Page 1 of 3 123 LastLast
Results 1 to 20 of 55
  1. #1
    Community Member
    Join Date
    Apr 2013
    Posts
    2,881

    Default Needed Change on the To-hit formula.

    Hi, the monster to-hit formula was changed to a ratio to avoid it being too lopsided with 95% avoidance.

    The ratio it was changed to is currently (monster to hit + 10.5) / ( 2 * Player AC).

    At the time it was changed, it made sense, and it made sense all the way through Stormhorns, since front line characters had around 130 AC at level 28 cap, and monsters with 150 to hit would have a 62% chance to hit such a player.

    Since that change, monsters to-hit has increased to 400+, but front line characters AC has only increased to around 170-200, which results in the lopsided monsters having 100% chance to hit players, capped at 95% due to missing on a 1.

    For example, here is a 200 AC character getting hit 95% of the time in Legendary Elite, while wearing Gaurdian of the Gates tank set and tower shield in Unyielding Sentinal:


    Here is a 170+ AC uber completionist character also getting hit 95% of the time against a dungeon boss with all AC past lives also in adamantine body (plate) with shield AC spell, EK % AC, and various other buffs:


    So now it is lopsided the other way, monsters now automatically hit not only through AC but literally every single defense an armored (non-AC stacked tank) character has. So I propose a new formula that doesn't go lopsided in any direction, that is:

    (monster to hit + 10.5) / ( monster to hit + Player AC).

    On a high end, where an 500+ AC stacked character has AC roughly equal to monster to hit, the result remains around 50% chance to hit. However, for everyone else, this means that their AC actually matters. Then 200 AC in plate with 500 mob to-hit is 28% avoidance, which is about the same as 25% dodge in cloth with 50 AC. A monk with 200 AC AND 25% dodge then has better defense than a wizard with 100 AC and 25% dodge.

    At level 28, plate had AC of 40 (Epic Calvary Plate level 28) and mobs had ~150 to hit. Now at 30 plate still has AC of 40 (Legendary Watch Captain's Platemail) but mobs have >400 to hit. Since you aren't updating player plate AC, you have to update the mob to hit formula. As a result, when I look at level 30 population it is a bit lean on the front line melee:



    Seriously, we can do better than 0 melee rogues, 0 barbarians, 0 melee rangers, 0 artificers, 1 medium/heavy armor fighter, 1 bard (unknown type), 1 druid, 2 monks, and 4 paladins, to a zillion casters and ranged. I mean, we can't do worse. Just try it.
    Last edited by Tilomere; 09-02-2020 at 08:24 PM.

  2. #2
    Community Member
    Join Date
    Jan 2015
    Location
    Seattle
    Posts
    548

    Default

    Yeah but then my first-level shifter monk's 113 AC doesn't do as much good... /sarcasm
    Last edited by Discpsycho; 09-04-2020 at 12:34 PM.

  3. #3
    Hero Propane's Avatar
    Join Date
    Mar 2006
    Location
    From Wisconsin, Live in Iowa, Vist Wisconsin
    Posts
    1,691

    Default

    I see you at R6 correct?

    I suspect MOB are buffed a lot...
    I suspect some defuffs on players that are not advertised...
    Sarlona - Guildmaster - Brotherhood of Redemption - ddoborguild.com - 2016 & 2017 Players Council --- Alts: Acetylene, Antimematter, CNG, Dilithium Crystal, EMF, EMPulse, Exothermic, Geothermal, Hexane, Hexyne, Hydropower, JA, Kerosene, LPG, Natural Gas, Nuclearpower, Propane, Solarpannel, Tidalpower, WASOB, Waulter, Windpower, Woodpile

  4. #4
    Community Member
    Join Date
    Nov 2009
    Posts
    546

    Default

    Agreed when things are too far out of balance people just tend to take the easiest route which is to never be in harms way in the first place hence lots of casters and ranged. Clearly the buff to melee 2h dps and cc via strikethrough and the nerf to ips was not enough to shift the meta for long. Anything that buffs melee non tank survivability slightly is a good thing.

  5. #5
    Community Member Hawkwier's Avatar
    Join Date
    Jul 2010
    Location
    Scotland
    Posts
    564

    Question What is going on here Devs?

    On the face of it, these result are very concerning.

    @ Tilo - I'd like to see some other data points at different difficulties please - Elite, R2, R4 - i.e. how does R factor into these results?

    @Devs - If these results are valid, then what is going on here?? It looks like AC is not providing the mitigation it should based on the stated formula, i.e. based on mob CR, which doesn't change with R.

    Simple question - Is this WAI?

    If not then get it fixed asap please.

    If so the please let us know.

    Many folks will have built AC into their mitigation, often at significant opportunity cost, and if it is indeed as useless as these result suggest, they have been misled. You need to be up front about this.

    The biggest impact, as noted above, is on melee, as they are the ones getting hit in the face most. That doesn't mean others are not affected though.

    If AC isn't any significant mitigation on damage based on R, then please let us know this.

    We can then likely figure out the "true" calculation for ourselves, and recognise AC as dead and a total waste of investment in R.

    I'm really hoping this is either an aberration in this data set, or an honest bug that needs a priority fix.

  6. #6
    Community Member
    Join Date
    Apr 2013
    Location
    Athens
    Posts
    2,186

    Default

    I did some testing for reaper:
    My character is level 8 with 38-39 AC. Against CR 11 monsters in Kind of a Big Deal (a kobold bruiser), I have a ~50% miss chance. (27.5 + 10.5) / (38 * 2) = 38 / (38 * 2) = 0.5
    Elite: 70 attacks, 30 hit, 40 miss (57% miss)
    R1: 60 attacks, 28 hit, 32 miss (53% miss)
    R2: 55 attacks, 27 hit, 28 miss (51% miss)
    R3: 50 attacks, 27 hit, 23 miss (46% miss)
    R5: 50 attacks, 24 hit, 26 miss (52% miss)
    R7: 40 attacks, 20 hit, 20 miss (50% miss)

    The error margin is big. However, if there was any reaper buff to the monster's to-hit, I would expect by R7 the miss chance would have gone way down to 20% or something. As such, I do not think that Tilo's results are due to reaper but likely due to the developers hard-coding a higher to-hit value of those bosses instead of using the formula of CR*2.5.
    My main server is Khyber. Have toons in almost every server for favor purposes. The Faltouts

  7. #7
    Community Member A-O's Avatar
    Join Date
    Apr 2013
    Posts
    467

    Default

    Quote Originally Posted by Faltout View Post
    I did some testing for reaper:
    My character is level 8 with 38-39 AC. Against CR 11 monsters in Kind of a Big Deal (a kobold bruiser), I have a ~50% miss chance. (27.5 + 10.5) / (38 * 2) = 38 / (38 * 2) = 0.5
    Elite: 70 attacks, 30 hit, 40 miss (57% miss)
    R1: 60 attacks, 28 hit, 32 miss (53% miss)
    R2: 55 attacks, 27 hit, 28 miss (51% miss)
    R3: 50 attacks, 27 hit, 23 miss (46% miss)
    R5: 50 attacks, 24 hit, 26 miss (52% miss)
    R7: 40 attacks, 20 hit, 20 miss (50% miss)

    The error margin is big. However, if there was any reaper buff to the monster's to-hit, I would expect by R7 the miss chance would have gone way down to 20% or something. As such, I do not think that Tilo's results are due to reaper but likely due to the developers hard-coding a higher to-hit value of those bosses instead of using the formula of CR*2.5.
    I also think they strayed from the formula with Legendary content. AC does not seem to do a hell of a lot in leg sharn, e.g.
    Formerly known as Absolute-Omniscience, co-creator of the old DPS calc.

  8. #8
    Community Member Hawkwier's Avatar
    Join Date
    Jul 2010
    Location
    Scotland
    Posts
    564

    Thumbs up

    Quote Originally Posted by Faltout View Post
    I did some testing for reaper:
    My character is level 8 with 38-39 AC. Against CR 11 monsters in Kind of a Big Deal (a kobold bruiser), I have a ~50% miss chance. (27.5 + 10.5) / (38 * 2) = 38 / (38 * 2) = 0.5
    Elite: 70 attacks, 30 hit, 40 miss (57% miss)
    R1: 60 attacks, 28 hit, 32 miss (53% miss)
    R2: 55 attacks, 27 hit, 28 miss (51% miss)
    R3: 50 attacks, 27 hit, 23 miss (46% miss)
    R5: 50 attacks, 24 hit, 26 miss (52% miss)
    R7: 40 attacks, 20 hit, 20 miss (50% miss)

    The error margin is big. However, if there was any reaper buff to the monster's to-hit, I would expect by R7 the miss chance would have gone way down to 20% or something. As such, I do not think that Tilo's results are due to reaper but likely due to the developers hard-coding a higher to-hit value of those bosses instead of using the formula of CR*2.5.
    Aye, that looks like constant results independent of R within error margins. Something else going on with Tilo's data set then. Thnaks!

  9. #9
    Community Member Powerhungry's Avatar
    Join Date
    Oct 2013
    Posts
    552

    Default tank?

    an end game, fully geared tank will have an AC well over 300 (as well as 300+ PRR and MRR)- my pajama wearing level 30 (non-completionist - non-optimized) monk has an AC of 210
    (Combat): You are hit by your knockdown.

  10. #10
    Community Member Bjond's Avatar
    Join Date
    Dec 2018
    Posts
    695

    Default

    Quote Originally Posted by Hawkwier View Post
    Aye, that looks like constant results independent of R within error margins. Something else going on with Tilo's data set then. Thnaks!
    Tilo's data is good. It's from L30. Faltout's data is inapplicable. It's from L8. The problem isn't with with AC in Heroic content or even in Epics. It's exclusively with L29+.

    My Thrower with about 160 AC & 41% dodge sub29 routinely stands in the middle of a herd, doesn't bother kiting, and doesn't get hit. The only stuff she has to worry about are very high STR mobs (that I assume have high to-hit from that STR). She can even stand in the whirling chains from Shadar Kai in Wheloon Mirror. Step up to L29+ quests and L29 gear (190 AC now) and everything will automatically hit her.

    I've seen Tilo's Bard (the one that he used for testing) also stand in stuff without taking damage in Wheloon, too. I don't know what stats it has sub29, though.

    Changing the L29+ to-hit from mobs would be very welcome. However, general changes to the way to-hit works would need very wide testing across all levels and difficulties. The potential for making things either too hard or too easy is high.
    Last edited by Bjond; 09-02-2020 at 11:33 PM. Reason: corrected poster reference

  11. #11
    Community Member timmy9999's Avatar
    Join Date
    Jan 2017
    Posts
    246

    Default You serious

    [QUOTE=Hawkwier;6356136]On the face of it, these result are very concerning.

    @ Tilo - I'd like to see some other data points at different difficulties please - Elite, R2, R4 - i.e. how does R factor into these results?

    @Devs - If these results are valid, then what is going on here?? It looks like AC is not providing the mitigation it should based on the stated formula, i.e. based on mob CR, which doesn't change with R.

    Simple question - Is this WAI?

    If not then get it fixed asap please.

    If so the please let us know.

    Many folks will have built AC into their mitigation, often at significant opportunity cost, and if it is indeed as useless as these result suggest, they have been misled. You need to be up front about this.

    The biggest impact, as noted above, is on melee, as they are the ones getting hit in the face most. That doesn't mean others are not affected though.

    If AC isn't any significant mitigation on damage based on R, then please let us know this.

    We can then likely figure out the "true" calculation for ourselves, and recognise AC as dead and a total waste of investment in R.

    I'm really hoping this is either an aberration in this data set, or an honest bug that needs a
    Who bothers with ac playing melee dps in this game? Its all about prr..and has been for a long time

  12. #12
    Community Member
    Join Date
    Apr 2013
    Posts
    2,881

    Default

    When I think of plate in DDO, this is what pops into my head now:


    It's too cute to unsee.
    Last edited by Tilomere; 09-03-2020 at 12:25 AM.

  13. #13
    Community Member A-O's Avatar
    Join Date
    Apr 2013
    Posts
    467

    Default

    Honestly. They just painted themselves into a corner with Martial past lives. They simply give waaaay too much AC. +60AC from past lives, when 3 action points give you 3 AC. Or a T4 twist for 5 AC. It gives more than the best full plate in the game... Including dex bonus.
    This makes it so that any build (regardless of class/items) will have a good standing AC, and if say, 200AC meant 50% mitigation. *everyone* could get there. Further, the past lives also absolutely trivialize early game (more so than any other past life)


    Thus I'd propose they significantly nerf the martial past lives to like 1AC/Past life (12 AC total). And also reduce mobs to hit by a significant amount of account for this. And then maybe we can see an endgame where random melees get hit maybe 60-70% of the time. Casters 95% of the time. Tanks 30% of the time. Or something like that.
    Formerly known as Absolute-Omniscience, co-creator of the old DPS calc.

  14. #14
    Community Member Bjond's Avatar
    Join Date
    Dec 2018
    Posts
    695

    Default

    Quote Originally Posted by timmy9999 View Post
    Who bothers with ac playing melee dps in this game? Its all about prr..and has been for a long time
    I read this in postings all over and bought into it as a new player. It's an actively harmful statement to new folk. The real truth is AC is King until slaughtered by Prince PRR at L29.

    Quote Originally Posted by A-O View Post
    Honestly. They just painted themselves into a corner with Martial past lives. They simply give waaaay too much AC. +60AC from past lives, when 3 action points give you 3 AC
    I'd support a Martial PL nerf, but only as part of a general AC and mob to-hit balancing. You're spot on that it's major effect is to render other sources of AC trivially ineffective by comparison.

  15. #15
    Community Member
    Join Date
    Apr 2013
    Posts
    2,881

    Default

    Quote Originally Posted by A-O View Post
    Thus I'd propose they significantly nerf the martial past lives to like 1AC/Past life (12 AC total).
    A better idea hit me overnight. If balance is the goal, we could just nerf dodge out of existence like self healing, AC, ghostly, and concealment. If we get the last 2monks, 4 paladins, druid, and medium armor fighter to reroll or quit, the game will have 0 melee characters, which would allow it to be perfectly balanced around ranged and casters.
    Last edited by Tilomere; 09-03-2020 at 09:35 AM.

  16. #16
    Community Member Avocado's Avatar
    Join Date
    Apr 2010
    Posts
    592

    Default This post is all sorts of misinformation.

    You cant make accurate statements about how ac interacts with mobs to hit when you only test it 6 times one raid boss and a few reaper 6 mobs. If you did this in real life you would the laughing stock of the science community and before that you wouldnt even get the funding to qualified to perform a study. You fail out high school stats before any of that if you did this.

    You need like at least 3 tests of 100+ hits to determine if ac is working correctly. Please do that before making invalidated claims about AC.

    AC is the 2nd most important physical defense factor after prr. Anecdotally, raid bosses on LH+ have way higher to hit then any most any r10 mob. I wont see many misses from rudus or gish but when you stick 300 ac in a reaper quest you will see many more. I can say that as a healer i see people with 500+ ac get hit way less in reaper PN and reaper content then my 300 ac tank does.

    Also do you have ANY proof that other then this baseless post of picture at 200 ac that the to hit formula for the reason you claim it was changed.
    At the very least I agree that AC is horribly implemented and needs to be better for those that dont have those ungodly op 12 ac past lives. AC for pls needs to lowered to maybe 2 per pl and then the ac curve vs attacker to hit needs to be smoothed out. Maybe formula 100/ (100+1/4ac) This gives 100 ac a 20% defense chance at 100 ac and and 50% at 400. But then how to incorparate to hit of attacker into it. IDK
    Last edited by Avocado; 09-03-2020 at 04:03 PM.

  17. #17
    Community Member Bjond's Avatar
    Join Date
    Dec 2018
    Posts
    695

    Default

    Quote Originally Posted by Avocado View Post
    as a healer i see people with 500+ ac get hit way less in reaper PN and reaper content then my 300 ac tank
    Reporting an impression based on 100s of attacks may not have the statistics for science, but it's still VERY useful to those of us who are wondering where the tipping point on AC might lie. It's not as if it's easy to just slap on a little gear and roam out to test 500+ AC.

    Without your anecdotal tip, making a tank is just "build and pray".

  18. #18
    Community Member
    Join Date
    Apr 2013
    Location
    Athens
    Posts
    2,186

    Default

    Quote Originally Posted by Bjond View Post
    Tilo's data is good. It's from L30. Faltout's data is inapplicable. It's from L8. The problem isn't with with AC in Heroic content or even in Epics. It's exclusively with L29+.
    I think there are several questions asked here:
    "Is the AC miss chance working as advertised?"
    - "Yes"
    - - "Does the to-hit value for monsters follow the observed formula of CR*2.5?"
    - - - "Yes
    - - - - "Are Tilo's results to be believed?"
    - - - - - "Yes"
    - - - - - - "Why do Tilo's results contradict the first two statements?"
    - - - - - "No"
    - - - - - - "What is that Tilo did to make his results invalid?"
    - - - "No"
    - - - - "Is the observed formula at least valid for most of the cases and what would be the common outliers?"
    - "No"
    - - "When and why did the AC miss chance change?"

    Starting with the inner questions:
    - "Why do Tilo's results contradict the first two statements?":
    Perhaps there is a bug with hich AC or high to-hit or some interaction with buffs. We can't do anything but bring it to the dev's attention.

    - "What is that Tilo did to make his results invalid?"
    Tilo posted reaper results and not a sequential battle log. I performed a test for reaper and saw that reaper does not screw with your miss chance. As for the sequential battle log, we'll have to take Tilo at his word that he did not ommit the times he was missed to post only the times he was hit. He also did not disclose whether the bosses had some kind of buff like dodge bypass, True Seeing or accuracy buff.

    - "Is the observed formula at least valid for most of the cases and what would be the common outliers?"
    That's for the devs to answer. However, my guess and what other people are saying is that certain special monsters have their stats hard-coded by the designers. It's entirely possible that the CR:63 in Kor Kaza is just for show and her actual stats were created having the 400 AC tanks in mind. Not being able to tank a certain boss with 173 AC is much less serious than the inability to tank ANYTHING in legendary content because the to-hit formula is blown out of proportions.
    My main server is Khyber. Have toons in almost every server for favor purposes. The Faltouts

  19. #19
    Community Member A-O's Avatar
    Join Date
    Apr 2013
    Posts
    467

    Default

    Quote Originally Posted by Avocado View Post
    Also do you have ANY proof that other then this baseless post of picture at 200 ac that the to hit formula for the reason you claim it was changed.

    I think most people that have played e.g. sharn know that 200 AC is not 50% mitigation. Not even close. Maybe, it's 10%, maybe it's 15%. That's what "3 tries of 100+ hits" would find out. But it's not like I need to flip a coin 300 times to know that it's roughly 50% chance of landing heads/tails.

    I doubt the formula has changed. Just that sharn mobs have another +50 tohit over ravenloft. Which has another 50 tohit over e.g. codex quests. I.e. it's no longer based on CR of the mob. In other words, there's also an "expansion" modifier of flat +X amount, and "red named" modifier, and "doom reaper" modifier, and "raid boss" modifier. Just conjecture. But I think everyone who plays a lot feel the same. There's a significant difference between codex quests, RL quests, and sharn quests. Even though the CR is practically the same.
    Formerly known as Absolute-Omniscience, co-creator of the old DPS calc.

  20. #20
    Community Member Stravix's Avatar
    Join Date
    May 2013
    Posts
    368

    Default

    Quote Originally Posted by A-O View Post
    I think most people that have played e.g. sharn know that 200 AC is not 50% mitigation. Not even close. Maybe, it's 10%, maybe it's 15%. That's what "3 tries of 100+ hits" would find out. But it's not like I need to flip a coin 300 times to know that it's roughly 50% chance of landing heads/tails.

    I doubt the formula has changed. Just that sharn mobs have another +50 tohit over ravenloft. Which has another 50 tohit over e.g. codex quests. I.e. it's no longer based on CR of the mob. In other words, there's also an "expansion" modifier of flat +X amount, and "red named" modifier, and "doom reaper" modifier, and "raid boss" modifier. Just conjecture. But I think everyone who plays a lot feel the same. There's a significant difference between codex quests, RL quests, and sharn quests. Even though the CR is practically the same.
    Which, if true, needs to be determined empirically. This is a game of numbers, so numbers like this are VERY important, and if they are not what we expect, can have dire consequences. This just means that testing needs to happen.

Page 1 of 3 123 LastLast

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  

This form's session has expired. You need to reload the page.

Reload