Mob Scaling Findings: Higher Armour is not better

[Avaree]

Since Sam already stated Devs will look into this, why not just enjoy the Weekend and wait what they come up with? We have Summer, Soccer WM, and the Big Event is near, let s all take a Rest!

Cheers!

[Corrupt]

crowsfeet why so mad? You act as if Sera insulted you directly. The devs noticed/ commented and are investigating, therefore Sera was successful in her mission. You mad?

[Ishtmeet]

Sera,
Low lv elites don't scale up. Okay agreed. Now tested it with 12 observations each and this time the result is even more close!
Elite jagged trail, best place to find a solo mob to test something on.
First try:360 damage 775 armour.
Screen shots:



2nd try :355 dmg 2005 armour.
Screen shots.


These were the screen shots.
Mathematical analysis.
Damage done by 775armour and 360damage
282
282
299
326
320
281
293
283
326
277
278
299
overall damage done= 3556
Average damage done= 296.33

Damage done by 2005 armour and 355 damage
305
278
299
311
289
286
300
314
276
287
296
278
Total damage done = 3518
Average damage done= 293.16

So, just 3 average damage difference is not a big deal at all. Now I'm pretty sure, the damage done is how much damage you have instead of what gear or armour value you have. If, I am again wrong correct me.
Gave you proof again. Let's wait for devs reply.

Sent from my Gpad_G1 using Tapatalk

[Serancha]

Any input provided can be helpful. In the interest of keeping things from turning into a contest or battle of data, I will leave any further info for the developers to handle when they investigate this issue.

Thank you all for participating in this very interesting conversation, and thank you to STS for your continued excellence in all matters technical.

[Crowsfoot]

I don't understand the "Battle of data" portion. Although he used small sample size and didnt account for error, Ish's data should show a much larger difference in damage output than shown based on the OP's claim. His screen shots cover any disputes over him fudging the data with buffs. I have no plans to do my testing unless someone has a hard issue with his.

Thanks for saving me time Ish!

[Crowsfoot]

crowsfeet why so mad? You act as if Sera insulted you directly. The devs noticed/ commented and are investigating, therefore Sera was successful in her mission. You mad?

I was frustrated with myself for losing my already recorded results (along with hundred of notes on things more important to me than AL which I am very concerned with the location of). Beyond that, within 5 minutes of posting the data I worked very hard to gather and translate into something useful, I was shot down as "misunderstanding the point" event though I noted that testing was only half way (if that) complete.

I was also annoyed that my critique of time being the dependent variable was entirely ignored (still hasn't been adressed) instead of disputed while my hard numbers "missed the point." I hit some breaks and chilled out. All good here.

Edit: it was some what addressed. However, testing with 30+ people over multiple months instead of a controlled group under defined circumstances just exemplifies why I think time is an unfit dependent variable for testing a causation between armor and damage output et prevae vice versa.

[Ardbeg]

I don't understand the "Battle of data" portion. Although he used small sample size and didnt account for error, Ish's data should show a much larger difference in damage output than shown based on the OP's claim. His screen shots cover any disputes over him fudging the data with buffs. I have no plans to do my testing unless someone has a hard issue with his.

Thanks for saving me time Ish!

but but but shouldn t this have been tested in a party, according to seranchas original hypothesis/observation? how are these results related to her findings and statements? she observed and investigated the scaling for *parties*. timed runs may or may not be the best tool, but the provided data from ish are from a *single player* test.
i m with serancha here, just let the data flood end and the devs have look at it!

[Ishtmeet]

I don't understand the "Battle of data" portion. Although he used small sample size and didnt account for error, Ish's data should show a much larger difference in damage output than shown based on the OP's claim. His screen shots cover any disputes over him fudging the data with buffs. I have no plans to do my testing unless someone has a hard issue with his.

Thanks for saving me time Ish!

Np at all. I used a non-buff build. Used juggernaut sometimes, because that mob was hitting me hard without armour. Juggernaut increases the hp, not str! So no extra damage was there and I even had damage elix all the time for keeping things fair.

Sent from my Gpad_G1 using Tapatalk

[Crowsfoot]

but but but shouldn t this have been tested in a party, according to seranchas original hypothesis/observation? how are these results related to her findings and statements? she observed and investigated the scaling for *parties*. timed runs may or may not be the best tool, but the provided data from ish are from a *single player* test.
i m with serancha here, just let the data flood end and the devs have look at it!

What figures has she provided?

Faster, better, less. There are no numbers.

Am I suppose to fill a party of four and a party of three and repeat my testing? Fine, party me up. There should be no difference whether the entire party has higher or lower armor. Give me one variable to test.

Speed is an impractical dependent variable. Beyond that, not a single time has been provided. I will take Ish' over hers simply because his is defined, uses a distinct methid, and is replicable.

Sera's is a deja vu of Nikola Tesla's earthquake machine. We here there was a test and that the results are huge, spectacular, earth shattering. But, we can't test it, observe the testing, or have any figures from it.

PS: his "earthquake machine" was finally rebuilt by his specifications with less than spectacular results. It was likely a publicity stunt by him to gain funds for his floundering wireless electrical transmission studies (I am not saying that Serancha was doing a publicity stunt. The only point in cluding this is to show that an experiment without hard numbers is rarely accurate).

[Bigboyblue]

Did you re enter the map with your different gear sets Ish? I believe that mobs scale to whatever stats your toon has upon entrance.

Also, Sera has stated that she isn't sure what causes runs to take longer. If damage remains constant it is possible that hp increases on mobs. This will slow down your run the same as it would with the same hp and less damage.

Very interesting thread. I always felt in my mind that running with mythic gear never really helped me. Legendary gear worked just as well.

[Milan Lame Man]

Ok so I did some testing on my own, sadly it's somewhat hard to find a partner for tests, so most tests are solo only.
ENEMY HEALTH

toon	elite jarl's mob	elite jarl	reg overgrowth
half myth rogue (1363 armor+lepre)	9003-9006		27-30k
legend rogue (1095 armor+lepre)	8840-9071	33364-33597	27-29k
half myth sorc (1022 armor+lepre)	8871-9157	32777-34000
legend sorc (all gear lv36, 856 armor)	8812-9096	33227-33500	~27k

Conclusion: enemy health probably does not scale with gear.
Note: I have measured Overgrowth with a party of 2, resulting in the same health, but the test may be invalid because the other player joined late and Overgrowth does not seem to reset properly.

ENEMY DAMAGE

toon	reg Tindirin archer	dmg/(1-armor/41)
half myth rogue (armor 1363)	267-400	400-600
legend rogue (armor 1095)	284-438	387-597
myth sorc+lepre (armor 1022)	291-447	387-595
legend sorc, no lep (armor 856)	306-475	387-600
2 myth rogues (armor 1363)	291-439	436-659

Conclusions:
- Enemy damage does not scale with armor rarity.
- Enemy damage is consistent with damage_dealt=damage_base*(1-armor/level) for maps that scale
- Enemy damage slightly increases for a party of 2 over solo.

About testing:
- Mobs and jarl's health was measured twice by adding flying numbers, without pet.
- Overgrowth was only measured once with each toon, with added estimate of damage dealt by pet.
- Mobs damage is measured using min-max as it seems to give more consistent results than average (thanks go to Fluff for inventing this method).
- When testing legendary gear, mythics were moved to another toon on the same account.
Thanks to Serancha and Amazonina for help.


Disclaimer:
However hard I tried, there may be errors.

The obvious question is, why would the run times be the same? I agree with Madnex and Rare: If you don't die, health+armor does not matter for run's time. The important stat is damage. Since damage is mostly dependent on the weapon, the difference in time caused by armor may be low.
For the Hauntlet example, maybe as the warrior switched to legendary, he got the aggro 100% time, which leads to better mob clustering & less time spent potting by DPS.

And it would not be me without a fun story, right.
During testing, I ran Overgrowth with legendaries first, and got killed by mobs quite a bit.
So I got the myths, swept through the map like a boss... and then realized that I unstashed the mythics but didn't equip them.
Which makes me believe that psychology and many other things have a big impact on run times.

What should be tested next (by someone else :-/)
- Measure dependence of mob's armor (real damage dealt vs. damage listed on Inventory page) on toon armor
- Measure crit+dodge rate for both attack and defence
- Measure mob health and damage with party of 2, 3, 4
- Repeat tests with warrior
- Find what makes mobs to focus on a specific toon. I ran Mount Spyr with a friend (rogue+sorcerer) and depending on the pets we equipped, all mobs went after me or her.

[Ishtmeet]

Did you re enter the map with your different gear sets Ish? I believe that mobs scale to whatever stats your toon has upon entrance.

Also, Sera has stated that she isn't sure what causes runs to take longer. If damage remains constant it is possible that hp increases on mobs. This will slow down your run the same as it would with the same hp and less damage.

Very interesting thread. I always felt in my mind that running with mythic gear never really helped me. Legendary gear worked just as well.

Why would I? All pics were took in one run.

Sent from my Gpad_G1 using Tapatalk

[Waheedski]

I really hope high armour is better because im geared that way. I personally would've thought high armour is better and that mobs only become harder depending on your level not your armour. If Serancha has given data to suggest otherwise and at least make the devs look into it then its all good. I can see what Serancha is trying to say by stating high armour creates stronger mobs and thus makes it harder for the party especially the squishes and intensifies the battle.

Both Crow and Ish have posted opposing views with data also, I hope devs look into this and give feedback immediately on the issue.

Out of curiosity - arent runs slow anyway with a tank in party ? I dont mind a slow run if deaths are none or very minimal I must be the only one that doesn't care about speed runs LOL, you should see my crew in elites we're like the keystone cops

[Ishtmeet]

Out of curiosity - arent runs slow anyway with a tank in party?

This!^^^
Ofc when you run with a tank in pt, the runs will be slow, and you would have less deaths and use less potions.





Sent from my Gpad_G1 using Tapatalk

[Raregem]

What figures has she provided?

Faster, better, less. There are no numbers.

Am I suppose to fill a party of four and a party of three and repeat my testing? Fine, party me up. There should be no difference whether the entire party has higher or lower armor. Give me one variable to test.

Speed is an impractical dependent variable. Beyond that, not a single time has been provided. I will take Ish' over hers simply because his is defined, uses a distinct methid, and is replicable.

Sera's is a deja vu of Nikola Tesla's earthquake machine. We here there was a test and that the results are huge, spectacular, earth shattering. But, we can't test it, observe the testing, or have any figures from it.

PS: his "earthquake machine" was finally rebuilt by his specifications with less than spectacular results. It was likely a publicity stunt by him to gain funds for his floundering wireless electrical transmission studies (I am not saying that Serancha was doing a publicity stunt. The only point in cluding this is to show that an experiment without hard numbers is rarely accurate).

Post two does provide numbers and time.

[Ardbeg]

What figures has she provided?

Faster, better, less. There are no numbers.

Am I suppose to fill a party of four and a party of three and repeat my testing? Fine, party me up. There should be no difference whether the entire party has higher or lower armor. Give me one variable to test.

Speed is an impractical dependent variable. Beyond that, not a single time has been provided. I will take Ish' over hers simply because his is defined, uses a distinct methid, and is replicable.

Sera's is a deja vu of Nikola Tesla's earthquake machine. We here there was a test and that the results are huge, spectacular, earth shattering. But, we can't test it, observe the testing, or have any figures from it.

PS: his "earthquake machine" was finally rebuilt by his specifications with less than spectacular results. It was likely a publicity stunt by him to gain funds for his floundering wireless electrical transmission studies (I am not saying that Serancha was doing a publicity stunt. The only point in cluding this is to show that an experiment without hard numbers is rarely accurate).

you can t verify or falsify a observation made about the levelling for a party by inserting data for a single toon.
it s answering a question no one asked. if the time isn t a valid indicator for you, that something might be not in balance with the leveling of elite runs for *this obviously can t be stressed enough party runs*, it s a good idea to come up with a party test setup. or let the devs look at the code. i don t see how inserting single player test data helps to explain or debunk seranchas observations.

for mored tesla humor, please look here :

http://theoatmeal.com/comics/tesla

[Crowsfoot]

Did you re enter the map with your different gear sets Ish? I believe that mobs scale to whatever stats your toon has upon entrance.

Also, Sera has stated that she isn't sure what causes runs to take longer. If damage remains constant it is possible that hp increases on mobs. This will slow down your run the same as it would with the same hp and less damage.

Very interesting thread. I always felt in my mind that running with mythic gear never really helped me. Legendary gear worked just as well.

Okay, let's stop playing shoot the fish in the barrel and define what the issue is.

Do you want to test damage output, damage reduction, mob HP, what? Pick a variable. Time simply isn't a reliant dependent variable.

Also decide whether you want tests done solo or in a party.

Furthermore, do you want us to reenter the map for each test? I didn't, it doesn't make sense to since mobs scale real time. We should all remember mobs becoming harder after a high level person enters the map. Or we can leave it up to Dev's right when an indisputable claim against the damage output claim was made.

PS: if "Sera has stated that she isn't sure what causes the runs to take longer," why on Earth would she make a thread on it? This thread makes it very clear that their are ways to test any hypothesis one may think of. There was ample time to test these as well since this testing had a 30+ man team working on it for months.

PPS: Milan tested the HP theory.

[Mihu]

I think this might be true, my warrior friend says that his elite runs take longer with maul which makes no sense but just happens. Also in km3 i can do 1min30sec runs easily with skilled legendary players but when i hav full myth team it usually takes almost 2mins.

[Ishtmeet]

Also in km3 i can do 1min30sec runs easily with skilled legendary players but when i hav full myth team it usually takes almost 2mins.

I would love to see this.



Sent from my Gpad_G1 using Tapatalk

[Mihu]

I was also suprised and thought why and now i might know the reason i see why u wars hate u love stacking that armor