Theck's MATLAB thread - Cataclysm/4.x

Warning: Theorycraft inside.

Moderators: Fridmarr, Worldie, Aergis, theckhd

Re: Theck's MATLAB thread - Cataclysm/4.x

Postby Klaudandus » Wed Nov 17, 2010 3:07 pm

AriKT wrote:From just messing around from a bit on the 4.0.3a PTR:
Crusader Strike is 115% Weapon Damage as a base.

Wrath of the Light Bringer is now 50%/100% increased Judgement and Crusader Strike damage.

This brings fully talented Crusader Strike down to 264.5% Weapon Damage.
From some quick napkin math, this puts HotR > CS.

Oh and the Boss dummy is level 88 now I believe. It shows as skull level to me. No logs for this yet, I'll see about getting some later.


Not this again. Damn you blizz...
The Element of Forum Hyperbole
Image
---
Flüttershy - Draenei Protection Paladin, Aerie Peak
Klaudandus - BE Protection Paladin, Feathermoon (Semi-retired)
User avatar
Klaudandus
 
Posts: 9359
Joined: Thu Apr 02, 2009 7:08 am
Location: Texas' Armpit

Re: Theck's MATLAB thread - Cataclysm/4.x

Postby Gamingdevil » Wed Nov 17, 2010 3:30 pm

AriKT wrote:Oh and the Boss dummy is level 88 now I believe. It shows as skull level to me. No logs for this yet, I'll see about getting some later.


A boss level mob, always counts as being 3 levels above you, regardless of it's intended level (hence you still need 8% hit for Ragnaros for instance) I assume dummies are no different.
If it is weak, kill it before it gets stronger. If it is strong, weaken it.
Image
User avatar
Gamingdevil
 
Posts: 1267
Joined: Mon Jan 14, 2008 8:16 am
Location: Belgium

Re: Theck's MATLAB thread - Cataclysm/4.x

Postby Klaudandus » Wed Nov 17, 2010 3:40 pm

Gamingdevil wrote:
AriKT wrote:Oh and the Boss dummy is level 88 now I believe. It shows as skull level to me. No logs for this yet, I'll see about getting some later.


A boss level mob, always counts as being 3 levels above you, regardless of it's intended level (hence you still need 8% hit for Ragnaros for instance) I assume dummies are no different.


Not anymore in Cata, where GC hinted at going with lvl 88 that scale per tier, kind of like 88, 88-1, 88-2, so that you will always want more stats without reaching an actual cap for whatever is the latest content. So, while the first bosses of the first raids might require a certain amount of hit, the next tier would need just a smidge more of hit.
The Element of Forum Hyperbole
Image
---
Flüttershy - Draenei Protection Paladin, Aerie Peak
Klaudandus - BE Protection Paladin, Feathermoon (Semi-retired)
User avatar
Klaudandus
 
Posts: 9359
Joined: Thu Apr 02, 2009 7:08 am
Location: Texas' Armpit

Re: Theck's MATLAB thread - Cataclysm/4.x

Postby Gamingdevil » Thu Nov 18, 2010 5:47 am

That doesn't necessarily mean that the boss' level will change, just that their stats will. They might just increase their dodge, parry and expertise for instance.
If it is weak, kill it before it gets stronger. If it is strong, weaken it.
Image
User avatar
Gamingdevil
 
Posts: 1267
Joined: Mon Jan 14, 2008 8:16 am
Location: Belgium

Re: Theck's MATLAB thread - Cataclysm/4.x

Postby culhag » Thu Nov 18, 2010 6:53 am

Gamingdevil wrote:A boss level mob, always counts as being 3 levels above you, regardless of it's intended level (hence you still need 8% hit for Ragnaros for instance) I assume dummies are no different.

Boss level mobs scale up but they don't scale down.
For example if you go fight Ragnaros now it'll be as if he's lvl 83, but if you'd try to fight him at level 50 he won't be lvl 53. He's coded as a level 60 boss so he'll always be at least level 63.
Likewise if the dummy is coded as a lvl 85 boss it will always be at least lvl 88.
User avatar
culhag
Maintankadonor
 
Posts: 1246
Joined: Wed Aug 06, 2008 7:50 am
Location: France

Re: Theck's MATLAB thread - Cataclysm/4.x

Postby Gamingdevil » Thu Nov 18, 2010 7:02 am

culhag wrote:
Gamingdevil wrote:A boss level mob, always counts as being 3 levels above you, regardless of it's intended level (hence you still need 8% hit for Ragnaros for instance) I assume dummies are no different.

Boss level mobs scale up but they don't scale down.
For example if you go fight Ragnaros now it'll be as if he's lvl 83, but if you'd try to fight him at level 50 he won't be lvl 53. He's coded as a level 60 boss so he'll always be at least level 63.
Likewise if the dummy is coded as a lvl 85 boss it will always be at least lvl 88.


Ah yes, forgot about that, good point.
If it is weak, kill it before it gets stronger. If it is strong, weaken it.
Image
User avatar
Gamingdevil
 
Posts: 1267
Joined: Mon Jan 14, 2008 8:16 am
Location: Belgium

Re: Theck's MATLAB thread - Cataclysm/4.x

Postby AriKT » Thu Nov 18, 2010 10:43 am

Sorry, didn't get logs last night and probably won't before the weekend. Thu, Fri, and Sat are raid days for me.

Another thing was, that SoT and Censure were doing far less damage on the PTR than on Live. This will probably make SotP less attractive.
AriKT
 
Posts: 42
Joined: Thu Jul 15, 2010 12:24 pm

Re: Theck's MATLAB thread - Cataclysm/4.x

Postby kristoferpally » Sun Nov 21, 2010 2:51 am

1) so to clarify is hotr> cs for single target as of now and will this change post 4.03a?

2) they are nerfing gbtl from 60% to 10% increase to wog :( Pally tanks already take the most damage of the tanks (maybe less than warriors) and on top of this they are nerfing our heal that wasnt even in need of a nerf since to use it we have to give up shield slam which is a huge dps loss. /sigh
kristoferpally
 
Posts: 13
Joined: Thu Oct 14, 2010 3:34 pm

Re: Theck's MATLAB thread - Cataclysm/4.x

Postby theckhd » Sun Nov 21, 2010 11:42 am

kristoferpally wrote:1) so to clarify is hotr> cs for single target as of now and will this change post 4.03a?

No, and probably not. CS is better right now, and despite the slight (~10%) nerf in 4.0.3a it will probably remain better due to the glyph, if only by a slim margin. At 85 I suspect that CS will win hands down, as it scales considerably better with AP (~45% compared to ~33% for HotR).

kristoferpally wrote:2) they are nerfing gbtl from 60% to 10% increase to wog :( Pally tanks already take the most damage of the tanks (maybe less than warriors) and on top of this they are nerfing our heal that wasnt even in need of a nerf since to use it we have to give up shield slam which is a huge dps loss. /sigh

Is there a question here? Because there's no cheese in this thread to go with whine.
"Theck, Bringer of Numbers and Pounding Headaches," courtesy of Grehn|Skipjack.
Simcraft 6.x, Call to Arms 6.0, Talent Spec & Glyph Guide 6.x, Blog: Sacred Duty
User avatar
theckhd
Moderator
 
Posts: 6211
Joined: Thu Jul 31, 2008 3:06 pm
Location: Harrisburg, PA

Re: Theck's MATLAB thread - Cataclysm/4.x

Postby tobit » Sun Nov 21, 2010 1:14 pm

theckhd wrote:CS is better right now, and despite the slight (~10%) nerf in 4.0.3a it will probably remain better due to the glyph, if only by a slim margin.


Is this the case only if you're using a 2.6 speed weapon, or does it apply to the (admittedly, outdated) faster tanking weapons? I've been trying to get a BVB or at least a Last Word while farming honor for a Wrathful, but as of right now I'm still using a Bonebreaker Scepter. -_-

Also, does 4.0.3a fix the glyph and set bonus bugs for HotR?
tobit
 
Posts: 4
Joined: Thu Oct 15, 2009 4:46 pm

Re: Theck's MATLAB thread - Cataclysm/4.x

Postby Gamingdevil » Sun Nov 21, 2010 1:43 pm

tobit wrote:
theckhd wrote:CS is better right now, and despite the slight (~10%) nerf in 4.0.3a it will probably remain better due to the glyph, if only by a slim margin.


Is this the case only if you're using a 2.6 speed weapon, or does it apply to the (admittedly, outdated) faster tanking weapons? I've been trying to get a BVB or at least a Last Word while farming honor for a Wrathful, but as of right now I'm still using a Bonebreaker Scepter. -_-

Also, does 4.0.3a fix the glyph and set bonus bugs for HotR?


Iirc, they fixed the glyph, not the set bonus. I'm using a Mithrios and my CS hits harder than HotR, so it's better with fast weapons as well, yes.
If it is weak, kill it before it gets stronger. If it is strong, weaken it.
Image
User avatar
Gamingdevil
 
Posts: 1267
Joined: Mon Jan 14, 2008 8:16 am
Location: Belgium

Re: Theck's MATLAB thread - Cataclysm/4.x

Postby theckhd » Sun Nov 21, 2010 3:30 pm

tobit wrote:Is this the case only if you're using a 2.6 speed weapon, or does it apply to the (admittedly, outdated) faster tanking weapons? I've been trying to get a BVB or at least a Last Word while farming honor for a Wrathful, but as of right now I'm still using a Bonebreaker Scepter. -_-

It should be true in either case. The gap might be narrower if you're using a slow weapon, but for tanking weapons CS should easily pull ahead. Also note that this is assuming a high stack of Vengeance.

Gamingdevil wrote:
tobit wrote:Also, does 4.0.3a fix the glyph and set bonus bugs for HotR?


Iirc, they fixed the glyph, not the set bonus.

The PTR data I've seen indicates they've fixed neither.
"Theck, Bringer of Numbers and Pounding Headaches," courtesy of Grehn|Skipjack.
Simcraft 6.x, Call to Arms 6.0, Talent Spec & Glyph Guide 6.x, Blog: Sacred Duty
User avatar
theckhd
Moderator
 
Posts: 6211
Joined: Thu Jul 31, 2008 3:06 pm
Location: Harrisburg, PA

Re: Theck's MATLAB thread - Cataclysm/4.x

Postby kristoferpally » Mon Nov 22, 2010 4:54 am

kristoferpally wrote:2) they are nerfing gbtl from 60% to 10% increase to wog :( Pally tanks already take the most damage of the tanks (maybe less than warriors) and on top of this they are nerfing our heal that wasnt even in need of a nerf since to use it we have to give up shield slam which is a huge dps loss. /sigh

Is there a question here? Because there's no cheese in this thread to go with whine.[/quote]



I was feeling crabby and needed to vent. There any butter here?
kristoferpally
 
Posts: 13
Joined: Thu Oct 14, 2010 3:34 pm

Re: Theck's MATLAB thread - Cataclysm/4.x

Postby allikat » Tue Nov 23, 2010 10:05 pm

Theck, since you've got the Shattering right now, and us Euros get our chance in 5 more hours (or so, we all know Blizz and patch day..) is there any real differences we should be aware of from a tanking paladin's perspective? Are the rotations and specs from 4.01-4.03 still 100% valid with the rebalancing?
Just planning ahead to our talent reset.
Vanilla paladins, never forget, never forgive.
Quel'serrar serrar... whatever I am, I tank :)
Image
allikat
 
Posts: 4
Joined: Wed Nov 17, 2010 9:42 pm

Re: Theck's MATLAB thread - Cataclysm/4.x

Postby theckhd » Tue Nov 23, 2010 10:26 pm

I suspect that nothing's changed appreciably enough to make the rotation different, no. Seals of the Pure is probably a little weaker than it was if the seal nerf went through, and the ShoR change should devalue AP slightly, but otherwise things are probably identical.
"Theck, Bringer of Numbers and Pounding Headaches," courtesy of Grehn|Skipjack.
Simcraft 6.x, Call to Arms 6.0, Talent Spec & Glyph Guide 6.x, Blog: Sacred Duty
User avatar
theckhd
Moderator
 
Posts: 6211
Joined: Thu Jul 31, 2008 3:06 pm
Location: Harrisburg, PA

Re: Theck's MATLAB thread - Cataclysm/4.x

Postby Treck » Wed Nov 24, 2010 2:15 pm

I seemed to consitantly get higher Hammer hits than Crusader strikes throughout the whole raid on single target, anyone else noticed this?
Ret are also doing way less dmg with crusader strike than they used to, dunno if related (all their dmg seemed less in fact)
User avatar
Treck
 
Posts: 1431
Joined: Thu Aug 21, 2008 8:10 am

Re: Theck's MATLAB thread - Cataclysm/4.x

Postby theckhd » Wed Nov 24, 2010 3:17 pm

More recent testing indicates that the Glyph of HotR bug has been fixed. It's possible that this means they've also fixed the Tier 10 bonus. This, along with the CS nerf, probably pushes HotR back ahead for single targets, which would explain your observations.

I might try and do a few quick edits to the local copy of the 4.0.1 code I have just to see if this is in fact the case.
"Theck, Bringer of Numbers and Pounding Headaches," courtesy of Grehn|Skipjack.
Simcraft 6.x, Call to Arms 6.0, Talent Spec & Glyph Guide 6.x, Blog: Sacred Duty
User avatar
theckhd
Moderator
 
Posts: 6211
Joined: Thu Jul 31, 2008 3:06 pm
Location: Harrisburg, PA

Re: Theck's MATLAB thread - Cataclysm/4.x

Postby knaughty » Wed Nov 24, 2010 3:48 pm

theckhd wrote:This, along with the CS nerf, probably pushes HotR back ahead for single targets.


Does anyone have Ghostcrawler's shipping address? I'd like to send him a copy of Excel and the "For Dummies" guide to go with it... :evil:

I cannot fathom how they keep fucking this up. If I was being paid to balance the damn game I would not be making shit up on the fly - I'd have the whole shebang modelled in a few Excel workbooks and I would know what DPS/TPS/HPS people would be able to do before I tweaked shit. If I had constraints like "CS should be > HotR for 1 target" then they'd be coded in as conditional formatting and if I fucked shit up cells on my summary sheet would go red.

The fact that external theorycrafters can work out balance problems before the dev team without access to the source code or internal models is verging on gross incompetence.
This isn't the "Offtankadin" forum. My MoP FAQ: http://tinyurl.com/FAQ-5-0
- Knaughty.
User avatar
knaughty
Maintankadonor
 
Posts: 1846
Joined: Mon Dec 17, 2007 10:06 pm
Location: Sydney, plotting my next diatribe against the forces of ignorance!

Re: Theck's MATLAB thread - Cataclysm/4.x

Postby Dantriges » Wed Nov 24, 2010 4:05 pm

9 to 5 job, design by committee, designers considering themselves artists with "artistic vision" and experts, basic miscommunication, egos who think they know better colliding with the player base, corporate structure messing around with work in progress, time constraints leading to quick fix solutions. There are probably dozens of reasons RPGs in general are screwed in the math department.

I am really wondering if someone is really reading threads like this. Whoever visits the board and is a member of Blizzard staff, probably stumbles over this and similar threads in 5 minute after entering the forum. The information is pretty much based on facts. Still I haven´t seen many changes based on the facts neatly provided by the community. This isn´t the usual WoW official forum whine thread, every fact here is based on simulations. Seems that Mr forum reader is part of another department or in a low position on the corporate totem pole. You know the intern who is sent for coffee and copy duties.

I´ve read fans arguig that barely understandable rules are a statement about the iimportance of trancendence in the setting or that the rules are modeled after set theory and that you don´t get them if you don´t have a PH.D in computer sciences, philosophy or you should read more about eastern or greek philosophy, the Gilgamesh epos or some other eastern epos from India or China (depending on what kind of game). I think my remark was something like "oh nice, this rules are barely useful for actually playing the game, I didn´t buy it for an statement about transcendence in the setting, someone thought about that?"

Get rid of the idea that you have actual programmers in the design department. It explains a lot. Designing grand visions are exciting, including cameos and jokes in the game are fun, math is boring, at least for artists and you have a bunch of wannabe artists as designers. Worse you have a bunch of wannabe artists embedded in corporate structures. The only saving grace is that WoW is more mainstream so you can´t include the stuff you need a PH.D in philosophy or too much free time to ever have more than passing knowledge.
Dantriges
 
Posts: 657
Joined: Wed Jan 30, 2008 12:39 am

Re: Theck's MATLAB thread - Cataclysm/4.x

Postby Treck » Wed Nov 24, 2010 4:15 pm

theckhd wrote:More recent testing[/url] indicates that the Glyph of HotR bug has been fixed. It's possible that this means they've also fixed the Tier 10 bonus. This, along with the CS nerf, probably pushes HotR back ahead for single targets, which would explain your observations.

My observations were without the glyph aswell, so should be pushing ahead even more with that aswell.
The differance in damage is so small, that it feels stupid even "choosing" cuz it really doesnt matter what you take, both are going to work.
During the beta testing CS seemed to come way ahead tho on single target, so this might just be unp untill cata.
User avatar
Treck
 
Posts: 1431
Joined: Thu Aug 21, 2008 8:10 am

Re: Theck's MATLAB thread - Cataclysm/4.x

Postby knaughty » Wed Nov 24, 2010 5:06 pm

Designers should be setting targets and goals. Someone who did maths, science or programming at University should be plugging the damn numbers into the spell database.

Targets and goals are things like:
  • Tank EH in equivalent gear should be within 3% of each other, ideally closer.
  • Overall mitigation should be within about 10% or so for BossModel A, B & C and within 25% for D, E & F. Where total damage received is noticeably higher it should be less spiky.
  • Randomly whacking things that are off CD should be at least 20% worse than the optimal rotation.
  • The tankadin single target and AE rotations should be different.
  • For each ranged DPS class, the best PvE spec should be within 10% of TargetValueForThisTier, ideally 5%.
  • Tanks should do about 60% of the damage of a melee DPS.
  • We don’t want Inquisition to be effective for single target use.
  • Grand Crusader should be a useful proc in virtually all circumstances – it lights up, you hit it, DPS increases.
  • It should be possible for mana to run out for prot and ret if the player is stupid. It should be impossible if they’re smart.
  • Prot should be mostly unviable in Arena, but marginally useful in rated BGs

Stuff like that is what designers should be doing - I'm not claiming my suggestions above are correct, by the way - but that's the sort of goals I'd be expecting to receive if I was the dude coming up with spell coefficients. Coming up with spell descriptions and mechanics should also a design thing “We’ll do a Captain America shield throw thingy for tankadins! It’ll be awesome!

Once you’re at the “plug numbers into the DB” you state your goals and constraints to the nerd with a maths degree and he determines the damn values. Basic failures like “OK, nerf warlocks by 12%” or “HotR beats CS single-target” or the utterly craptastic tuning of the Vengeance mechanic should not be possible if your design process was correctly built.

I’m an IT Security Risk Advisor – one of my key roles is “control design” – IE: this is how you build a process so it’s idiot proof and reliably meets its design goal. The design and balance process that Blizzard use is fundamentally flawed – they basically just make educated guesses as to what might work and pick a few numbers, then randomly tweak shit when it’s wrong in an attempt to get “close enough”. Then the next tier of content happens, every scales wrong and they randomly tweak shit hoping to get it right again.

I find this sort of shit annoying and have a pathological desire to fix inefficient processes. Thus: The day job. If I was running the show I'd have the entire goddamn Cataclysm expansion tuning already done through to Tier-14. You'd be able to ask me how much damage a warlock will be doing at Deathwing-hard-mode, and I'd be able to say "this much".
This isn't the "Offtankadin" forum. My MoP FAQ: http://tinyurl.com/FAQ-5-0
- Knaughty.
User avatar
knaughty
Maintankadonor
 
Posts: 1846
Joined: Mon Dec 17, 2007 10:06 pm
Location: Sydney, plotting my next diatribe against the forces of ignorance!

Re: Theck's MATLAB thread - Cataclysm/4.x

Postby Klaudandus » Wed Nov 24, 2010 5:08 pm

Only problem I see, is that the moment they finally realize HotR is ahead of CS, rather than buff up CS, they're gonna nerf HotR.
The Element of Forum Hyperbole
Image
---
Flüttershy - Draenei Protection Paladin, Aerie Peak
Klaudandus - BE Protection Paladin, Feathermoon (Semi-retired)
User avatar
Klaudandus
 
Posts: 9359
Joined: Thu Apr 02, 2009 7:08 am
Location: Texas' Armpit

Re: Theck's MATLAB thread - Cataclysm/4.x

Postby theckhd » Wed Nov 24, 2010 6:44 pm

Treck wrote:During the beta testing CS seemed to come way ahead tho on single target, so this might just be unp untill cata.

I suspect that the scaling of weapon damage will make this true at 85, which is what this patch is balanced around. But since we're still level 80, we're in a weird temporary state that the abilities weren't balanced around.

knaughty wrote:Once you’re at the “plug numbers into the DB” you state your goals and constraints to the nerd with a maths degree and he determines the damn values. Basic failures like “OK, nerf warlocks by 12%” or “HotR beats CS single-target” or the utterly craptastic tuning of the Vengeance mechanic should not be possible if your design process was correctly built.


I think a job description of "do all sorts of neat math modeling about WoW" sounds like a dream job. Especially if I didn't have to figure out all of the spell coefficients empirically. On the other hand, since we already do all of this without getting paid, there's not much incentive for Blizzard to pay someone to do it. I bet the kinks would get worked out a lot faster that way though. Just imagine if they had a Theck on staff for every class to give them immediate feedback on numbers tuning?

(Was that too subtle?)
"Theck, Bringer of Numbers and Pounding Headaches," courtesy of Grehn|Skipjack.
Simcraft 6.x, Call to Arms 6.0, Talent Spec & Glyph Guide 6.x, Blog: Sacred Duty
User avatar
theckhd
Moderator
 
Posts: 6211
Joined: Thu Jul 31, 2008 3:06 pm
Location: Harrisburg, PA

Re: Theck's MATLAB thread - Cataclysm/4.x

Postby Dantriges » Wed Nov 24, 2010 6:55 pm

Do you want to send in an application? :wink:

The sad thing is, they probably don´t listen so much to these kind of guys. I mean there must be someone on the staff who is good with this kind of things. How big is the staff involved with the design and development?

Ah well probably no one really wants to argue thiskind of things if it is his job. How often do people keeptheir mouth shut if they think their boss just had some bad ideas.
Especially considering that most people won´t have a large knowledge about every class to begin with. Got a lot of twinks and only know perhaps six to seven specs with a level of competence that I believe would be sufficient to actuall say something about it. And it´s a hobby for me. Unless your job is WoW design, it´s not in the job description to know much about how the game is actualy played.
Dantriges
 
Posts: 657
Joined: Wed Jan 30, 2008 12:39 am

Re: Theck's MATLAB thread - Cataclysm/4.x

Postby theckhd » Wed Nov 24, 2010 7:08 pm

Dantriges wrote:Do you want to send in an application? :wink:


Shit, I'd have to basically completely re-write my C.V. for that. Or an addendum. "The first three pages cover my day job, including work history, scientific publications, etc. And the next four pages after that cover my night job, which is WoW theorycrafting."
"Theck, Bringer of Numbers and Pounding Headaches," courtesy of Grehn|Skipjack.
Simcraft 6.x, Call to Arms 6.0, Talent Spec & Glyph Guide 6.x, Blog: Sacred Duty
User avatar
theckhd
Moderator
 
Posts: 6211
Joined: Thu Jul 31, 2008 3:06 pm
Location: Harrisburg, PA

PreviousNext

Return to Advanced Theorycraft and Calculations

Who is online

Users browsing this forum: Majestic-12 [Bot] and 1 guest

Who is online

In total there are 2 users online :: 1 registered, 0 hidden and 1 guest (based on users active over the past 5 minutes)
Most users ever online was 380 on Tue Oct 14, 2008 6:28 pm

Users browsing this forum: Majestic-12 [Bot] and 1 guest