Warrior Threat Testing, Stress Test Feb 1, 2015

Share your secret tips about your favorite class here !

Re: Warrior Threat Testing, Stress Test Feb 1, 2015

by Viper » Sun Feb 01, 2015 10:22 pm

"impressive" is the right word :)
User avatar
Viper
Administrator
Administrator
 

Re: Warrior Threat Testing, Stress Test Feb 1, 2015

by Hatson » Sun Feb 01, 2015 11:04 pm

Image

Very good work. :)
The White Blacksmith.
Sometimes I stream stuff: http://www.twitch.tv/travallama/
User avatar
Hatson
Stone Guard
Stone Guard
 

Re: Warrior Threat Testing, Stress Test Feb 1, 2015

by Undertanker » Thu Feb 12, 2015 3:32 pm

Great test. Does anybody else think demo shout is too low on threat? I remember it was a great too for aoe grabbing in retail vanilla. Just a thought, could be wrong.
Undertanker
Stone Guard
Stone Guard
 

Re: Warrior Threat Testing, Stress Test Feb 1, 2015

by boomerjp » Fri Mar 06, 2015 4:36 am

This kind of test can be performed very easily with some code allowing for a GM account player to use an in game command to view current threat on a target. Probably not even that hard to implement for a test server.
boomerjp
Private
Private
 

Re: Warrior Threat Testing, Stress Test Feb 1, 2015

by Armilus » Fri Mar 06, 2015 4:38 pm

boomerjp wrote:This kind of test can be performed very easily with some code allowing for a GM account player to use an in game command to view current threat on a target. Probably not even that hard to implement for a test server.


Unfortunately that does not test the research of the dev team. They aren't just writing code and making up class mechanics. The goal is to replicate the same threat mechanics that existed in vanilla retail and that means they have to do a LOT of research to find out how those mechanics worked. In fact, the majority of the work involved in fixing Mangos bugs is figuring out how it should be working.

The dev's can easily test their own code, that wasn't the goal of this test at all. The goal was to test their research on the threat mechanics and their thoroughness in looking for mangos bugs. If I didn't think something was working correctly and I simply said "hey dev's, demo shout is broken". That is pretty much useless to them. But if I can say "Demo shout is causing the correct amount of threat; however, it is being divided among all targets in combat instead of being applied to each target affected. You can see that it should not be divided by watching this video of a warrior kiting a crap ton of mobs with demo shout. This would not be possible if demo shout worked with the current mechanics". Now that is something they can work with!
Cadmus - Priest, Nostalrius Begins PvP
Stabsington - Rogue, Nostalrius Begins PvP
Armilus
Sergeant Major
Sergeant Major
 

Re: Warrior Threat Testing, Stress Test Feb 1, 2015

by boomerjp » Sat Mar 07, 2015 3:07 am

Armilus wrote:Unfortunately that does not test the research of the dev team. They aren't just writing code and making up class mechanics. The goal is to replicate the same threat mechanics that existed in vanilla retail and that means they have to do a LOT of research to find out how those mechanics worked. In fact, the majority of the work involved in fixing Mangos bugs is figuring out how it should be working.

The dev's can easily test their own code, that wasn't the goal of this test at all. The goal was to test their research on the threat mechanics and their thoroughness in looking for mangos bugs. If I didn't think something was working correctly and I simply said "hey dev's, demo shout is broken". That is pretty much useless to them. But if I can say "Demo shout is causing the correct amount of threat; however, it is being divided among all targets in combat instead of being applied to each target affected. You can see that it should not be divided by watching this video of a warrior kiting a crap ton of mobs with demo shout. This would not be possible if demo shout worked with the current mechanics". Now that is something they can work with!


Yes, I understand the reasons for testing, and I applaud you for your effort. I just think that if the Nostalrius team was able to add an in game command for you to use on a test server it would go a lot faster! You could also do it with just one character. In my experience, threat is usually working fine on private servers and doesn't need to be extensively tested (mangos default I mean), but I have seen some serious issues with Taunt that can break some fights with threat reduction mechanics. So I would advise that you test taunt if you haven't already.
boomerjp
Private
Private
 

Re: Warrior Threat Testing, Stress Test Feb 1, 2015

by Lecat » Wed Mar 11, 2015 8:45 am

That kind of testing is a great help for devs!
User avatar
Lecat
Sergeant
Sergeant
 

Re: Warrior Threat Testing, Stress Test Feb 1, 2015

by Armilus » Tue Mar 24, 2015 5:01 pm

Update with some demo shout evidence:

Here is a video of a warrior kiting adds on razorgore with demo shout (courtesy of Graves): https://www.youtube.com/watch?v=qVeIEDZkXrs

There are 3 ways that demo shout threat could work:
1) Threat is divided among ALL targets in combat, like healing threat
2) Threat is divided among all targets HIT by the debuff (~42 divided by number of targets affected)
3) The same amount of threat is applied to each target hit with the debuff (~42 threat to each target)

In the video we see that the warrior auto attacks the mob to get initial agro. After that, the warrior kites the mobs in a group and uses demo shout and piercing howl to keep agro over the healers. We also see a second warrior kiting an even larger group of legionnaires.

I think we can definitely rule out #1 from this video and I don't think nost is using this method at all.

So, the next question is, would it be possible to hold agro on this many mobs if demo shout threat is divided among all targets hit? We can see that the warrior does receive some healing so healers are definitely causing threat on all targets. Is it a matter of healer threat being divided among such a large number of mobs that demo shout still works even with mechanic #2?

He isn't exactly spamming demo shout, seems to only use it 1-2 times on each platform.

I can't make out the numbers on the threat meter in the video but it looks like several healers have AT LEAST over 1000 threat on razorgore. Assuming that the threat meter properly divides threat among all targets, the healers have at least 1000 threat on the mobs kited since the beginning of the event. It's hard to tell but it looks to me like some healers have over 4000 threat.

I don't think it is possible for a warrior to be kiting mobs like this if demo shout threat is divided among all the targets hit.
Cadmus - Priest, Nostalrius Begins PvP
Stabsington - Rogue, Nostalrius Begins PvP
Armilus
Sergeant Major
Sergeant Major
 

Re: Warrior Threat Testing, Stress Test Feb 1, 2015

by Armilus » Wed Apr 29, 2015 6:03 pm

Finally found some additional evidence for the demo shout threat.

This is the latest version of Kenco's threat guide that I can find. This is a guide that was originally posted on the official wow forums.
http://www.cgalliance.org/forums/showth ... wow-forums

Battle Shout generates 55 threat for each player that is buffed; up to 5 people in your party and their pets, as long as they are on the relevant mob's threat list. For example if you buff a player that is out of combat, no threat is generated. Therefore in a tightly packed group, Battle Shout can rival Sunder Armor for threat, but as a buff the threat is split amongst all the mobs that are aware of you. On the other hand, Demoralizing Shout gives 43 threat to each affected mob. As long as there are 6 or more mobs affected, it will hold aggro as well or better than Sunder Armor.
Cadmus - Priest, Nostalrius Begins PvP
Stabsington - Rogue, Nostalrius Begins PvP
Armilus
Sergeant Major
Sergeant Major
 

Re: Warrior Threat Testing, Stress Test Feb 1, 2015

by Faustz » Wed Apr 29, 2015 8:04 pm

Great work. Hopefully we can get demo shout working as intended.
User avatar
Faustz
Sergeant
Sergeant
 

Previous

Return to Mechanics and classes