant stats were covered in the blurbs. The Dot DK, even if it "sucks", does more than 5% with those abilities in that time frame.

    I do understand taking damage. I do not understand taking that much damage from people that also take an exceptionally small amount of damage. That doesn't jive.

    "Magblade is bad" isn't an answer to "good tooltip with almost 30k pen crits for 1.7k dmg". You might think it does. It doesn't. It's an unhelpful, intellectual lazy nonanswer that lets you stick to your scripts instead of actually engaging with the question you don't have an answer to.

    If you either lack the ability or the desire to actually read and engage with the material because all you have are prepared talking points and a poster refuses to engage with your unhelpful script, just say that and go.

    There is alot of information missing, no answer to anything, no thoughts, no videos or content you can provide to answer my question. You're just like every PvPoster that thinks they are good, but doesn't know that they miss the point.


    Just say "I don't know", because it's kind of obvious you don't.

    It was a mistake checking this. Peace.
    Dead account. Y'all deserve each other.

    GLHF.
  • Daffen
    Daffen
    ✭✭✭✭
    Again, you are providing nothing to the conversation for anyone to help you out.

    10k damage over 10 seconds is only 1k dps which is easy enough to heal through with hots, even if it was 40k over 10 seconds its still easy.

    You dont need alot of damage to kill people this patch, thats why people are tanky while still doing damage, i was killing players without damage sets (not experienced players they need damage set to kill)

    And your magblade statement theres too much information that has gone missing, i cant help you other than what @fred4 said about blocking, corrosive, snb ult, ironblood or other defensive abilities. When i had a pvp permablock build i was so tanky ppls dizzy were hitting for 200 through my block or force pulse doing 50 per element.
  • Theignson
    Theignson
    ✭✭✭✭✭
    This is one of the eternal eso pvp questions: how can that guy take so little damage but deal such huge damage?

    The assumption here is they take little damage because they are tanky. But if they are tanky, how can they deal huge damage? You can't have it all, right?

    Well, in some cases, they take little damage from you because your weapon damage/crit/ penetration at the point of attack is low. Conversely, they do big damage because theirs is high. Their hots/buffs are up, yours are not.

    But if their WD is high, shouldn't they take more damage from your attacks even though your WD is lower? It seems they should. And if your resists are high, shouldn't you take less dmg even if their weapon damage is higher? It seems you should.

    One thing I noticed is I didn't start really hurting these guys until I was getting 7k wd and 15k or more penetration. Then you could see some damage.

    I do think some players cheat. There have been many posts about it, but the most reliable indicator is they always perfectly weave their LA/HA no matter what conditions. This is achieved with a short macro that sets the perfect delay between your LA/HA and the ability so that they land together. You push one button, and both the lA and ability hit at almost the same time. You can see these people in Combat metrics "always" have the same (almost non existent) delay between the LA/HA and the ability hitting. The key is that the standard deviation of the delay is very low. This is the statistical sign of a macro. "Oh but what about lag it isnt a reliable statistic"-- not so. The lag affects all players equally yet the SD is still low despite lag. Humans , most humans, can't do this.

    Anyway simple macros wouldn't explain why they take no damage even when 5-10 people chasing them.

    But whenever it is brought up a bunch of people deny cheating is possible and say it is all just skill. I think 99% or more of players dont cheat.
    3 GOs, a Warlord, and bunches of prefects etc-- all classes...I've wasted a lot of time in PVP
  • fred4
    fred4
    ✭✭✭✭✭
    ✭✭✭
    Theignson wrote: »
    This is one of the eternal eso pvp questions: how can that guy take so little damage but deal such huge damage?

    The assumption here is they take little damage because they are tanky. But if they are tanky, how can they deal huge damage? You can't have it all, right?

    Well, in some cases, they take little damage from you because your weapon damage/crit/ penetration at the point of attack is low. Conversely, they do big damage because theirs is high. Their hots/buffs are up, yours are not.

    But if their WD is high, shouldn't they take more damage from your attacks even though your WD is lower? It seems they should. And if your resists are high, shouldn't you take less dmg even if their weapon damage is higher? It seems you should.
    And don't forget that high WD may also result in high healing. I'm a bit out of touch as to how everything scales now. I believe healing sets now scale with max stat. Healing skills, however, I believe still scale like damage skills. High weapon damage = large Vigor ticks.

    I don't really play tanky stamina builds enough to say how it's done. Even my stam DK wears Coward's Gear in addition to Pariah and is in 5x medium. It has a sustain and movement component to it's defense rather than the extra healing you get from heavy armor. I did, however, put together a 55K max mag warden - and that's before Northern Storm. It's next to impossible for a single player to take me down on that build. One guy managed. I think it was a stamplar. Stamblades could not get me. I subsequently adjusted my skills and sustain. It's a completely squishy light armor build that requires active defense. At that amount of magicka it felt like shielding builds of old, at least that day in IC. Strong 1v1, brittle against more players. Not terribly viable outnumbered, because it lacks the mobility of sorc. This is, however, an example of a build that hits hard and that appears really tanky 1v1, even though it is anything but. If you don't manage your sustain and your active defense, you're suddenly dead. I also see this from tanky stam brawler builds from time to time. They seem invincible until, suddenly, they go from 100 to 0 in 2 seconds. That's a sign of people having good defensive rotations and eventually running out of resources or otherwise screwing up without the number of enemies beating on them necessarily changing.

    EDIT: I guess the above could also be a sign of an eventual health desync happening to them, but I think it's a bit of both. It's most obviously the player's fault when they have mismanaged their stamina and can't break free from a CC, but I reckon there are other ways people screw up from time to time. Grrr. As I write this I realise people complain about unbreakable CC as well, of course. I have the benefit of mostly playing in IC (no lag), however I think I can tell an unbreakable CC from someone running out of stamina. When I experience unbreakable CC it generally means something like two seconds. It looks and feels different from a low stamina CC that can last longer and is never accompanied by weird visual effects, such as you lying flat on the floor.
    Edited by fred4 on 3 August 2021 15:34
  • Daffen
    Daffen
    ✭✭✭✭
    fred4 wrote: »
    Theignson wrote: »
    This is one of the eternal eso pvp questions: how can that guy take so little damage but deal such huge damage?

    The assumption here is they take little damage because they are tanky. But if they are tanky, how can they deal huge damage? You can't have it all, right?

    Well, in some cases, they take little damage from you because your weapon damage/crit/ penetration at the point of attack is low. Conversely, they do big damage because theirs is high. Their hots/buffs are up, yours are not.

    But if their WD is high, shouldn't they take more damage from your attacks even though your WD is lower? It seems they should. And if your resists are high, shouldn't you take less dmg even if their weapon damage is higher? It seems you should.
    And don't forget that high WD may also result in high healing. I'm a bit out of touch as to how everything scales now. I believe healing sets now scale with max stat. Healing skills, however, I believe still scale like damage skills. High weapon damage = large Vigor ticks.

    I don't really play tanky stamina builds enough to say how it's done. Even my stam DK wears Coward's Gear in addition to Pariah and is in 5x medium. It has a sustain and movement component to it's defense rather than the extra healing you get from heavy armor. I did, however, put together a 55K max mag warden - and that's before Northern Storm. It's next to impossible for a single player to take me down on that build. One guy managed. I think it was a stamplar. Stamblades could not get me. I subsequently adjusted my skills and sustain. It's a completely squishy light armor build that requires active defense. At that amount of magicka it felt like shielding builds of old, at least that day in IC. Strong 1v1, brittle against more players. Not terribly viable outnumbered, because it lacks the mobility of sorc. This is, however, an example of a build that hits hard and that appears really tanky 1v1, even though it is anything but. If you don't manage your sustain and your active defense, you're suddenly dead. I also see this from tanky stam brawler builds from time to time. They seem invincible until, suddenly, they go from 100 to 0 in 2 seconds. That's a sign of people having good defensive rotations and eventually running out of resources or otherwise screwing up without the number of enemies beating on them necessarily changing.

    EDIT: I guess the above could also be a sign of an eventual health desync happening to them, but I think it's a bit of both. It's most obviously the player's fault when they have mismanaged their stamina and can't break free from a CC, but I reckon there are other ways people screw up from time to time. Grrr. As I write this I realise people complain about unbreakable CC as well, of course. I have the benefit of mostly playing in IC (no lag), however I think I can tell an unbreakable CC from someone running out of stamina. When I experience unbreakable CC it generally means something like two seconds. It looks and feels different from a low stamina CC that can last longer and is never accompanied by weird visual effects, such as you lying flat on the floor.

    The way i do it on my stamdk is using pariah + vamp stage 3 (original i know), imperial so abilities cost the same. Lava foot soup for sustain full infused wep damage, lady mundus, 5m2h, bloodspawn, snb defending wep, hp+stam+phys resist pot. Gets me to 40k resists at 100% and with the wep damage i get 3.5k vigor ticks 5.6k crit next to keep, 7k average cauterize and rally heals hard af. Enjoy.
  • Sluggy
    Sluggy
    ✭✭✭✭✭
    Theignson wrote: »
    This is one of the eternal eso pvp questions: how can that guy take so little damage but deal such huge damage?

    The assumption here is they take little damage because they are tanky. But if they are tanky, how can they deal huge damage? You can't have it all, right?

    Well, in some cases, they take little damage from you because your weapon damage/crit/ penetration at the point of attack is low. Conversely, they do big damage because theirs is high. Their hots/buffs are up, yours are not.

    But if their WD is high, shouldn't they take more damage from your attacks even though your WD is lower? It seems they should. And if your resists are high, shouldn't you take less dmg even if their weapon damage is higher? It seems you should.

    One thing I noticed is I didn't start really hurting these guys until I was getting 7k wd and 15k or more penetration. Then you could see some damage.

    I do think some players cheat. There have been many posts about it, but the most reliable indicator is they always perfectly weave their LA/HA no matter what conditions. This is achieved with a short macro that sets the perfect delay between your LA/HA and the ability so that they land together. You push one button, and both the lA and ability hit at almost the same time. You can see these people in Combat metrics "always" have the same (almost non existent) delay between the LA/HA and the ability hitting. The key is that the standard deviation of the delay is very low. This is the statistical sign of a macro. "Oh but what about lag it isnt a reliable statistic"-- not so. The lag affects all players equally yet the SD is still low despite lag. Humans , most humans, can't do this.

    Anyway simple macros wouldn't explain why they take no damage even when 5-10 people chasing them.

    But whenever it is brought up a bunch of people deny cheating is possible and say it is all just skill. I think 99% or more of players dont cheat.

    There certainly are cheaters in the sense that many folks have been found to take advantage of exploits: self-ressing, broken sets, broken set-skill interactions, other campaign mechanics that would occasionally not work correctly, etc. But I think macros would be extremely low on the list of things that could be detected reliably.

    Your macro theory is very likely flawed reasoning based on the assumption that you are being told the timings of those attacks when the occurred. You are in fact being told the timing of those attacks when your machine received them and your client registered them. It is entirely possible for the server to be buffering up multiple messages and sending them off in a single package whereby your client then unpacks and relays them to you. It's also entirely possible that your client is buffering messages or simply processing them in a way that isn't how they were originally executed. Heck, the client could be buffering them before it ever sent them off to the server in the first place! (Not unlikely since many games tend have regular fixed intervals where all communications are packaged up and sent off in one go, often at a framerate that is much lower than what you'd expect)

    Without some detailed look at the actual data being sent and received as well as a look at how the client processes this data (which would require reverse engineering so I'm not one to be bothered with that) I couldn't say this for sure. But I just wanted to point out that the timing info as informed by the game client will almost certainly never reflect exactly how your opponent actually performed those attacks.
  • fred4
    fred4
    ✭✭✭✭✭
    ✭✭✭
    Sluggy wrote: »
    Theignson wrote: »
    This is one of the eternal eso pvp questions: how can that guy take so little damage but deal such huge damage?

    The assumption here is they take little damage because they are tanky. But if they are tanky, how can they deal huge damage? You can't have it all, right?

    Well, in some cases, they take little damage from you because your weapon damage/crit/ penetration at the point of attack is low. Conversely, they do big damage because theirs is high. Their hots/buffs are up, yours are not.

    But if their WD is high, shouldn't they take more damage from your attacks even though your WD is lower? It seems they should. And if your resists are high, shouldn't you take less dmg even if their weapon damage is higher? It seems you should.

    One thing I noticed is I didn't start really hurting these guys until I was getting 7k wd and 15k or more penetration. Then you could see some damage.

    I do think some players cheat. There have been many posts about it, but the most reliable indicator is they always perfectly weave their LA/HA no matter what conditions. This is achieved with a short macro that sets the perfect delay between your LA/HA and the ability so that they land together. You push one button, and both the lA and ability hit at almost the same time. You can see these people in Combat metrics "always" have the same (almost non existent) delay between the LA/HA and the ability hitting. The key is that the standard deviation of the delay is very low. This is the statistical sign of a macro. "Oh but what about lag it isnt a reliable statistic"-- not so. The lag affects all players equally yet the SD is still low despite lag. Humans , most humans, can't do this.

    Anyway simple macros wouldn't explain why they take no damage even when 5-10 people chasing them.

    But whenever it is brought up a bunch of people deny cheating is possible and say it is all just skill. I think 99% or more of players dont cheat.

    There certainly are cheaters in the sense that many folks have been found to take advantage of exploits: self-ressing, broken sets, broken set-skill interactions, other campaign mechanics that would occasionally not work correctly, etc. But I think macros would be extremely low on the list of things that could be detected reliably.

    Your macro theory is very likely flawed reasoning based on the assumption that you are being told the timings of those attacks when the occurred. You are in fact being told the timing of those attacks when your machine received them and your client registered them. It is entirely possible for the server to be buffering up multiple messages and sending them off in a single package whereby your client then unpacks and relays them to you. It's also entirely possible that your client is buffering messages or simply processing them in a way that isn't how they were originally executed. Heck, the client could be buffering them before it ever sent them off to the server in the first place! (Not unlikely since many games tend have regular fixed intervals where all communications are packaged up and sent off in one go, often at a framerate that is much lower than what you'd expect)

    Without some detailed look at the actual data being sent and received as well as a look at how the client processes this data (which would require reverse engineering so I'm not one to be bothered with that) I couldn't say this for sure. But I just wanted to point out that the timing info as informed by the game client will almost certainly never reflect exactly how your opponent actually performed those attacks.
    It's actually even "worse" than that. The server buffers keystrokes - not just network packets - for legitimate reasons, e.g. for the game to feel responsive. It possibly buffers network packets and bundles data as well, like you have suggested. However you don't need to look that far to explain why the macro theory is entirely bogus.

    As we all know there is a GCD, the so-called global cooldown. You may only cast one skill per second. However you may and will pre-cast skills, so they automatically fire when the global cooldown is up. Every player does this all the time, whether they use macros or not and whether they are aware of it or not. After you have pressed a skill key there is a very short delay, something like 100ms or 200ms, where another, different keypress will be ignored. If you are spamming the same key over and over, I believe that timeout may be a bit longer, but still below 500ms. Let's take the first case though. 2 different skills / keys are involved:
    1. The player presses skill key A and skill A fires.
    2. For the sake of the argument the player presses skill key B 500ms later.
    3. The server says "Aha! The player wishes to fire skill B next." However the GCD is not yet up.
    4. The server buffers skill B until the GCD is up and fires it exactly then.
    5. If you change your mind and press a different skill key before the timeout is up, the server fires that skill instead.
    In other words, everyone can and will execute a perfect rotation if they press the next skill key any time within almost a full 1 second window. Furthermore they can just press that key repeatedly. This will result in the server firing the action exactly when the next GCD is up.

    Things get more complicated when you take light-attack weaving into account. I believe light attacks are on their own, independent 0.7 second timeout or something like that. In low lag it's still entirely possible to weave perfectly, however, unless you are what ZOS call a low-APM (actions per minute) player, for example if you are physically handicapped. That said, PvEers weave perfectly, or nearly so, on target dummies all the time. I believe the real difficulties come from indecisiveness, e.g. not knowing what to do as a new player and not having committed your combos and rotations to muscle-memory yet. They also come from the other demands the game places on you, such as the need to move, break free, dodge and to deal with lag.

    There is one edge case where I'd say macro use may be detectable and that's partially charged heavy attacks. If they came out the exact same every time, I might find that suspicious. In reality, though, the charge time should be detected at the server, otherwise a hacked client could just tell the server "Here: I've detected a full heavy attack" when that was not the case. Therefore heavy attack charge time should be dependent on the ping you had at that moment. Damage values are also highly dependent on your and the target's buff status. Damage of the same attack can vary tremendously based on things like Empower, 2H follow up buff, Clever Alchemist etc., weapon / spell damage enchants, Vulnerability, different penetration values, different target resistances on front / back bar or due to armor buff running out, and so on, and so on.
    Edited by fred4 on 4 August 2021 08:15
  • WeylandLabs
    WeylandLabs
    ✭✭✭✭✭
    3 words -

    Build * For * Lag *
  • Theignson
    Theignson
    ✭✭✭✭✭
    The whole point of statistics is to separate signal from noise.

    Let's start with the basics so we all know what we are actually saying.

    First: the point of LA or HA weaving is for the LA and ability to land at the same time, or nearly the same time. If done perfectly: LA, Pause (eg 600 msec), ability. It is easy to see that they land almost the same time on the target (test this on your dummy).
    The differential time between when the LA lands and the Ability lands, we call delta.
    Combat metrics sends a time stamp down to .001 seconds, for all actions. (You could ask whether the server or the client supplies the timestamp, but I think it is timestamped by the server-- would like to have this confirmed-- it doesnt really matter for the statistics).
    Delta is calculated simply by (time LA) - (time ability.

    You can make a simple macro with any programmable keyboard or mouse: 1) LA 2) pause 600 milliseconds 3) ability 1. You can test the pause length to determine the ideal pause which creates a delta = zero.

    Any experiment involves an experimental group and a control group. You can make your own control group by hitting your target dummy (X)times with your LA/ ability weave. Then, calculate Delta for each attack. The experimental group would be a player using the macro.

    The standard deviation (SD) is defined by using the difference between a sample value and the population value, divided by the number in the sample. So a bigger sample shrinks the SD, and obviously if most samples are close to the total mean of all the samples, your standard deviation is low.

    So you can imagine making a macro and repeating the test with the dummy. Then calculate the mean delta for each group and the standard deviation for each group.

    Not surprisingly, the SD of the experimental group, under the same conditions, is much lower than the control group. Put simply: a computer can perform the actions with much more uniformity than a person. That is why using macros is a cheat!.

    Next you can compare the means etc of two groups and arrive at a T statistic to test the probability of the null hypothesis. This gives you the likelihood that the difference between the control and experimental group is due to "chance".

    The main point is that in this, as in every experiment, there are factors that distort the numbers. In statistics this is described as the difference between a sample mean and the "true" population mean.

    The key in science is to ask whether the conditions affecting the numbers (in this case lag, server buffering, etc) are different between the control and experimental groups. In other words: is the expected SD the same for the two groups? You can do the statistics whether they are expected to be the same or not. But in general, with a large enough sample sizes, we expect that all the factors extrinsic to the actual experimental test are randomly distributed and thus (in a large enough sample!) evened out between the control and experimental group.

    That is why, in this scenario as well as in actual scientific experiments, you can easily distinguish the signal from the noise.

    TLDR: statistics easily allows you to prove that any given player is using a macro, because the standard deviation of his time gap between light attack hitting and ability hitting, is very low.

    To put it very simply: if a player lands his LA and ability only .001 seconds apart(one one thousandth of a second) again and again (in a large enough sample-- not just once), every time in PvP while jumping, dodging, fighting 4 other players, sprinting , running and blocking (as opposed to when standing in front of a target dummy or standing in front of a tria boss being healed continuously), then it is suspicious for cheating.

    Yes and the analysis of Heavy attacks is similar
    3 GOs, a Warlord, and bunches of prefects etc-- all classes...I've wasted a lot of time in PVP
  • Waffennacht
    Waffennacht
    ✭✭✭✭✭
    ✭✭✭✭✭
    OP clearly has their mind made up.

    Im thinking if its cheating its only one or two people - in which case move on

    If its a large amount of the population than clearly its either their build or their skill - not way a ton of people are cheating.

    On console there isn't really a worry about cheating, ur pretty sure when u die its because of legit reasons
    Gamer tag: DasPanzerKat NA Xbox One
    1300+ CP
    Battleground PvP'er

    Waffennacht' Builds
  • Raeyleigh
    Raeyleigh
    ✭✭✭
    OP clearly has their mind made up.

    Im thinking if its cheating its only one or two people - in which case move on

    If its a large amount of the population than clearly its either their build or their skill - not way a ton of people are cheating.

    On console there isn't really a worry about cheating, ur pretty sure when u die its because of legit reasons

    Legit server dysfunction
  • divnyi
    divnyi
    ✭✭✭✭✭
    Theignson wrote: »
    To put it very simply: if a player lands his LA and ability only .001 seconds apart(one one thousandth of a second) again and again (in a large enough sample-- not just once), every time in PvP while jumping, dodging, fighting 4 other players, sprinting , running and blocking (as opposed to when standing in front of a target dummy or standing in front of a tria boss being healed continuously), then it is suspicious for cheating.

    I did macro for other games before, if you do keystrokes 0.001 seconds apart, the game won't recognize them at all. You need much higher delays.

    If you gonna differentiate macro based on delays and if the person sometimes drops LA, then it is very easy to program too :)

    Doing perfect light attack weaving after 2 years of play is a question of a muscle memory. Go ahead and claim that if a person does comboes in a fighting without dropping them, it is because they always do macro.
  • Theignson
    Theignson
    ✭✭✭✭✭
    "I did macro for other games before, if you do keystrokes 0.001 seconds apart, the game won't recognize them at all. You need much higher delays."

    Sorry that's not what I wrote.

    The .01 or .001 delta = (timestamp of Ability)-(timestamp of light attack). To get that with a macro you need a delay of around 0.6 seconds between keystrokes (I don't know the exact number since I never wrote macroes for ESO).

    "Doing perfect light attack weaving after 2 years of play is a question of a muscle memory. Go ahead and claim that if a person does comboes in a fighting without dropping them, it is because they always do macro."

    Again, not what I wrote. Everyone (even me lol) can land a "perfect" (delta = .01 seconds or less) weave sometimes.

    What you need to do is look at the standard deviation of the mean delta.

    That means you need a decent sample size.

    If a person "always" lands a perfect weave, even when hounded by 4-5 players, dodging, jumping, blocking etc etc -- again, it their weave is "always" perfect-- Meaning, one lasttime, that the standard deviation of the mean of a large number of their attacks is very low--

    That is when you suspect a macro. Like all statistics, it never "proves" it, but you can get a P value, which is an estimate that that the low SD arises randomly.

    If you never studied or used statistics, then the argument will be hard to understand.

    3 GOs, a Warlord, and bunches of prefects etc-- all classes...I've wasted a lot of time in PVP
Sign In or Register to comment.