Quantifying the Effect of Knockdown Zones

So I've been trying to quantify how effective knockdown zones are at keeping enemies occupied. I've noticed some work better than others but never really known why. I'm hoping to gain a better understanding.

This is what I know. There are three main variables that appear to be at play:
1) Knockdown Chance - the chance of knockdown each pulse
2) Max Number of Targets - the number of targets the power can hit
3) Pulse Rate - how often the power 'tics'

There is also a less obvious variable as well:
4) Duration of Knockdown Animation - how long the enemy is occupied by a knock. I assume this value is a constant. Things will be tricky if it is not.

The powers I chose to look at are Ice Slick, Earthquake, Freezing Rain, Liquefy and Oil Slick Arrow.

Base Numbers
- It turns out that the Pulse Rate of all the zones is 0.2 seconds.
- The duration of the knockdown animation is not known to me. I assumed that it is 2 seconds long.
- I'm not sure if Arcanatime is a factor in the calculation, or how to begin incorporating it


Results Table



- Pulses per second shows how fast the zone pulses (0.2 seconds for all)

- KD chance per Tick shows the % chance of a knockdown.

- Max Targets shows the max targets the power can hit, according to Red Tomax.

- Streak Chance refers to the chances that at at any given point in time an enemy inside the zone is currently occupied by the knock animation. Note that Streak Chances are very low when the zone first activates, and don't reach full effectiveness until the length of time equal to knockdown animation duration. This number assumes Max Targets hasn't been exceeded.

- % Standing At Any Time is the inverse of Streak Chance. It shows the odds that a given enemy is standing at any particular point in time.

- # Standing At Any Time refers to how many enemies are currently standing up at any random point of time if there are 16 of them in the zone and we ignore the Max Targets parameter.

- % Standing At Any Time (Max Targets) refers to the percentage of enemies left standing if we assume 16 enemies are in the zone and the knockdown effect can only hit as many as Max Targets allows.

- # Standing At Any Timerefers to the number of enemies currently standing up if there are 16 enemies in the zone and the Max Targets parameter is used.



Now, I'm not totally sure about some of my assumptions here. The longer the knockdown animation time is, the more effective the zones are. The other parameter I'm not sure I have right is the Max Targets piece. If there are more than the number of max targets in an AoE, how are targets selected? I assumed it is always the first 10 targets, but if it is a random selection process then the numbers will be very different.

Thoughts?

---

One other thing. When I came up with these numbers, I assumed that an enemy in any stage of the knockdown animation can be knocked down again. Reality may be somewhat different. How long do you think the lockout period is between consecutive knockdowns?

---

From my experience chain knockbacking many foes on teh way to 50 with my FF defender, my observation is that you can't knock something back again until it finishes standing up from the last one. You have to time it so that the new knockback effect hits a split second after the standing up animation finishes. If you're too early, they fly through the air but don't fall over (maybe even shooting at you along the way), which looks really funny but doesn't provide any mitigation.

Since knockdown is just low magnitude knockback, I would think it would work in the same way, so they couldn't be knocked down again until they finish standing up from the last one.

However, that said, I'm not actually sure it *does* work that way for knockdown patches. Reason being, although I never paid much attention to any individual foe, I'm almost certain I remember seeing foes being chain KD'd without ever finishing standing up by my dom's ice slick. Further testing is in order, I think.

As for the target cap, my bet would personally be on it randomly selecting 10 targets out of the 16 each time it ticks. I don't have anything except gut feeling to go on for this, though. We definitely know, though, that when AoE powers hit multiple targets, the game handles the hits sequentially - we can tell this because we can see the streakbreaker forcing hits in the combat logs if you miss too many of the targets in the AoE. Presumably, the game picks a target, rolls to hit (if applicable, obviously not for a power like ice slick), notes the effects to be applied, and then moves on to the next target, stopping when it runs out of targets or hits the target cap for the AoE. The big question, though, is 'how does it pick the next target - randomly, or in some set order?' My guess is randomly, but that's only based on the fact that that's the way *I* would have wanted to do it if it were up to me.

If we really wanted to test this, I would think it would be best to go to a PvP zone, line up several players in a line, and then hit the guy on the end with a large radius TAoE and see in what order the tohit rolls show up in the combat logs (presumably, this order reflects the order which the game rolled the hits). If it does go by distance from the caster or distance from the focus point, it would be fairly easy to tell the difference between that and a random order.

If it does select targets randomly, then calculating stuff isn't too bad - any one mob simply has a (target cap)/(# of targets) chance of being targeted by the power each time it ticks. Figuring out how to deal with the lockout period, if there does turn out to be one, will be more complicated, but it should be able to be done.

If, on the other hand, it picks targets in some non-random order, well, I'm not sure what to do in that case.

---

@MuonNeutrino
Student, Gamer, Altaholic, and future Astronomer.

This is what it means to be a tank!

From my observations (not testing and timing) it appears that the faster base movement a mob has the faster it regains it's feet after being knocked down/up/back.

Also from observation a mob's powers are only locked out for the duration of the first applied KD/U/B and they are able to animate an attack immediately after regaining their feet. Once they regain their feet they can be knocked again and have their powers locked out until they again regain their feet.

If they are knocked again during the original "getting up period" they continue in the getting up animation even though they are in motion.

This leads to the phenomena I have seen before of knocking a mob just before it regains it's feet and even though it flys backwards it lands on it's feet and can be activating attacks during the travel period of the second knock.

EDIT - Doh Muon beat me to it but our observations concur.

Also judging from my experiences with blaster Ice patch (max targets = 5) it's completely random which mob gets knocked when the target cap is exceeded.

---

-Insanity is doing the same thing over and over again and expecting different results. - Albert Einstein.
-I do not feel obliged to believe that the same God who has endowed us with sense, reason, and intellect has intended us to forgo their use. - Galileo Galilei
-When injustice becomes law, resistance becomes duty. - Thomas Jefferson

I played a little bit with demorecord tonight but didn't figure much out about the timing yet. Watching the animation of enemies fall, though, the graphic implies that they are protected from subsequent knockdowns up until they start trying to stand up again, but can get knocked down while the trying to stand up animation is playing. I haven't figured out if there is a standard speed that they stand up or fall at.

I did, however, find out that Ice Slick has 1070 HP.

---

I'm afraid I can't comment on the math/method, but as far as targetting when over the cap is concerned:

I think the way it works is that each pulse it'll check on its target cap worth of targets, so say you've got 20 mobs on a 10capped patch. It'll check 10 of them each pulse, but each pulse can only check/affect a mob that wasn't checked/affected by the previous pulse. So if a mob was KDed/is getting up then it can't be checked for KD by the next pulse.

So assuming targets numbers 1-10 all got checked for KD/KDed by the first pulse, targets 11-20 would be checked for the second pulse.

At least, I think that's how it works >.>

---

Nice work, Oedipus_Tex!

---

Based on rule of thumb alone I think you guys are right that the selected targets are random. My reasoning is only that if it selects enemies sequentially what we should be seeing from Ice Slick is 10 guys down 80% of the time and a bunch of people conspicously standing up. On the other hand I swear Ive seen cases where there us that one guy who manages to truck all the way from one side of the slick to the other without falling. It could be just luck. I think it's pretty easy to test just by finding a group bigger than 16 and see if they all fall. To eliminate distance from centerpoint as a variable I can immob them first and wait for the -kb to wear off. Sitting in the airport now but will look onto it when I get home.

---

As far as repeating KD goes I have actually observed two different things - the first is the same one made by Muon and M_K, where a standing mob is semi immune to KB until they have just finished the standing up animation. The second is in the split second right after they are knocked down just before they start standing up - if you hit them again right then with a second KD they kind of 'bounce' up and down again, extending the KD period.

I noticed it mostly with my bane who will open with a shatter, knocking the first target down and then frequently follow up with a crowd control if he is in a group, but I have also seen it with ice slick where sometimes a mob will be knocked down then bounce 1 or 2 times - probably because the RNG is in a low streak so the chance to KD pinged a couple times in a row.

This does appear to extend the KD time although it is hard to say how long.

---

Globals: @Midnight Mystique/@Magik13

---

The Bacon Compels You.

---