Mongoose Questions

Get help with your character's gear

Moderators: Fridmarr, Worldie, Aergis, Sabindeus, majiben, lythac, Digren

Re: Mongoose Questions

Postby Avengeance » Sat Apr 24, 2010 11:47 am

Mongoose still gives the small haste boost element that +26 agility isnt giving.
Avengeance of Boulderfist-EU
Guild Master of High Tempest
Avengeance
 
Posts: 251
Joined: Sun Feb 08, 2009 6:30 pm

Re: Mongoose Questions

Postby tlitp » Sat Apr 24, 2010 4:22 pm

cds4850 wrote:Rather than speculating on this, can anyone explain to me how I can search a combat parse for uptime? If so, can you toss me the parameters that I should use while compiling the data set (ie should I run it hit cap/exp cap, naked, wearing a wedding dress?)

Code: Select all
+0.1 base swing timer >> +20 expertise rating (to softcap) > +20 hit rating (to softcap) > +20 expertise rating (to hardcap)

Slower weapons are by far the best option to optimize ppm-based effects (check, for reference, the Vindication discussion).

fuzzygeek wrote:(...) check against real-world data (...)

Agreed. The average 3-5 min long kill isn't relevant, though.

(goose,bw,zerk) :
- 1 ppm
- not haste-normalized
- same triggers (AA,HotR,J/debuff/JotJ)
Obviously, a (stealth-)hotfix is always a possible option. Then again, it's quite reasonable to assume that someone would have noticed, after more than three months away from the last known change in mechanics (early January).

Digren wrote:(...) 22% uptime gives Mongoose an average agility of 120 x .22 = 26

Do remember that post-DR avoidance is not a linear map. If Mongoose would have the quoted 26/120 (~21.67%) average uptime, its corresponding (time-averaged) avoidance output would be strictly lower than the passive output of 26 agility.

A simple numerical example should make this quite clear. Let's say a Paladin has no defense/dodge/parry rating from his/ger gear, and no non-base agility.
  1. A passive effect that grants 100 dodge rating will lead to dr(dodge=100) = ~2.25% avoidance gain.
  2. An active effect that grants 500 dodge rating, with 20% average uptime, will lead to dr(dodge=500)*0.2+dr(dodge=0)*0.8 = ~2.04% time-averaged avoidance gain.
While the average input is identical (100*1=500*0.2), the output is not.
User avatar
tlitp
 
Posts: 556
Joined: Mon Jul 27, 2009 3:25 pm

Re: Mongoose Questions

Postby Digren » Sat Apr 24, 2010 7:22 pm

Avengeance wrote:Mongoose still gives the small haste boost element that +26 agility isnt giving.

But that makes it a threat ability only, and a poor one at that because it's threat is uncontrollable. If I'm not hit capped I'd rather have Accuracy on my threat weapon, and Blood Draining on my EH weapon, and Agility on my block cap weapon. Mongoose isn't even on the list.
User avatar
Digren
Moderator
 
Posts: 2139
Joined: Sat Jul 26, 2008 2:41 pm

Re: Mongoose Questions

Postby Shoju » Tue Apr 27, 2010 10:01 am

I guess I'm seeing higher than average uptimes on my Mongoose then possibly attributed to tanking with a DPS weapon? Going through parses of boss fights, I'm between 27 and 31% uptime.
User avatar
Shoju
 
Posts: 6396
Joined: Mon May 19, 2008 7:15 am

Re: Mongoose Questions

Postby cds4850 » Tue Apr 27, 2010 12:49 pm

I hadn't even thought of searching for the lightning speed proc to evaluate uptime. I'll check logs later in the week, but I'm skeptical that the average agility over time with recent reports will exceed the static +26agi enchant. Might just be time to dust of that Blood Draining vellum...
cds4850
 
Posts: 135
Joined: Thu Sep 03, 2009 7:49 am

Re: Mongoose Questions

Postby Ardrhyst » Tue Apr 27, 2010 8:49 pm

I just checked my World of Logs for tonight, and my uptime was pretty abysmal. Overall, I've got a 10% uptime, but that includes some standing around in fights. On constant combat fights like Marrowgar and Saurfang, I'm still at only around 20%.

Unfortunately, I think I'll still be sticking with it. Blood Draining seems unnecessary, seeing as how I haven't felt like tank death was a danger in awhile, and the 15% buff only reinforces that. My threat's not an issue, and that makes Accuracy seem fairly lackluster. Exceptional Agility would be a decent option, except that it's got a lame glow. I will not sacrifice my sexiness for the well-being of the raid. It's amazing that the Wrath enchants are so meh that there's barely any impetus to change. Even if I did go to 26 agi (Which seems like the "winner"), it ends up being almost the same effect in the grand scheme, as has been walked through in this thread.
Biggest Game Hunter, as well as collector of pets and mounts. Currently hunting paladins' former glory.
User avatar
Ardrhyst
Maintankadonor
 
Posts: 129
Joined: Tue Jan 06, 2009 2:16 pm

Re: Mongoose Questions

Postby Digren » Wed Apr 28, 2010 7:38 am

And glow color triumphs once again.

Accuracy's sexy white is pretty hot.
User avatar
Digren
Moderator
 
Posts: 2139
Joined: Sat Jul 26, 2008 2:41 pm

Re: Mongoose Questions

Postby Digren » Wed Apr 28, 2010 7:57 am

I reworded the comments in the weapon enchant section of my guide to link to this thread, and to mention its new lower uptime.
User avatar
Digren
Moderator
 
Posts: 2139
Joined: Sat Jul 26, 2008 2:41 pm

Re: Mongoose Questions

Postby theckhd » Fri Apr 30, 2010 6:30 am

Digren wrote:I reworded the comments in the weapon enchant section of my guide to link to this thread, and to mention its new lower uptime.

I'm curious whether this is a real or perceived effect. I've been using Accuracy for quite a while now, so I don't have any recent logs with Mongoose to look at for uptimes. Would anyone mind linking a few? I don't have a lot of time lately, but if I find some I'd like to see if there's been any stealth changes to the proc triggers that would explain the uptime dropping by about half.

If the observed uptimes are correct, then Agility becomes a more attractive option than Mongoose for survival. Mongoose is still going to be a little ahead in the threat department though (but not as high as Accuracy).
"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: 8030
Joined: Thu Jul 31, 2008 3:06 pm
Location: Harrisburg, PA

Re: Mongoose Questions

Postby tlitp » Fri Apr 30, 2010 7:03 am

theckhd wrote:
Digren wrote:I reworded the comments in the weapon enchant section of my guide to link to this thread, and to mention its new lower uptime.

I'm curious whether this is a real or perceived effect. (I)
I've been using Accuracy for quite a while now, so I don't have any recent logs with Mongoose to look at for uptimes. Would anyone mind linking a few? I don't have a lot of time lately, but if I find some I'd like to see if there's been any stealth changes to the proc triggers that would explain the uptime dropping by about half. (II)

(I) Defective induction. People see results of ~5 min long "tests". Real-world tests.
(II) The said results can be explained by a careful examination of the workframe :
  1. Is the target a heavy caster ? If so, player's parryhaste will have a diminished significance.
  2. Does(/can) the player follow religiously the 6/9 cycle, without "wasted" GCDs, or delays, or lag ? If not, the expected uptime drops.
  3. Is the uptime of various dynamic effects (WFT/SR/HP) constant ? If not, the expected uptime drops.
  4. Is the time on target constant ? If not, the expected uptime drops.
We return to square one. Depending on the encounter at hand, people will see "abnormally low" values of ppm-based effects. Does that necessarily mean that their mechanics have been altered ?
As always, the answer will be found in proper testing. :P
User avatar
tlitp
 
Posts: 556
Joined: Mon Jul 27, 2009 3:25 pm

Re: Mongoose Questions

Postby theckhd » Fri Apr 30, 2010 7:29 am

(I) is exactly what I meant by real vs. perceived. A 5-minute log is not enough data to base anything on, especially since 969 adherence and all of the factors listed in (II) can drastically affect the uptime. That said, if I get a lot of logs with different people adhering to 969 very strictly and see low uptime on all of them, then it's worth investigating further.

I think the preferred method to do so would be to look at proc triggers again rather than real-world combat parses. The only thing that would cause an uptime change is a proc trigger change or a ppm nerf. Both of those can be tested "in the lab" so to speak, given enough time and target dummies. And that removes most of the outside influences of (II) that can screw up the data.

<edit> I guess I should add that real-world data is still important for fight mechanic reasons. If a fight prevents you from meleeing for a while (or slows your swing speed, like Sindragosa), your mongoose uptime will obviously be lower, and that impacts the value of the enchant. So that's another reason for requesting parses to examine.
"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: 8030
Joined: Thu Jul 31, 2008 3:06 pm
Location: Harrisburg, PA

Re: Mongoose Questions

Postby cds4850 » Fri Apr 30, 2010 8:33 am

Aren't our 5minute encounters the "real world" situations from which we should derive conclusions since they are in fact the actual encounters that we are gearing for? No one here is going to argue that we are not hitting on target dummies in Icecrown. Instead our targets are casting, moving, booooooooonnnne stooooooorrrm-ing, swapping, etc.

That being said, if parses from those encounters consistently reveal lower uptime of the proc (and so averaging less agility over time than the static 26) can we not simply close the book on mongoose being viable for this current content?
cds4850
 
Posts: 135
Joined: Thu Sep 03, 2009 7:49 am

Re: Mongoose Questions

Postby Ardrhyst » Fri Apr 30, 2010 9:49 am

http://www.worldoflogs.com/reports/0oe3lsuk554e1rea/ is the log that my RL did on Tuesday. I didn't weapon switch like I normally do on runs, and instead used my standard gear and standard weapon (with Mongoose) for the entire raid. We had a crappy turnout, so there isn't as much boss coverage as I would like.

Obviously, it's a limited data source, but I'd still expect to see some higher uptime, especially on a fight like Saurfang, where I'm constantly in melee, even if I have to hold my HotR for sections of the fight.

EDIT: I will update this post with more logs as they are made. Hopefully they will be of some use.
Biggest Game Hunter, as well as collector of pets and mounts. Currently hunting paladins' former glory.
User avatar
Ardrhyst
Maintankadonor
 
Posts: 129
Joined: Tue Jan 06, 2009 2:16 pm

Re: Mongoose Questions

Postby tlitp » Fri Apr 30, 2010 10:05 am

cds4850 wrote:5minute encounters (...) "real world" situations (...) conclusions

Real-world situations are fine and dandy. 5-min encounters are not : sLLN.
User avatar
tlitp
 
Posts: 556
Joined: Mon Jul 27, 2009 3:25 pm

Re: Mongoose Questions

Postby fuzzygeek » Fri Apr 30, 2010 10:14 am

tlitp wrote:
cds4850 wrote:5minute encounters (...) "real world" situations (...) conclusions

Real-world situations are fine and dandy. 5-min encounters are not : sLLN.


The obvious solution is to get a thousand parses of 5-min encounters, yes?

This seems more useful than strict lab testing, but it's also kind of addressing a different question.

Q: What is the uptime on Mongoose?
A: Test dummy.

Q: How useful is Mongoose in current content?
A: X parses, where X is a large number.

Also to consider is your own individual situation and adherence to rotation, gearing, and raid execution.
Image
User avatar
fuzzygeek
Maintankadonor
 
Posts: 5130
Joined: Wed Aug 01, 2007 1:58 pm

PreviousNext

Return to Gear Discussions and Advice

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