|
Post by bazukar on Apr 27, 2016 22:33:23 GMT
If AA was to bypass crit imm it should be from an ability and probably at a reduced level, like half multiplier or 1.5 crits. Any AA worth their salt with such an ability and a 19-20 crit range will be rocking OC/DC and keen ready to slay the beast.
I still think AA needs a viable swift action. Whether its an ability they already have or something new, they need to be able to weave something in between auto attacking.
|
|
|
Post by tomaan on Apr 28, 2016 7:29:41 GMT
Monks can strip crit immunity in certain situations, so it's not entirely without precedent. Once per day, against a single enemy, on a successful save. That's a lot different from ignoring crit immunity entirely. Correct, hence the part about the "certain situations".... ...but that's not important right now. What is important is that all agree it's not a viable solution. Inflicting vulnerability (damage/spell/otherwise) would seem to be the best option, but it's not a particularly unique buff so it would have to be significant to make an AA stand out. As boring as it sounds, buffing damage may be the way to go. AA already has some cool abilities - insta-kill, aoe, elemental focus, auto-hit - instead of trying to reinvent the wheel, why not give those abilities more punch?
|
|
|
Post by Paradoom on Apr 29, 2016 4:57:16 GMT
Since this needs some more hard data here are 2 full runs. Bazukar played his Arcane Archer which is purely build for damage with paragon weapon spec and I my Slinger. Both toons have OC/DC/IC/PC with a crit range from 18-20. AA has a 4x crit mod and the slinger a 3x. AA was 1x demi and Slinger 7x, but the AA has such high AB it didn´t matter. Listen was better on the Slinger, but only when buffed with babel fishes. EDIT: Both have 8 attacks per round and AA has the ego. Toon names: Leyn Teh'Pohk .:Crack Slinger:. Malad AA vs SlingerNessus AA vs SlingerOn a side note: Semner was on his Staffy in Nessus. Why that class got nerf bashed this extreme on damage output I really don't understand.
|
|
|
Post by MurphysLawAgain on Apr 29, 2016 7:27:37 GMT
Thanks for posting the logger data Paradoom. I read it as showing that there is very little difference in damage output between the AA and slinger. One of them was presumably adding party healing, kd & breach bullets, miracles, mass rez, weapon buffs and other benefits, the other possible interposing hands. Its pretty clear that AA's are suffering even in relation to slingers which aren't overpowered by anyones standards.
Just out of my interest, what classes are Malleausa and Rumpshakir?
|
|
|
Post by Paradoom on Apr 29, 2016 7:33:31 GMT
Malleausa was sorc and rumpshakir pariah. Interposing hand was usually done by me since it is a skill with the bullets. One trouble I have with it is like also the breach, is that it respects SR. That makes it impossible to put on many superior and elite mobs (e.g. Narzugons and Aspirants).
EDIT: Something that just came to mind. OC/DC was achieved via items on both characters. While an AA is capable to get that even in the build, this is impossible for a slinger. Without such an item the damage output for the slinger would drop even further and a level 60 one would suffer also from the low ab. Still the overall damage of an AA is too low as these logs show painfully. And, as murphy pointed out, the utility the slinger has is worlds beyond in comparison.
EDIT: mistakingly said rumpshakir was pariah. mixed that up with the actual one. he was as ssj pointed out a barbarian. my bad.
|
|
|
Post by Raj on Apr 29, 2016 8:34:18 GMT
A brief but necessary premise before going on ranting: AAs/Staffy/Slingers dmg in the current module suck, and the formers don't even have the excuse of support duty. Having incredible AB means nothing when you hit on 2+ on last attack even on ''uni'' tanks, having incredible single-element damage means even less when monsters outside desert haven't a single vulnerability but take good damage from multiple sources. AAs and Staffmasters have never been nerfed, they simply stopped receive freebies or changes got rolled back; other classes moved on, mostly thanks to free dmg buffs in the spirit of variety that altered the balance dramatically (I'm looking at you, integrated weapon buffs and loreboosted dmg). I don't think this is a (quasi)class fault, more a monsters design issue. Now onward to logs rant. Those logs, like all end of the run total logs, mean next to nothing. At most they're telling me some people were doing a terrible job, knowing their class and what such classes are able to do when good played, but the raw comparison between damage dealers is unfair. For those who weren't there it says nothing about efficient targeting and damage and is very dependant on player skill, reactivity, gear and survivability. Being the first one to attack or the one who needs the less time to reposition or rebuff means a lot, not to mention you could simply target some hamatula that is going to be imploded the next round instead of a pit fiend, or just never switch target/weapon (ye the latter means little for AAs, more for others). When you have weak casters (or when casters focus more on support/disable due to little confidence or fragile partymembers) then the numbers are obviously inflated: different Nessus parties might face the same amount of enemies and do about half total damage, because stuff gets instaed before aoe damage or bad tanks targeting (lemures! crits!) The next time you want to post some anecdotal evidence, better check specific number of attacks, hit rate and damage against a particular enemy or two (being Nessus, let say pit fiends/snakes/ichors and their paragon variations) that you are expected to deal swiftly as a dps toon. Even that won't say much to people who weren't there, because a guy might go afk, or die, or be busy pt2/rez people while his tunnel-visioned friend keeps attacking (the cursed, drained and stoned lemure), but would appease number crunching nerds a bit more. The best comparisons you can make are theoretical ones, because in practice there're too many variable and most of them are player-dependant. On the same side note: hey at least the useless staffy got a 99% dodge rate on 2259 attacks received, even if that says nothing about the importance of that tanking duty; might have been on some crucial chokepoint and prevented a wipe, or total unnecessary when everybody in the party is godlike geared: I'm going to assume it was the latter, and not having party support abilities makes a good tank nothing more than a glorified bbod if there're no taunt mechanics. Nonetheless, in a full man Nessus party being a high-survivability, low-damage toon is not going to contribute as much as some glass cannon. Different parties, different situations, different efficiency. EDIT: Something that just came to mind. OC/DC was achieved via items on both characters. While an AA is capable to get that even in the build, this is impossible for a slinger. Without such an item the damage output for the slinger would drop even further and a level 60 one would suffer also from the low ab. Still the overall damage of an AA is too low as these logs show painfully. And, as murphy pointed out, the utility the slinger has is worlds beyond in comparison. Let just say that talking about AA/Slinger damage performances in Nessus and add a OC/DC comment says a lot about bad targeting. Your hit rate also basically reflects the enemy listen-dropped concealment, quite high for pf/maleb/asmo, when I'm sure AB is of no concern. Edit: I wanted to doublecheck my strong assumptions, with a bit of old logs digging and a quick arena testing. Turned out most hell creatures (big boyz included) have AC of about 80 (EIGHTY!) before curse song, the highest ac I can find anywhere is Graz'zt at 100-105 probably. Still, with paragon levels and the availabe buffs those are hilarious ''defensive stats'' for monsters who rely on concealment only pretty much. Even against level 60 builds. Since a screen might be worth more than a lot of bs: I had to unequip my weapon to get low enough ab on the last attack and still didn't matter AAs big pro is high ab? really?
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Apr 29, 2016 9:09:30 GMT
The best comparisons you can make are theoretical ones, because in practice there're too many variable and most of them are player-dependant. I agree very strongly with this. The vastly more reliable method is to simply post typical attack progression + average damage per hit with respective 'ideal weapons' (can do AB also, but assuming AB is always high enough will be a close enough estimate IMO), then use those to directly calculate average damage per round against the classic tank targets and compare it to other classes; Pit Fiend, Malebranche, Marilith and Balor would be a decent start. I believe the above numbers would be what Funky and co would prefer to see before considering a 'buff of X% to damage' or anything similar. More on topic, I do tend to agree that AA's need a buff - a swift action Ranged Disarm could indeed be nice, being able to take advantage of AA's relatively high AB, and it seems somewhat thematic too (super accurate archer shoots their target's weapon/hand directly). As stated above, I would prefer to see accurate damage numbers before commenting on a direct boost to damage. Edit: Are you sure Rumpshakir is a Pariah? He has a 66% hit-rate from 3000+ attacks, and a 94% dodge-rate.
|
|
|
Post by Raj on Apr 29, 2016 9:21:43 GMT
AB is a no-factor, maxed listen is used any serious endgame build. It should reach 120+ with ease, then pens would affect most toons in a similar way, but overall differences between fully buffed LSA: listen/class skill users are minimal. Prell-tags and abyss reward bonus damage, prayer, bardsong, various buffs are the same for everybody. We can also safely assume high enough gmw to pierce mobs dr as well (+15 gmw/+16 iof are usually availabe, AA bows are +16, most enemies in hells and abyss have +14 dr, some +15). Damage immunities datas for pretty much everything are availabe to anybody who downloads HGX and checks into the ...\hgx2.8.5\data\characters.d directory. If you don't post anything in the next hours I might do some comparisons between standard builds, but I'd have to find those first
|
|
|
Post by Paradoom on Apr 29, 2016 9:51:26 GMT
Well Rajah since you are so against slingers and aa toons anyway I am not very surprised about that biased outburst on your end. Player-skill, targets attacked, quicker reaction? If we had such noob problems I wouldn´t go into the trouble of making runs like this. Why do you think I posted the WHOLE log files? If you want to have detailed information against specific mobs you can get it out of there. E.g. Slinger did quite more damage on Pit fiends while AA got more on the Malebranches. Clever targeting and micromanagement is something that Baz and I are really up too and the amount of attacks is as well as the hit rating overall pretty much the same level. A very solid basis to compare and see how they do.
Salek is working on a detailed report for those logs in the meantime, but it will take a bit.
|
|
|
Post by Raj on Apr 29, 2016 9:55:07 GMT
You didn't get it.
|
|
|
Post by Paradoom on Apr 29, 2016 9:58:12 GMT
You neither.
|
|
|
Post by Raj on Apr 29, 2016 10:11:10 GMT
No look, I was going to calculate precisely theoretical best dmg output and actually am arguing for your pet toons to receive a damage buff, but as it stands I mostly complained about your useless logs attitude. And I'm more than happy to skip some unrewarded work and let you ramble with numbers you have no clue about.
Even better, hand checking the full log shows that you focus on A LOT of fodder monsters while Semner, Baz or Bad are focusing more problematic monsters so, while waiting for more detailed infos that are such a pita to extract now that run ended, I'll just smile when randomly swimming through logs while you act so defensive.
Elite pf! let sneak on the Barbazu instead...
A Ichor! Whole party attacks it! Ah wait, Doom is on the erinnie.
Note: my rant was pointed at how logs can be manipulated and in general aren't to be trust to define the viability of a particular class, this could be a fine example as well. If you want to argue more about classes who deserve a buff feel free to provide some less casual numbers, because, guess what, I lost interest in supporting your cause.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Apr 29, 2016 10:12:06 GMT
It has nothing to do with a particular individual being a 'noob' or not, the fact of the matter is that total damage from loggers is a weak metric to use. While it is *a* metric, the huge variance based on party size/composition/skill/play-style as well as rewarding sub-optimal play (ie: hitting trash) makes it inferior next to a direct theoretical comparison.
Raj was essentially saying that if you insist on using logs (despite all the valid reasons listed why they are poor for evaluating effectiveness), at least do '#stat.vs Pit Fiend' and other key mobs to get data for the *entire run* against those specific tank targets, which would moderately improve reliability of the data. This is not information that comes readily from a full log (fun fact, the log DID at least show that Rumpshakir is a Barbarian, not a Pariah).
Think of it like submitting a science paper - you want to provide information with as little variance as possible, and more importantly, be able to convince the reader that the variance has indeed been minimised. Anecdotal evidence of 'in our opinion, we were playing ~optimally' is not particularly helpful in this regard, even if it was true.
|
|
|
Post by Paradoom on Apr 29, 2016 10:36:44 GMT
Alright I think we can all calm down now. First off: Myrphy was asking us to make logs and post them. We tried to make a fair ground level by taking two top tier builds which cheese in what ever they can to max out their performance/damage and baz and I did in the respective class (AA and Slinger) which were the discussion. Secondly: Detailed extrapolations of these will follow but need as said time yet. Thirdly: What do you expect Raja. The first thing you do is basically jump on the provided data and make it look like it is basically a big bunch of crap and now talking of support... Did your really expect another reaction? Also you claim that theoretical numbers would provide a better basis for discusion vs actually data from runs. I think if even then only both together would.
|
|
|
Post by bazukar on Apr 29, 2016 11:03:07 GMT
Can we stop with the passive aggressive attitudes? It's not bringing anything to the discussion. Make fun of target selection if you want, but several times that run I myself had to go rail on something that wasn't a PF because it was beating on me and the rest of us were otherwise occupied. There are plenty of factors regarding what you are smacking at any given time. Sheesh. This is some basic numbers to help illustrate the issue. I'm all for more data and dissemination. Let's focus on that rather than sniping.
|
|