Difference between revisions of "Dodge 1 (Source)"

Latest revision as of 19:24, 9 March 2009

As promised, I spent several hours this weekend testing our new Defensive Mods. While there is still a lot of stuff that I'd like to test, the results of the tests that I ran this weekend were really interesting. Here's a quick summary of what I found:

1) Your opponent's accuracy greatly affects your ability to dodge his attacks. This is something we already knew, but it was confirmed by these tests. If your opponent is very accurate, you dodge far less than you would against an inaccurate opponent. You'll dodge the most against inaccurate mobs and you'll dodge the least in PvP, because human players are generally far more accurate than PvE mobs.

2) Dodge is weapon-specific. If you do not have a Pistol equipped, your Dodge modifiers will not work. I ran over 200 tests, both with a pistol equipped and with nothing equipped, and the results are conclusive. This probably means that the Dodge skill from Fencer does not stack with the Dodge skill from Pistoleer, although I have not tested that and thus cannot be certain.

3) Against an accurate opponent in PvP (Master Bounty Hunter), dodge is effective -- as a Master Pistoleer (+105 Dodge), I dodged around 20% of his attacks.

4) Against an inaccurate PvE mob (Cowardly Gurreck), I dodged FAR more (as a Master Pistoleer) than I did against the PvP opponent.

5) Defense v. Knockdown probably requires you to have a pistol equipped as well, and thus probably does not stack with the Defense v. Knockdown from other professions. I'll need more data to be completely certain of this conclusion, but the data I do have strongly suggest that Defense v. Knockdown works like Dodge -- you need a pistol equipped.

6) Defense v. Knockdown (50%, Master Pistoleer) was effective against an opponent who was using a CDEF Carbine to knock me down. The knockdown only works 50% as much as it does when a pistol is not equipped. More data is needed to get a precise answer here.

7) Defense v. Knockdown (50%, Master Pistoleer) was NOT effective against an opponent who was using a Laser Carbine to knock me down. More data is needed to be sure of this conclusion, but based on the tests I ran it doesn't look like Defense v. Knockdown does anything if the other guy is using a good Laser Carbine.

8) Based on the above, it appears that your ability to knockdown your opponent is affected by the amount of damage that your weapon is spitting out. The Laser Carbine was far more likely to knock me down than the CDEF when I had a pistol equipped. A lot more testing is needed to be sure of this ... it's something I can work on in upcoming weeks.

As you can tell by my conclusions, I focused on Dodge and Defense v. Knockdown because my impression was that people were the most anxious to get those tested. I'll keep on testing other things, so feel free to give me suggestions of other stuff that you would like me to take a look at. I'm going to test Defense v. Blind and get more data on Defense v. Knockdown as soon as I can get a chance. (As a side note, the Dodge animation is still a major problem -- it stops you dead in your tracks when you're moving, which is often a bad thing when you're trying to get away from a mob.)

Below are the actual test results, so you can see where all my conclusions come from. For folks who are new to this sort of thing, here is how I run tests: First, I use /log to capture the chat logs (including the spam from the Combat window). Then I copy-and-paste the log into Microsoft Excel -- it's just a text file -- and get rid of all the non-Combat chatter. Then I use Microsoft Excel to analyze the data.

DODGE TEST RESULTS

TEST ONE: Dodge with pistol UNEQUIPPED versus a PvP opponent

Defender = me (Master Pistoleer + BH pistol line, which has no Dodge) Attacker = Master Bounty Hunter, using a CDEF Pistol (ideal range = 15m) Range = 15m

HITS: 265 MISSES: 7 DODGES: 1 (out of 273 attacks)

I dodged less than 1% of attacks, as a Master Pistoleer, when I didn't have a pistol equipped.

TEST TWO: Dodge with pistol EQUIPPED versus a PvP opponent

Defender = me (Master Pistoleer + BH pistol line, which has no Dodge) Attacker = Master Bounty Hunter, using a CDEF Pistol (ideal range = 15m) Range = 15m

HITS: 220 MISSES: 11 DODGES: 66 (out of 297 attacks)

With a pistol equipped, I dodged roughly 22% of attacks from a Master Bounty Hunter (+120 Pistol Accuracy).

TEST THREE: Dodge with pistol EQUIPPED versus a highly inaccurate PvE opponent

Defender = me (Master Pistoleer + BH pistol line, which has no Dodge) Attacker = Cowardly Gurreck (as you can see, a very inaccurate mob) Range = melee

HITS: 17 MISSES: 87 DODGES: 50 (out of 154 attacks)

The Cowardly Gurreck almost never hit me anyway, but when he did I dodged his attack. Extremely effective.

DEFENSE V. KNOCKDOWN TEST RESULTS

It was difficult and time-consuming to test Defense v. Knockdown, largely because of the knockdown timer. (The knockdown timer prevents you from knocking down an opponent more than once every 30 seconds.) For this reason, the best way we came up with to test Defense v. Knockdown was to test HOW MANY knockdown attacks it took to knock the Pistoleer down.

TEST ONE: FireKnockdown using CDEF Carbine against Master Pistoleer with pistol UNEQUIPPED

Defender = me (Master Pistoleer + BH pistol line, which has no Defense v. Knockdown) Attacker = Master Bounty Hunter, using FireKnockdown with a 17-35 CDEF Carbine (ideal range = 35m) Range = 35m

Number of FireKDs that it took to knock me down (10 tests): 3, 2, 2, 1, 1, 1, 2, 1, 4, 2 (AVERAGE = 1.9 shots to knock me down if I had NO pistol equipped)

TEST TWO: FireKnockdown using CDEF Carbine against Master Pistoleer with pistol EQUIPPED

Defender = me (Master Pistoleer + BH pistol line, which has no Defense v. Knockdown) Attacker = Master Bounty Hunter, using FireKnockdown with a 17-35 CDEF Carbine (ideal range = 35m) Range = 35m

Number of FireKDs that it took to knock me down (10 tests): 4, 10, 10, 6, 3, 6, 1, 1, 1, 4 (AVERAGE = 4.6 shots to knock me down when I had a pistol equipped)

This is a significant difference from Test One, which leads me to believe that Defense v. Knockdown -- like Dodge -- only works when a pistol is equipped. Although more data is needed, it looks like Defense v. Knockdown stops around 50% of successful knockdown attacks from a CDEF Carbine. Much more data is needed, though, because as you can see the knockdown results seem to be all over the map.

TEST THREE: FireKnockdown using Laser Carbine against Master Pistoleer with pistol UNEQUIPPED

Defender = me (Master Pistoleer + BH pistol line, which has no Defense v. Knockdown) Attacker = Master Bounty Hunter, using FireKnockdown with a 42-403 Laser Carbine (ideal range = 50m) Range = 35m

Number of FireKDs that it took to knock me down (10 tests): 2, 1, 1, 3, 2, 8, 1, 1, 3 (AVERAGE = 2.3 shots to knock me down if I had NO pistol equipped)

It seems to take roughly two shots on average to knock down a opponent with NO Defense v. Knockdown using FireKnockdown with either the CDEF Carbine or the Laser Carbine. Actually, based on this data it appears that it is slightly more difficult to knock the opponent down if you use a Laser Carbine (as opposed to a CDEF) but the difference is so small that it's well within the fairly high margin of error (given that we only have 10 data points). My guess is that the chance of knocking down an opponent with either gun is roughly equivalent if the target doesn't have Defense v. Knockdown.

TEST FOUR: FireKnockdown using Laser Carbine against Master Pistoleer with pistol EQUIPPED

Defender = me (Master Pistoleer + BH pistol line, which has no Defense v. Knockdown) Attacker = Master Bounty Hunter, using FireKnockdown with a 42-403 Laser Carbine (ideal range = 50m) Range = 35m

Number of FireKDs that it took to knock me down (10 tests): 4, 2, 7, 1, 1, 1, 2, 3, 1, 3 (AVERAGE = 2.5 shots to knock me down when I had a pistol equipped)

This is an interesting result -- it seems that Defense v. Knockdown does nothing when the opponent is using a high damage Laser Carbine. This leads me to believe that your ability to knockdown your opponent is affected by the amount of damage that your weapon is spitting out. More data to test that conclusion, although the stark differences in the tests I ran are pretty significant. More tests are needed, however, to confirm that 50% Defense v. Knockdown is ineffective against an enemy who is using a high damage weapon.

I hope this data is helpful to all of you. I'll keep testing further when I have the time, but I think that my hard work from this weekend yielded some interesting results and gives us some hard numbers to point to.