Spiral Knights - Enemy Health

Enemy Health

18 replies [Last post]

Fri, 10/07/2011 - 23:04

Quailia

Hi,

I'm wondering if an article has been made outlining the detailed aspects of what determines an enemies numerical health value. So far as I know, when I'm on a level, I can determine the range of it's HP based on the number of hits I deal to it before it dies, but I'm wondering if this has ever been documented and explained in the wiki. Clearly there is some determination based on the enemy type, the level you're on, and perhaps also a hidden X factor; but as far as I know, I cannot find an article explaining what factors determine the value set.

Sat, 10/15/2011 - 19:07

Equinox

Game Master

Various folks have tried to

Various folks have tried to figure out what they can, but from the discussions I've seen monsters may heal over time, so just using strikes to kill may be too inaccurate. Maybe someone else recalls where this was discussed? I can't recall if it was here in the forum on somewhere on the wiki.

Sat, 10/15/2011 - 22:35

Toastyspiralknight

Hmmm?

You may be able to kill a enemy and count its number from damage it took.But the damage may become a overkill.Which is why telling apart which enemy has more health and which enemy has less health is nearly impossible.Oh and I forgot.Hail admin Equinox!Well thats it.Case closed.Now off to the graveyard you go.

Mon, 10/17/2011 - 11:33

Quailia

Would it be worthwhile

Would it be worthwhile to set up some sort of data collection thread? I mean, the most people would have to do is watch YT videos of people playing solo, and counting the numerical health based on the enemies seen there...

Also, I don't think this topic is "case closed." That's giving up without any effort :\

Mon, 10/17/2011 - 00:35

Buckdida

Well... There would be an

Well...

There would be an "easy" way to document enemy health. I say "easy" because it's a simple plan, but the way it works is tedious and long. That way is using the lowest damage weapon in game to finish off enemies- the Proto Gun (or a low star gun with a damage type the enemy is resistant to). This way, you can get the closest possible value for the enemy's HP. This would be particularly annoying, with, say, T3 wovlers...

Anyway, it would take far longer than you think to figure out all the enemy health values, because they vary greatly depending on the situation. Enemy HP is affected by the following:

-Enemy species
-Tier/Stratum/depth (yes, down to depth I'm pretty sure)
- # Of party members

We'd need a lot of people roaming the clockworks and working together in order to get this done. We can start with solo health values- that would be easier, and you can even do that on a normal-ish run.
With Different party members... well, it doesn't look like it's something simple like "the health doubles" because that isn't the case. For example, I can 3-hit wolvers playing solo with my sword- and I can 3-hit wolvers when I'm playing in a 2-man, or even a 3-man. In a 4-man, the HP seems to raise drastically. It's hard to say how party size changes HP, and that's what we'd need to work hardest one- and you'd need massive cooperation from everyone involved.

As for monsters healing over time, that's an interesting theory, and would need some testing. Simple enough to do- whack an enemy a couple of times until it's near death, or you know the next hit will finish it- then wait. And wait. And wait! Then whack again. Killed it? No regen. Didn't? Yes regen. (Then we'd have to figure out the regen rate, which would be near impossible, and ruin the "easy" plan above.) Personally, I don't think there's any passive regen based on the fact that we're usually shown when an enemy is healing, for example, Royal Jelly, Silkwings, Electrified Quicksilvers.

Toasty, this case is far from closed. Equinox was simply bringing up a different point to consider.

Where we can start is solo running T1, T2, and T3, with people getting approximate numbers with their standard weapons on different monsters. It'll give us a place to start if this gets off the ground.

Mon, 10/17/2011 - 01:37

Trias

I'm not sure if any large

I'm not sure if any large scale effort has been tried. But I think it is worth a try. I think the first thing that should be investigated is if there is any passive healing effect. That shouldn't be too hard.

I think the next step should be to look at a single monster, and gather data on its health. This should allow us to put up conjectures on how depth and party number effect the monster health. It also allows us to check the consistency of the methods used. (A good candidate for this would be jelly cubes, they appear plentiful (and consistently in RJ palace) and are easy to manage, therefore allowing precision measurements with low yield weapons.)

As for methodology. (provided that monsters do not heal over time) You can use any weapon to provide a window where the health must lie by recording the cumulative damage done to the monster and noting the total damage done to the monster before AND after the last hit that killed it. Using measurements from different players should provide different windows, with the true value somewhere in the intersection.

For each monster record: (Monster type, party size, depth, dmg before depth, dmg after kill)

Somethings to look out for:
- Make sure there are no healers around.
- If playing in a party, make sure everybody is onboard with the measurement, and make sure that the enemy you are measuring was not damaged by any of your party members.

This thread should work fine as a start for data collection.

Mon, 10/17/2011 - 16:23

Antistone

I also think that HP varies

I also think that HP varies from depth to depth, even within a stratum, but it's hard to be sure since damage dealt by weapons also varies from depth to depth.

HP definitely varies by the size of the party. I presume that all monsters have their health multiplied by the same amount, but I don't know the exact multiplier for each party size. I think a full four-knight party yields monster HP that is roughly double that for a single knight (it's certainly less than quadruple). This is kind of understandable, but I really think they took it too far with the Jelly King, whose regeneration is annoying for a party of four but totally batsnot insane for a solo knight.

Some experiences have also made me suspect that monsters that have already spawned do not have their health adjusted if players join or leave.

I'd be very interested to know the scaling rules for different party sizes. Knowing the exact HP totals of different monsters seems less interesting, but if you want to mount a project to discover them, I certainly won't stop you.

I noticed that my Glacius takes 3 hits to kill Gun Puppies at depth 15-16 when fighting solo with damage bonus: very high, but that drops to 2 hits with damage bonus: maximum. That probably narrows their HP to within +/-5% or so, but I don't happen to know the actual numbers off-hand. And I believe the "very high" version takes 4 hits in a party of 4, which is consistent with HP doubling since the third hit does a little more than the first two.

Tue, 10/18/2011 - 00:35

Sk-Tactics

It would be worth it

I believe it would be a worthwhile effort to document enemy health as this would increase the usefulness of having damage tables. :-)
However, its going to be a HUGE undertaking so I think it helps to be organized about this...

Currently, what we know about enemy health are:
1) Party Size. (Larger party = more enemy health)
2) Monster Class. (Lumber can take more damage than other constructs & Alphas are definately harder to kill than normal wolvers.)
3) Depth. (Enemies have more health the deeper you venture into the clockworks)

What we are unsure about:
1) Whether monsters have passive regeneration.
2) Whether health varies by depth or stratum. (Personally, I believe it varies by depth too but lets test it just to make sure.)

Testing & Documentation:
Since there's a lot of data collection to be done, I think we should put up a table on the wiki somewhere so that people can contribute data to it on an ad-hoc basis. Instead of have a fixed health value for each monster (like the damage tables), I believe we should express it as a range of values. E.g. It takes 2 mega magnus normal shots (dmg 76) to kill a wolver at D10, so the table should read 77 - 152 for wolvers at D10. Here's a simple table I've whipped up to help visualize.

If, on some other day, somebody else were to confirm that 1 mega magnus charged shot (dmg 125) cannot kill a wolver at D10, then the entry can be updated to become 126 - 152 for D10 wolvers instead. As time goes by & more people update the table with newer values, we should end up with a very good approximation of the health values for each monster.

Compared to testing with a proto gun, it'll take more iterations to get to the point where we have an accurate value. However, I think it will be easier to get more people to contribute data since they can use their favorite weapon.

The draft table provided above is classified by Monster Type / Depth for easy comparison between the different classes. We'll need to have a separate table for each party size to document everything. (total 4 tables)
Alternatively, we can classify it by Party Size / Depth instead & put 1 table on each of the monsters' talk page. Example here.

Complications:
1) Sometimes, some monsters can be "pulled in" from lower depths. (Most notably the alpha wolvers at the last party button in "wolvers den")
Since they take less damage than usual compared to the other monsters in that level, I believe that they will probably have more health than usual too. We need to make sure they are excluded from testing so our data will not be corrupted.

2) Lichens might be able to increase their max hp by merging, so I don't know how we should handle them.

Disclaimer: Having said all that, I really really don't want to spearhead this project. Already have my hands full with weapons testing. However, I don't mind contributing a few values here & there if you guys were to go ahead with this. (^^)/

Tue, 10/18/2011 - 08:39

Juances

Instead of testing each

Instead of testing each monster with each combination of player maybe we should establish a percentage that show how stronger an enemy gets based on # of players and just calculate using that. This, of course assumes that all monsters get the same increased life effect.
It could save some work.

Tue, 10/18/2011 - 16:22

Veroldin

How about the devs just tell

How about the devs just tell us how much health each monster has. That might work.

Sat, 10/22/2011 - 22:18

#10

Sk-Tactics

Update - Monster HP varies by depth

I've done some HP testing on wolvers in tier 1 during weapon damage testing...

Results
Depth 1:
Wolver health = between 26 - 40

Depth 3:
Wolver health = between 77 - 78

Think its safe to say that enemy health varies down to individual depth within a tier.

Sat, 10/22/2011 - 23:54

#11

Antistone

Wow

That's...a big difference. Just to double check, you weren't in a danger room, and the party size was the same in both cases? (And what was the party size?)

Mon, 10/24/2011 - 10:02

#12

Sk-Tactics

Confirmation

I wasn't in a danger room & the party size is one. (Have tried it on several wolvers)
Erm.... actually, its not a big difference if you consider that weapon damage increases as you go deeper.

Depth 1:
Mega Magnus dmg 25/40 vs beasts
Takes 2 normal shots or 1 charged shot to kill a wolver.

Depth 3:
Mega Magnus dmg 32/51 vs beasts
Takes 3 normal shots to kill a wolver.
(I was able to narrow down the range using a voltech mk ii. Dmg 4/6/2)
Either 2 x M.Magnus + 4 x Voltech OR 2 x M.Magnus + 3 x Voltech + 1 x shock will kill a wolver.

Sidenote:
The implication of this system is that if you are using a weapon with star rating equal or greater than the stratum,
both your weapon damage & enemy's health will increase as you go deeper.
The number of strikes needed to kill an enemy increase very slowly so you get a slow & steady increase in difficulty.

If you are using a weapon with star rating less than the stratum,
your weapon will deal less & less damage while enemies get more & more health as you venture deeper.
The number of strikes needed to kill each enemy increases very fast & you get a big spike in difficulty.

I believe this is something that we have all experienced but never really thought about. :-)

Fri, 10/28/2011 - 00:41

#13

Quailia

I am THRILLED

I am thrilled that people have responded in such enthusiasm to this cause. It's a long and tedious one for sure (perhaps more so than the weapon damage project sorry to say), but I think it's a good endeavor to at least dabble in from time to time.

As we all know, the PK event has caused a number of prime cases for each of us to farm certain strata in hopes of finding the elusive creeper. This said, I can assure you that I have at least found one discovery in my own excursions:

When Solo: not only are beast's health less than 40 HP in D1, but also ALL monsters are less than 40 HP (at least at that depth). I've found wolvers, devilites (basic, non upgraded), gun puppies and zombies (etc.) all to be 1-shotted by an Ascended caliber at lvl 10 dealing EXACTLY that value. All that's to be done is to go in smaller increments of damage from there :)

I also believe that the number of players may have a logarithmic effect on enemy health, but that is yet unconfirmed (I was no longer 1-shotting them when a teammate then joined at D1)

As was mentioned above, I believe there may be some differences to be found based on "species" (e.g. your basic wolver vs Ash tail, Frostifur, or Voltail or Rabid variants). Has anyone found two of a single family within t1d1?

Fri, 10/28/2011 - 10:01

#14

Quailia

more random data

Zinqf who has attempted (and I think succeeded?) a 0* twins run came up with this bullet point:

"- Scuttlebots seem to have about 375 HP in depth 17. That equates to about 26 normal hits or 17 charged hits. In other words, they aren't dying unless they are taking rocket hits from the twins."

dunno how he came to that conclusion, but I'm posting it here for reference

Sat, 11/05/2011 - 01:24

#15

Sk-Tactics

...

@Quailia
erm... Can't be 100% sure, but I think Alpha Wolvers at D1 have more than 40hp when soloing...
Believe that there are at least 3 "degrees of toughness" as far as enemy health are concerned.

Wimpy - Menders. (Their health have been reduced since this patch)
Strong - Alpha Wolvers, Lumbers & Retrodes. (Had a level where I took 6 shots to kill retrodes while scuttlebots took 4)
Average - Rest of the enemies.

Now for the bad news...
Was doing some tests in Tier 1 while soloing & got some disturbing results:

D1: Kats have exactly 39hp
D1: Chromalisks between 40 - 41hp

D3: Wings have exactly 76hp
D3: Wolvers have between 77 - 78hp (previous observations)

It seem kind of unlikely that their health would differ by 1 point...
Possible explanations are:
1) I've made a mistake somewhere.
2) There's an rounding off error. (Dmg to Kats done with "prisma mk2 UV low vs undead". Dmg 22 vs constructs, 23 vs undead)
3) Enemies regenerate health with time. (Was shooting at the chromalisks with the prisma mk2, so it took some time to kill)
4) Each type of enemy have different amount of hp. (Which means we have to conduct tests on the 20+ enemies to get the full data)

Sat, 11/05/2011 - 11:43

#16

Littlerbit

Needs moar estimation.

It's safe to assume that enemy health increases according to strata, depth, and tier. Finding EXACT enemy health would be painstakingly difficult, as you have to note the strata and depth, and half to make an estimate. There's also a myriad of monsters in Spiral Knights, which adds to the job of recording health for that monster in every area of the Clockworks.

Sun, 11/06/2011 - 15:52

#17

Pauling

The magic of graphs

Rather than reporting only an average, I would suggest collecting ALL raw numbers for enemy health and displaying the data as a scatter plot (health vs depth) for one enemy type to start. This would better show the spread of error, and by fitting a curve to the baseline values, we could begin to devise an analytical formula that would mathematically predict enemy health at each depth.

Wed, 11/09/2011 - 11:22

#18

The-Ark

Shufflebots

I used my silversix on shufflebots on depth 16, they have 350 health.

Search

Enemy Health