Theck's MATLAB TPS analysis - WotLK/3.x

Warning: Theorycraft inside.

Moderators: Fridmarr, Worldie, Aergis, theckhd

Re: Theck's MATLAB TPS analysis (A Jonesy derivative work)

Postby raenac » Wed Aug 05, 2009 11:16 am

theckhd wrote:
Raenac wrote:Hi Theck,

I'm not sure if I missed where it is, but was there a comparision of enchant/food for best mitigation since our threat pretty much rules already.

Nope, for exactly the reason Jasari pointed out. It's pretty much AGI or STR (for BV). Pick your poison, but AGI is almost certainly going to be the winner there.


Ok thanks!
raenac
 
Posts: 15
Joined: Thu Nov 20, 2008 6:57 am

3.2 Weapon Speed Analysis

Postby theckhd » Wed Aug 05, 2009 11:43 am

On the heels of the last analysis, let's take a look at weapon speed.

3.2 Weapon Speed Analysis

We know from previous work on SoB tanking that procs tend to favor slow weapons. Since SoV now has a proc, we might wonder how significant this effect is. As a first stab at the question, let's compare a few of the tanking weapons from 3.1 and earlier.

First, let's look at the question in the last post that I suggested nobody cared about: Will a slow weapon help SoCom catch up to SoR?

Setup
Rotation: 969
Seals: SoV, SoC, SoR
Glyphs: V, J, or R according to seal
Talents: Various
Gear: Theck's 8.5 Progression Set, given above
Weapons: Titanguard (TG), Broken Promise (BP)
File: calc_weaps_speed.m

I chose to only represent a few talent specs, so that this plot didn't get too busy. I picked ones that would be both popular, and gave us an idea how useful Reckoning was in this sort of situation.

Image
There are a few things that stand out here. First of all, Broken Promise is generating more threat than Titanguard fairly consistently, regardless of spec. Since SoCom isn't affected by SotP, switching to a slow weapon like BP actually helps SoCom catch up for the heavier crusade specs.

Note also that Reckoning helps BP more than TG, as expected.

The final thing to note is that no matter how slow you get, neither of the other seals will catch SoV. Maybe if we get a 3.6 speed tanking weapon down the line...

Next, let's look at how the rest of the tanking weapons fit into this plot. We'll focus only on SoV based on the last results.
Setup
Rotation: 969
Seals: SoV
Glyphs: V
Talents: Various
Gear: Theck's 8.5 Progression Set, given above
Weapons: Titanguard (TG), Broken Promise (BP), Last Laugh (LL), Shiver (ilvl 232 version) (SH), Sorthalis, Hammer of the Watchers (ST), and Vulmir, the Northern Tempest (VM).
File: calc_weaps_speed.m

Image
As we saw with the SoB analysis, slow DPS weapons like Vulmir will generate massive amounts of threat. Going from a 1.6 weapon (SH) to a 2.6 weapon (VM) nets you 280 TPS, or 28 TPS per 0.1 weapon speed. Not really much of a surprise, but it's always nice to have numbers.

What's perhaps more interesting is that as before, Broken Promise is now generating more threat than Shiver or Titanguard no matter what spec we choose. In fact, for SotP specs with 3 points in Reckoning, it's matching Sorthalis, and for the 5S+5R spec it's actually better than Sorthalis for threat. It's still behind Sorthalis for Crusade specs, though, and has weaker survival stats.

If you still have your BP, you may want to swap between a Shiver/TG for survival (enchanted for Blood Draining or an avoidance enchant maybe) and a BP with Accuracy (Berserkinglol?) for threat.

TLDR summary:
  • Slow DPS weapons are good for threat. The conversion is roughly 280 TPS per 1 weapon speed, if weapon DPS is the same.
  • Broken Promise is still hanging in there, being the 2nd-highest TPS tanking weapon for most specs (Sorthalis being the top).
Last edited by theckhd on Mon Aug 10, 2009 9:57 am, edited 1 time in total.
"Theck, Bringer of Numbers and Pounding Headaches," courtesy of Grehn|Skipjack.
MATLAB 5.x, Simcraft 6.x, Call to Arms 6.0, Talent Spec & Glyph Guide 5.x, Blog: Sacred Duty
User avatar
theckhd
Moderator
 
Posts: 7950
Joined: Thu Jul 31, 2008 3:06 pm
Location: Harrisburg, PA

Re: Theck's MATLAB TPS analysis (A Jonesy derivative work)

Postby theckhd » Wed Aug 05, 2009 11:47 am

Amaranthea wrote:Armor pen showing up as a better dps stat than hit for us when the only thing affected by armor pen is autoattack doesn't raise any red flags?

Armor penetration rating has a stat value of 1.0 - the same as strength or hit rating, as a quick look at the PVP necks can confirm. The wowwiki page and the EJ thread both appear to be out of date.

Edit: Also red gems. 16 str or 16 armor pen or 16 expertise. Same value.

I'm sure you're correct about this one. In the flurry of trying to get everything updated, I haven't had as much time as usual to step back and ask "does this even make sense." Sorry for letting that one slip by. It confused me when I stumbled across it during PTR, which is why I investigated (and corrected) the boss armor and armor penetration calculations. I never thought to double check the stat mod though.

Y'know, one of us should really be going in and editing wowwiki as we discover these things. I guarantee that 6 months from now I won't remember this, and I'll be wondering why my statmod value for ArPen is wrong again...

I'll re-run the calc and update the plots and discussion appropriately. I don't think it's worth re-posting the whole thing again, since it's only been a page or two.
"Theck, Bringer of Numbers and Pounding Headaches," courtesy of Grehn|Skipjack.
MATLAB 5.x, Simcraft 6.x, Call to Arms 6.0, Talent Spec & Glyph Guide 5.x, Blog: Sacred Duty
User avatar
theckhd
Moderator
 
Posts: 7950
Joined: Thu Jul 31, 2008 3:06 pm
Location: Harrisburg, PA

Re: Theck's MATLAB TPS analysis (A Jonesy derivative work)

Postby Sputterduck » Wed Aug 05, 2009 5:30 pm

Good Lord Theck, I have no idea what I'd do without you and your contributors to all this number-crunching. I hearby award you with both ears and the tail of this numerical bullfight
"experience is what you get when you don't get what you want"
User avatar
Sputterduck
 
Posts: 206
Joined: Sun Jun 01, 2008 9:50 pm
Location: California

Re: Theck's MATLAB TPS analysis (A Jonesy derivative work)

Postby jere » Wed Aug 05, 2009 7:31 pm

theckhd wrote:Y'know, one of us should really be going in and editing wowwiki as we discover these things. I guarantee that 6 months from now I won't remember this, and I'll be wondering why my statmod value for ArPen is wrong again...

Comment your code!!!! :mrgreen:
User avatar
jere
 
Posts: 2963
Joined: Fri May 11, 2007 5:12 pm

Re: Theck's MATLAB TPS analysis (A Jonesy derivative work)

Postby theckhd » Wed Aug 05, 2009 7:52 pm

jere wrote:Comment your code!!!! :mrgreen:

I've been trying!

I figure I'll have to make a serious pass over it and comment it up really well in a few weeks when things die down. Maybe while I translate it to lua and php.
"Theck, Bringer of Numbers and Pounding Headaches," courtesy of Grehn|Skipjack.
MATLAB 5.x, Simcraft 6.x, Call to Arms 6.0, Talent Spec & Glyph Guide 5.x, Blog: Sacred Duty
User avatar
theckhd
Moderator
 
Posts: 7950
Joined: Thu Jul 31, 2008 3:06 pm
Location: Harrisburg, PA

Re: Theck's MATLAB TPS analysis (A Jonesy derivative work)

Postby Nadir » Wed Aug 05, 2009 11:21 pm

Theck, would you mind adding Hellscream Slicer/ Stormpike Cleaver (the new ilvl 245 slow weapons) into the mix? No it doesn't have defense, but avoidance wise it's within 0.01%-0.02% of Broken Promise when fully raid buffed. Its got a nice chunk of hit, to boot.

http://www.wowhead.com/?item=47314
Image
Nadir
 
Posts: 266
Joined: Mon May 04, 2009 8:37 am

Re: Theck's MATLAB TPS analysis (A Jonesy derivative work)

Postby Candiru » Thu Aug 06, 2009 4:19 am

Very interesting Theck. Is this with the Judgement-proccing SoV (current behaviour if you have Judgements of the Just specced)?

Also, that means dropping Judgements of the Just for other threat talents is probably a really really stupid idea now, rather than just a stupid one :)

Very strange behaviour, SoR procs of JoR still without JotJ? Does it double proc with JotJ? I'm going to go and test all my seals and judgements and specs tonight I think!
Image
Candiru
 
Posts: 2479
Joined: Mon May 28, 2007 12:21 pm

Re: Theck's MATLAB TPS analysis (A Jonesy derivative work)

Postby theckhd » Thu Aug 06, 2009 5:24 am

Nadir wrote:Theck, would you mind adding Hellscream Slicer/ Stormpike Cleaver (the new ilvl 245 slow weapons) into the mix? No it doesn't have defense, but avoidance wise it's within 0.01%-0.02% of Broken Promise when fully raid buffed. Its got a nice chunk of hit, to boot.

http://www.wowhead.com/?item=47314

I could, but it's pretty clear what will happen there. It's got similar (though slightly better) stats to Vulmir, with a higher damage range and the same speed. It'll be higher than Vulmir across the board.
"Theck, Bringer of Numbers and Pounding Headaches," courtesy of Grehn|Skipjack.
MATLAB 5.x, Simcraft 6.x, Call to Arms 6.0, Talent Spec & Glyph Guide 5.x, Blog: Sacred Duty
User avatar
theckhd
Moderator
 
Posts: 7950
Joined: Thu Jul 31, 2008 3:06 pm
Location: Harrisburg, PA

Re: Theck's MATLAB TPS analysis (A Jonesy derivative work)

Postby theckhd » Thu Aug 06, 2009 6:59 am

My curiosity got the better of me. Since I was adding Crusader's Glory (the new tanking weapon) to the simulation, I threw in Hellscream Slicer as well. Here's the updated plot:

Image
Crusader's Glory (CG on the plot) barely edges out Sorthalis across the board. It's only 1.5 speed though, so seal procs won't be quite as large.

Interestingly, it also beats out Broken Promise for everything but the 5S+5R spec. Those points in Reckoning have gone from around 28 TPS per point to nearly 40 for a slow weapon like BP. Still not quite as good as Conviction, but getting close.
"Theck, Bringer of Numbers and Pounding Headaches," courtesy of Grehn|Skipjack.
MATLAB 5.x, Simcraft 6.x, Call to Arms 6.0, Talent Spec & Glyph Guide 5.x, Blog: Sacred Duty
User avatar
theckhd
Moderator
 
Posts: 7950
Joined: Thu Jul 31, 2008 3:06 pm
Location: Harrisburg, PA

Re: Theck's MATLAB TPS analysis (A Jonesy derivative work)

Postby Candiru » Thu Aug 06, 2009 7:07 am

If only I had more defence on gear I would enchant berserking onto my Razorscale Talon http://www.wowhead.com/?item=45298 no one wanted and swap it in if people were getting close on threat :)
Image
Candiru
 
Posts: 2479
Joined: Mon May 28, 2007 12:21 pm

Re: Theck's MATLAB TPS analysis (A Jonesy derivative work)

Postby raenac » Thu Aug 06, 2009 7:37 am

theckhd wrote:My curiosity got the better of me. Since I was adding Crusader's Glory (the new tanking weapon) to the simulation, I threw in Hellscream Slicer as well. Here's the updated plot:

Image
Crusader's Glory (CG on the plot) barely edges out Sorthalis across the board. It's only 1.5 speed though, so seal procs won't be quite as large.

Interestingly, it also beats out Broken Promise for everything but the 5S+5R spec. Those points in Reckoning have gone from around 28 TPS per point to nearly 40 for a slow weapon like BP. Still not quite as good as Conviction, but getting close.

Hmmm
tribute chest can you say ZOMG
http://www.wowarmory.com/item-info.xml?i=47506
raenac
 
Posts: 15
Joined: Thu Nov 20, 2008 6:57 am

Re: Theck's MATLAB TPS analysis (A Jonesy derivative work)

Postby theckhd » Thu Aug 06, 2009 9:42 am

Raenac wrote:Hmmm
tribute chest can you say ZOMG
http://www.wowarmory.com/item-info.xml?i=47506

Those stats are insane. It's nearly as good as Vulmir:
Image
"Theck, Bringer of Numbers and Pounding Headaches," courtesy of Grehn|Skipjack.
MATLAB 5.x, Simcraft 6.x, Call to Arms 6.0, Talent Spec & Glyph Guide 5.x, Blog: Sacred Duty
User avatar
theckhd
Moderator
 
Posts: 7950
Joined: Thu Jul 31, 2008 3:06 pm
Location: Harrisburg, PA

Re: Theck's MATLAB TPS analysis (A Jonesy derivative work)

Postby Nadir » Thu Aug 06, 2009 2:59 pm

I apologize for increasing your amount of work, but here's the ilvl 258 version of Hellscream Slicer finally (with a nice gem slot):
http://www.wowarmory.com/item-info.xml?i=47475
Image
Nadir
 
Posts: 266
Joined: Mon May 04, 2009 8:37 am

Re: Theck's MATLAB TPS analysis (A Jonesy derivative work)

Postby Artanas » Sun Aug 09, 2009 6:16 pm

just curious is there a legend for the all the abriviations your using?
Artanas
 
Posts: 8
Joined: Mon May 11, 2009 12:50 pm

PreviousNext

Return to Advanced Theorycraft and Calculations

Who is online

Users browsing this forum: No registered users and 1 guest

Who is online

In total there is 1 user online :: 0 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: No registered users and 1 guest