Boss armor values

[Ezmode]

(03-28-2012 12:50 AM)LagunaD Wrote:  

(03-27-2012 06:09 PM)Rystet Wrote:  Dont use the pts training dummies to test def either, they dont have any defense atm.

/facepalm

I figured this was the case when someone showed me parser results with 105% accuracy and no misses...

[Rystet]

I believe that a couple things have caused trouble locking down the 10% Defense previously.

I mentioned earlier the issues with Combat Text not displaying all results. As a Juggernaut many of my attacks actually strike the target multiple times, and each strike is rolled separately. So often its only a portion of an attack that is defended not the entire attack. This can cause confusion when trying to track it without logs. Then on top of that SCT is combining numbers, and who knows whats actually happening behind the scenes.

The other thing that causes many people to not realize they are being defended is that many of their key abilities can not be defended. Recently we have seen a lot of talk about Tanks in PVP and not actually using Tanking gear. PVP has shown us that all Force/Tech attacks (the Key abilities for most classes) can not be defended... only melee/ranged attacks. This cuts down the usefulness of the stat in PVP substantially. This holds true in PVE, bosses can't defend them either.

This is from the new Flashpoint, at 98.97% Base Accuracy, and 108.97% Special, the data parsed from those runs clearly shows my special Melee attacks are being Defended (Dodged/Parried). It did not happen very often with my current accuracy level, but I was defended 2 to 3 times with each of my Special Melee attacks when I compiled all of the Data. My force attacks as expected did not Miss, or get Defended. The only attack that actually "missed" was my basic Assault and this also only happened a few times.


Melee Attacks

Sundering Assault - Hit 481 times, 3 Defended

Crushing Blow - Hit 154 times, 2 Defended

Ravage - Hit 172 times, 2 Defended

Force Charge - Hit 124 times, 1 Defended (damage component only)

Vicious Slash - Hit 41 times, 1 Defended

Sweeping Slash - Hit 248 times, 2 Defended

Saber Throw - Hit 59 times, 0 Defended


Force Attacks

Backhand - Hit 79 times, 0 Defended

Smash - Hit 323 times, 0 Defended

Force Scream - Hit 216 times, 0 Defended

Force Choke - Hit 63 Times, 0 Defended

Force Push - Hit 2 Times, 0 Defended


As a reminder many of these abilities hit multiple times when activated. Each hit is rolled separately. The Hits Listed are not the number of times the ability was activated, but how many times a portion of the ability did damage. So, Often times the Defend is not the whole ability, but just one portion of the attack.

[Kaedis]

(03-28-2012 03:53 AM)Rystet Wrote:  This is from the new Flashpoint, at 98.97% Base Accuracy, and 108.97% Special, the data parsed from those runs clearly shows my special Melee attacks are being Defended (Dodged/Parried). It did not happen very often with my current accuracy level, but I was defended 2 to 3 times with each of my Special Melee attacks when I compiled all of the Data. My force attacks as expected did not Miss, or get Defended. The only attack that actually "missed" was my basic Assault and this also only happened a few times.


Melee Attacks

Sundering Assault - Hit 481 times, 3 Defended

Crushing Blow - Hit 154 times, 2 Defended

Ravage - Hit 172 times, 2 Defended

Force Charge - Hit 124 times, 1 Defended (damage component only)

Vicious Slash - Hit 41 times, 1 Defended

Sweeping Slash - Hit 248 times, 2 Defended

Saber Throw - Hit 59 times, 0 Defended


Force Attacks

Backhand - Hit 79 times, 0 Defended

Smash - Hit 323 times, 0 Defended

Force Scream - Hit 216 times, 0 Defended

Force Choke - Hit 63 Times, 0 Defended

Force Push - Hit 2 Times, 0 Defended

As a reminder many of these abilities hit multiple times when activated. Each hit is rolled separately. The Hits Listed are not the number of times the ability was activated, but how many times a portion of the ability did damage. So, Often times the Defend is not the whole ability, but just one portion of the attack.

This data gives an average of 0.8527% miss chance, which gives an estimated defense cap of 109.82, consistent with the newly suspected 10% cap. Your Assault ability should have had a 10% rate of being defended and a 1.03% rate of missing, on average, since Accuracy seems to consume "miss chance" before it does "defense chance".

[LagunaD]

(03-28-2012 11:11 AM)Kor Wrote:  This data gives an average of 0.8527% miss chance, which gives an estimated defense cap of 109.82, consistent with the newly suspected 10% cap. Your Assault ability should have had a 10% rate of being defended and a 1.03% rate of missing, on average, since Accuracy seems to consume "miss chance" before it does "defense chance".

And the 95% CL range of his defend rate is 0.0043-0.0153, which is consistent with 10% defense. Being defended at all makes the data inconsistent with 8% defense, obviously.

This pretty much nails it for me, unless conflicting data of similar quality emerge.

P.S. Give the man some rep (I don't think I'm exalted enough to do it...)

[hxxr]

Fair enough. One last question remains though - this is the PTR, what evidence is that there was no change between Live and 1.2?

[Izod]

(03-27-2012 06:09 PM)Rystet Wrote:  In regards to the boss defense question presented earlier, I had a developer at the guild summit come right out and say it was 10%.

Yes to Kor's point they also said Taunt works as a Fixate, which testing found was not how it is working.


RE hxxr: Correct me, if I'm wrong but I believe CMF's original findings came from Live - although they are unposted. Regardless it won't matter much in a couple of weeks anyways. I don't really see why they would change that stat.

[hxxr]

Well, no one who knows what cmf did tells anything about it, that's why this wasn't settled pages ago.

And yes, when 1.2 hits Live all of this doesn't matter.
There might be a reason in upping the boss defense to buff the value of accuracy.

[frmorrison]

(03-29-2012 05:54 AM)hxxr Wrote:  There might be a reason in upping the boss defense to buff the value of accuracy.

The devs are supposed to change the scaling of all the stats, so if the "buff" accuracy like the other ones, then you will get more % to hit out of the stats you already have, so it makes sense if it was buffed. Alternately, like has been said many times the pop-up text isn't always 100% correct, so that could cause mistakes.

[Gorlough]

Pretty interesting.
This allows for an accurate calculation on how many accuracy can be traded for other secondary stats.
In Rystets post we see a clear distribution on skills used.
One could add up the damage on those abilities (averages by parser) on current gear, and derive the damage outcome on lower/higher accuracy ratings and other stats (rearranging the given pool of stat points).
I'm pretty sure (just guessing - no solid maths behind that), that we may see a stat distribution, that favors less than 110% special accuarcy, but is highly dependant on your usage of melee/ranged attacks and thus very individual.

[cmf]

I'm afraid a few wires got crossed at some point – I did perform additional tests to confirm my earlier findings, however all tests were conducted on live servers and contrary to Kor's post above I will not be releasing those results. As others have stated 1.2 will be out fairly soon after which independent confirmation of the hit caps of various bosses will become possible.

Next, I'd like to clarify a statement I made earlier – when I stated the claims made should be fairly easy to test under controlled conditions I meant through the use of the same sort of brute force tests used earlier in this thread. That is, by measuring the effective damage reduction or miss rate of a particular attack against a boss, possibly as debuffs are varied. I'm afraid I have not come up with a clever testing methodology I could share that would dramatically reduce the effort it would take to test those claims.

Based on the follow-up tests I conducted however, I can offer the following: the first group of units in a hard mode Black Talon contains an Imperial Lieutenant (elite) and three Black Talon Marines (strong). The three marines will not respawn when the player exits the area and the melee/ranged resistance of the lieutenant is 10%. Based on a very limited set of brute-force tests I can confirm misses were observed on this unit on special attacks with a special attack accuracy of over 108%. Using this unit as a practice dummy it should at the very least allow players to fairly easily verify that misses occur while using a particular attack at accuracy levels close to 110%.