Enemy Camp Destruction Times

Fist of all, thank you for this post.
A lot of it has come as quite a revelation to me.

Quote:

Originally Posted by Lesson

The amount of damage a camp can sustain before destruction is as follows (derived emperically):

Code:

Building Damage Bandit camp 250 Watchtower 1000 Reinforced T. 1500 Stone tower 2000 White castle 2000 Witch tower 2000 Bone church 2000 Dark castle 2000

Now, are these numbers mostly a result of your personal testing or do they come from the combat sim?

Really great work guys. Thanks from us dummys.

Quote:

Originally Posted by Lamosos

Fist of all, thank you for this post.
A lot of it has come as quite a revelation to me.

Now, are these numbers mostly a result of your personal testing or do they come from the combat sim?

I verified timing on quite a few combinations personally, after that it was mostly sampling sim checks. And there could be other factors involved behind the scenes that cannot really be observed, but I think this should be pretty close. And I only personally verified results against regular camps and Witch Towers, Dark Castles and White Castles. I did not check the numbers versus the Watchtowers and Reinforced.

I am very confident that the unit type only needs to be 'represented' to contribute to the building destruction, and that Cannons deal double their Damage.

Asipak's numbers also seem to assume Archers (Bowmen, Longbowman and Crossbowman) get an accuracy bonus for attacking a Tower (I had always assumed the bonus only applied to the units within the tower) so that they would always hit for their max damage. I have NOT verified if this is correct... so if you are seeing longer actual destruction times when using Archers then Asipak's Sim is indicating, this might be the reason.

Definately test the different scenerios yourself, but you should have a good starting point now.

seams the longest battles are against archer tower which are wiped out buy calv. in the first round, go figure

Yea, no doubt.
As far as Asipak is concerned, it seems like he's following a similar concept to what you have described.
But his numbers are really hard to judge, since we haven't seen the actual math behind them.
Although I can say with confidence that those numbers are off, I have absolutely no idea where the errors are.
On a side note, I find it extremely strange that BB wouldn't just release this information.
What are they afraid of, that we block too many camps?
Who cares if its an undocumented feature?
Who cares if it didn't exactly turn out the way it was planned?
It's a good thing, it makes the game more interesting, so suck it up and embrace it!

Quote:

Originally Posted by xethilus

seams the longest battles are against archer tower which are wiped out buy calv. in the first round, go figure

Well, like its been said a few posts above, towers have more HP than regular camps, and cav does the least amount of damage of all the troops, so do the math :)

Quote:

Originally Posted by Lesson

Since i was asked about this recently, and no one else has taken the time to explain how the camp destruction works, I thought I would take a stab at it.

This description was derivied from empirical data I gathered in actual battles and many iterations of Asipak's Sim.

The destruction of a camp is the additional amount of time needed to destroy a camp after you have defeated the enemy that occupied it. It is in addition to the actual battle where you defeated the camp, and is represented by the 'Health Bar' that appears when the inital winning battle is over. The amount of time needed to destroy the camp is calculated the same way as the Battle itself: 20 seconds per round (10 seconds for fast generals like Vet, Grim and BHG). Side Note: This addtional camp destruction time is why it normally is better to have to Winning Advanced Block versus a losing one; Winning in 5 rounds will last just as long as losing in 7 rounds assuming you are not doing something silly like blocking with Cannons.

The number of Rounds it takes to destroy a camp is determined in much the same way as the battle was calculated and is dependent on the camp type and the unit type(s) originally sent to the battle.

The amount of damage a camp can sustain before destruction is as follows (derived emperically):

Code:

Building Damage Bandit camp 250 Watchtower 1000 Reinforced T. 1500 Stone tower 2000 White castle 2000 Witch tower 2000 Bone church 2000 Dark castle 2000

Now comes the important part: The amount of damage you can inflict per round is entirely dependent on the unit types represented in the actual battle. They do not have to be alive at the end of the battle to help with the destruction of the building, they just had to be a part of the garrison assignment at the start of the battle. It also does not matter how many are included to represent; as far as the building destruction is concerned, sending 1 cannon is the same as sending 100. You'll see a few examples of this tactic used in some of my guides (mostly Roaring Bull), where I have a mix of unit types being required to take the towers down faster.

Lesson Guides

So as an example:

Suppose you win a battle using your Vet general against some camp, and you had sent 100R and 150K(Cannons). Each round, you would hit between 255 and 390 in damage based on the following Unit Types that are represented:

Recruit (15-30)
Cannon (120-240) * remember cannons get double the effective damage against buildings
General (120)

So in this case a regular Camp is only going to last 1 round and a Dark Castle or Witch Tower 6-7 rounds depending on your accuracy rolls.

Why is all this important?

If, in this same example you were to replace 3 of those recruits (or cannons) with a Militia, a Soldier and an Elite, suddenly you are hitting for 60-120 additonal damage each round on the building. Now that Witch Tower will only last 4-5 Rounds.. you just shaved 20 seconds off your potential destruction time so that Advanced Block may work after all!

-Lesson

Nothing to add here. It's completely correct.

Quote:

Asipak's numbers also seem to assume Archers (Bowmen, Longbowman and Crossbowman) get an accuracy bonus for attacking a Tower (I had always assumed the bonus only applied to the units within the tower) so that they would always hit for their max damage. I have NOT verified if this is correct...

That isn't correct. Why do you think so?

Quote:

Originally Posted by Asipak

Nothing to add here. It's completely correct.

That isn't correct. Why do you think so?

I think this has to do with something he and I were discussing yesterday. We were discussing the upper right dark castle on BK and a combination he had that did not seem to calc out the same for destruction time as the sim.

The combination was 1r 22m 1s 1c 110b 1lb 114a

Link for sim results:
http://settlersonlinesimulator.com/d...=15#sim-result

The sim shows 5-6 rounds for distruction but if you manually add the hit points for all the troop types per round as follows:

recruit = 15 - 30 hit
militia = 20 - 40 hit
soldier = 20 - 40 hit
calv = 5 - 10 hit
bow = 20 - 40 hit
lbow = 30 - 60 hit
xbow = 45 - 90 hit
gen = 120 hit

You get - Hit point range per round = 275 - 430

If a Dark Castle takes 2000 hit points for destruction then

Min Destruction Rounds = 2000 / 430 hit per round = 4.65 = 5 rnds (which sim has)

Max Destruction Rounds = 2000 / 275 hit per round = 7.27 = 8 rnds (sim shows 6 instead of 8)

I'm hoping my above info of 2k and hit point valuations is correct but I believe the question here is the sim showing range of 5-6 destruction rounds rather than 5-8 rnds etc.

So do the Bow, Lbow and Xbow minimum hit points get a bonus each round against castles etc? If so, and the mins doubled, that would add 95 hit points to each round causing the max number of destruction rounds to be 6 as the sim states.

I think that is the question and point because I believe he saw the destruction of the Dark Castle take longer than the max of 6 rounds that the sim shows.

So the bottom line is either my info somewhere is flawed or it's something else.

Quote:

Originally Posted by Asipak

Nothing to add here. It's completely correct.

That isn't correct. Why do you think so?

Because it was a possible explaination as to why your min-max calculations were not coinciding with mine when I was trying figure out why an example I had with Archers in the mix did not seem to jive with yours. But since then I started looking at other examples and your results are not matching with what I was expecting, so I am now assuming some of my basic assumptions must be incorrect.

For example: If I calculate a Tower destruction with a Garrison of all soldiers, with the assumptions I made, I have to following:

Each round I would do the following in dmg to the building:

Soldier type (20-40)
General (120)

So I am looking at 140-160 in damage each round.

Your sim indicates a range of 13-14 rounds to destroy the tower.

However, if I fail on my accuracy rolls everytime, I only get 14 * 140 damage for a total of 1960 in damage which would mean it would take another round to take out the tower IF 2000 is the correct amount of Health and my expected damage per round is correct.

So my question is: for your sim, what is your expected damage per unit type? What is the amount of Health you are using for the tower?

I certainly didn't do enough self tests to be absolutely confident in my assumptions, so I would love to know what numbers your are using for your sim, as well how you determined them.

Please calc the average damage of your units.

Code:

2000 / 275

implies that you will have 8 rounds with the worst case. That's very unlikely.

[edit]

Okay, you've edited your posting, but the meaning is the same. You've anticipated the worst case.

Ahhh... sorry, it's another posting.

Quote:

What is the amount of Health you are using for the tower?

Your values are correct.

I'm off now. I'll answer later or tomorrow.

Quote:

Originally Posted by Asipak

Please calc the average damage of your units.

Code:

2000 / 275

implies that you will have 8 rounds with the worst case. That's very unlikely.

[edit]

Okay, you've edited your posting, but the meaning is the same. You've anticipated the worst case.

Ahhh... sorry, it's another posting.

Your values are correct.

I'm off now. I'll answer later or tomorrow.

Oh! Well that would explain it if the range in rounds you display in your sim is the expected average number of rounds and not the possible Minimum and Maximum. Good to know.

Question then: What is your threshold for determining if the possible Min/Max should be included or not?

Quote:

uestion then: What is your threshold for determining if the possible Min/Max should be included or not?

10000 repetitions of the battle against the building! If you don't find a round in the result, that would have been part of the worst-case, its chance is lower than 1/10000*100 = 0.01%. That's good enough, I think.

Quote:

Originally Posted by barenski

this is fun and all but enemy camp destruction times do not vary from attempt to attempt. its always 10 sec per round + 10 sec bar, double time for tower and castle. that is my experience with the game. always 30 sec for 2 round leader

Nice Troll attempt, but I'm not biting. Maybe somebody else will though.

Quote:

Originally Posted by barenski

well, i m lvl49 and done my share of adventures and im yet to see a different kill time for same leader with same troops attakin. please provide specific example of leader and attacking formation that yeiled different kill times in different attenmps, given they were same number of rounds. im aware that some recruits setups allow for quite a spread in number of rounds, so plz give example where number of rounds is same

Really, mate?
You have 5 different people telling you that this is how it works, but you still need to see a specific example?

Quote:

this is fun and all but enemy camp destruction times do not vary from attempt to attempt. its always 10 sec per round + 10 sec bar, double time for tower and castle. that is my experience with the game. always 30 sec for 2 round leader

Nothing to add here! It's completely wrong!

Example: http://www.youtube.com/watch?v=XZ3rQD-ad5U

Time: 240 seconds to destroy the building!

Enjoy!

Quote:

but why would you care predicting destruction time of stuff unless you want to blok on it

That's not the question!

Quote:

all im sayin that i never had any difference in camp desruction time

Because you didn't pay attention to it. You need special army settings to defeat a normal camp in 1 round (=10s normal general), so in most cases you'll always observe the same time of 20s (normal general).

Quote:

i think the difference that you're observing is caused by server lag

No. I've got many of those videos to proove that lags aren't responsible for these times. But i won't upload them just because 1 guy isn't believing me. ;)

Quote:

formula 10sec per round + 10 sec victory bar, double that for dark castle, works 100% of attempts for me for me

Wrong!

Quote:

plus this game is really easy

No argument!

Quote:

usually bloks go way over the leader battle and the bloks that cut short like in db no one ever does, when i tell people that i blok every other leader in db, they dont believe me, i think

Hmm, I don't get the point here.

Quote:

anyways, good luck with your guides i hope it helps you get lucky or something

Thanks. ;)

Regards

Indeed, why bother trying to understand something that has somehow been working thus far.
What if you jinx it with your stupid math, tables, and other mambo-jumbo, and it suddenly stops working?
What then??
WILL WE ALL HAVE TO LEARN TO COUNT?!!
SUCH BLASPHEMY!!!

Quote:

Originally Posted by barenski

victory.jpg

At this point I find myself with nothing left to say on the matter other than to quote an earlier post from this very thread.

http://cdn.gunaxin.com/wp-content/up...a-costume1.jpg

Quote:

lets just aks bb if its 10 sec per round + 10 sec victory bar

Well, BB just told us, on the german forum, that it's 20s per round.

I'm out of this discussion. It is all said!

Regards

I just had a very strange experience in SFTR.
It was a 10 round white castle fight that was supposed to take 200-220 seconds according to this simulation.
It took 120 seconds, including the destruction.
The health bar resembled a regular bandit camp with the speed at which it was diminishing.
I wish I was recording at the time.
Any ideas on what could've happened there?

Hmm, no, I have no idea. Perhaps it was a lag. I really like to see a recording of the time. In general, without lags, the visualization of the time in the game is very exact.

I'm not sure about the hit points of this building. Maybe it's only 250 HP. Extract of the game settings file:

Code:

<Building name="tent_traitors_leader" hitPoints="250" constructWithoutSettler="true" maxUnits="25" isWarehouse="true" > <BuildingUpgradeBonuses></BuildingUpgradeBonuses> <Blocks> <Block x="0" y="0" blockingType="1"/> <Block x="0" y="-1" blockingType="2"/> <Block x="0.5" y="-0.5" blockingType="2"/> <Block x="1" y="0" blockingType="2"/> <Block x="-0.5" y="-0.5" blockingType="2"/> <Block x="0.5" y="0.5" blockingType="2"/> <Block x="-1" y="0" blockingType="2"/> <Block x="-0.5" y="0.5" blockingType="2"/> <Block x="0" y="1" blockingType="2"/> </Blocks> </Building> [...] <Building name="Castle" hitPoints="2000"> <Costs> <Cost name="Marble" count="150"/> <Cost name="RealPlank" count="150"/> </Costs> <Blocks> <Block y="1.5" x="0.5" blockingType="1"/> <Block y="1" x="1" blockingType="1"/> <Block y="0.5" x="0.5" blockingType="1"/> <Block y="0" x="0" blockingType="1"/> <Block y="0.5" x="-0.5" blockingType="1"/> <Block y="1" x="-1" blockingType="1"/> <Block y="1.5" x="-0.5" blockingType="1"/> <Block y="1" x="0" blockingType="1"/> <Block y="2" x="0" blockingType="1"/> <Block y="2.5" x="0.5" blockingType="2"/> <Block y="2.5" x="-0.5" blockingType="2"/> <Block y="3" x="0" blockingType="2"/> <Block y="2" x="-1" blockingType="2"/> <Block y="1.5" x="-1.5" blockingType="2"/> <Block y="1" x="-2" blockingType="2"/> <Block y="0.5" x="-1.5" blockingType="2"/> <Block y="0" x="-1" blockingType="2"/> <Block y="-0.5" x="-0.5" blockingType="2"/> <Block y="-1" x="0" blockingType="2"/> <Block y="-0.5" x="0.5" blockingType="2"/> <Block y="0" x="1" blockingType="2"/> <Block y="0.5" x="1.5" blockingType="2"/> <Block y="1" x="2" blockingType="2"/> <Block y="1.5" x="1.5" blockingType="2"/> <Block y="2" x="1" blockingType="2"/> </Blocks> </Building>

I'm not sure if the building named „Castle“ is the white castle. As the resource names are:

Code:

<ResourceDefinition id="0" name="Population" MaxLimit="50">  <Creation CreationId="0" type="CreatedAlways" CreationTime="900"/> <ExpandMaxLimit ByBuilding="SimpleResidence" Amount="10"/> <ExpandMaxLimit ByBuilding="NobleResidence" Amount="30"/> <ExpandMaxLimit ByBuilding="tent_traitors_leader_Deco" Amount="50"/> <ExpandMaxLimit ByBuilding="WitchTower_Deco" Amount="100"/> <ExpandMaxLimit ByBuilding="Dark_Castle_01_Deco" Amount="100"/> </ResourceDefinition>

, it could be possible that the white castle is „tent_traitors_leader“. And this building only has got 250 HP.

It would be great if you can do some more attacks on this building to check out its real hit points.

I'll add a hint to this building in the simulator that the values aren't confirmed.

Regards

[edit]

Hmm, yes, I think it's only 250 Hp.

Code:

<Building name="tent_traitors_leader" hitPoints="250" constructWithoutSettler="true" maxUnits="25" isWarehouse="true" > <BuildingUpgradeBonuses></BuildingUpgradeBonuses> <Blocks> <Block x="0" y="0" blockingType="1"/> <Block x="0" y="-1" blockingType="2"/> <Block x="0.5" y="-0.5" blockingType="2"/> <Block x="1" y="0" blockingType="2"/> <Block x="-0.5" y="-0.5" blockingType="2"/> <Block x="0.5" y="0.5" blockingType="2"/> <Block x="-1" y="0" blockingType="2"/> <Block x="-0.5" y="0.5" blockingType="2"/> <Block x="0" y="1" blockingType="2"/> </Blocks> </Building> <Building name="tent_traitors_leader_Deco" hitPoints="250" constructionDuration="90" subType="Deco" > <BuildingUpgradeBonuses> <Buff level="0" tradable="false" buffType="Upgrade" /> <Buff level="1" tradable="false" buffType="Upgrade" /> </BuildingUpgradeBonuses> <Blocks> <Block x="0" y="0" blockingType="1"/> <Block x="0" y="-1" blockingType="2"/> <Block x="0.5" y="-0.5" blockingType="2"/> <Block x="1" y="0" blockingType="2"/> <Block x="-0.5" y="-0.5" blockingType="2"/> <Block x="0.5" y="0.5" blockingType="2"/> <Block x="-1" y="0" blockingType="2"/> <Block x="-0.5" y="0.5" blockingType="2"/> <Block x="0" y="1" blockingType="2"/> </Blocks> </Building>

I'll change it in the simulator. Thanks for this hint.

Hey Asipak,

Last couple of days, that skull that you click to get to the camp combos and such has disappeared, anything happen to the ref image file?

Also, just like to point out, in your map popup, in the upper right section you have a camp showing as 120nomads and 80 composite bow but i believe it's 120 nomads and 80 Lance Riders, since those types and numbers are put into the ride side for sims, someone who isn't double checking or paying attention may mess up a bit.

Quote:

Hey Asipak,

Last couple of days, that skull that you click to get to the camp combos and such has disappeared, anything happen to the ref image file?

It should be a cache problem. Please delete your browser cache or press [F5].

Quote:

Also, just like to point out, in your map popup, in the upper right section you have a camp showing as 120nomads and 80 composite bow but i believe it's 120 nomads and 80 Lance Riders, since those types and numbers are put into the ride side for sims, someone who isn't double checking or paying attention may mess up a bit.

I'll check it.

Quote:

Originally Posted by Asipak

It should be a cache problem. Please delete your browser cache or press [F5].

I'll check it.

Thanks Asipak - Yep clearing browser cache brought it back ok.

Hi Asipak,

Thought I'd link the following issue here.

I was examining the difference in the destruction of the middle witch tower in RB with an xbow combination and a cannon combination. According to the sim, the combination of 40R 120B 1LB 1XB 88K shows a one round battle with total destruction from battle start as:

Witch tower 4-5 80-100 40-50 100-120 50-60 10000

and indeed it turned out to be 50 seconds from start to finish.

I then ran the same battle against the same tower with an xbow combination of 38R 1M 1S 120B 1LB 89XB and it also was a one round battle (as I verified from the Battle Report replay) where the sim showed:

Witch tower 5-6 100-120 50-60 120-140 60-70 10000

However the actual total destruction time that took place was 80 seconds instead of the 60-70 secs expected (as you can see in the below video capture), so I"m wondering what is happening here in terms of it taking an apparent 10secs or extra round for the xbow version. Any ideas?

I'm providing links to video capture of both the cannon and the xbow battles. By the way, the last attacking general is the Vet hitting the tower, with the tower battle starting around the 2:18 mark for the xbow video.

Cannon Video Link: youtube.com/watch?v=CYMWW_-QsaM

Xbow Video Link: youtube.com/watch?v=EI7jIQXyqvw

Thanks for your thoughts in advance.

Quote:

Originally Posted by barenski

so if i had all xbow and just 1 cannon, will i still benefit?

Short answer: yes.
Long answer: read the thread, it's all been explained.

Hmm. That's strange. I've repeated the battle against the building another 10000 times, but there is not a 7th round.

You're right, the battle against the tower takes 70s but I have no idea why. Perhaps it's because of some dead unit types. But I have a video to proof that also the dead types have to take part of the battle:

Simulation

Video: http://www.youtube.com/watch?v=TMGjh...layer_embedded

Time: 50s

If I simulate a battle with cannoneers only, the battle takes 6-7 rounds.

Strange. Do you guys have an idea?

Regards

[edit]

Calculation (estimated; well below the average):

Code:

HP | G R M S B LB A 2000 / (120 + 22.5 + 30 + 30 + 30 + 45 + 67.5) = 5.797 (6 rounds)

This would be the case, if the hit percentage of the units was 50%. We all know, it's much better, so the calculation isn't wrong. There must be another reason.

Quote:

Originally Posted by Asipak

Hmm. That's strange. I've repeated the battle against the building another 10000 times, but there is not a 7th round.

You're right, the battle against the tower takes 70s but I have no idea why. Perhaps it's because of some dead unit types. But I have a video to proof that also the dead types have to take part of the battle:

Simulation

Video: http://www.youtube.com/watch?v=TMGjh...layer_embedded

Time: 50s

If I simulate a battle with cannoneers only, the battle takes 6-7 rounds.

Strange. Do you guys have an idea?

Regards

[edit]

Calculation (estimated; well below the average):

Code:

HP | G R M S B LB A 2000 / (120 + 22.5 + 30 + 30 + 30 + 45 + 67.5) = 5.797 (6 rounds)

This would be the case, if the hit percentage of the units was 50%. We all know, it's much better, so the calculation isn't wrong. There must be another reason.

I know there has to be a reason in there somewhere. I think that 40 you have above is 45 for median btw but that would bring the 50% average rnds only down to 5.79 (6 rnds) still.

If all minimums hit which i know is not realistic then it would be 2000/270 = 7.407 rnds destruction (8rnds) worst case. But as you mentioned above in 10,000 iterations you never saw even 7 rnds.

I did run this combo on the left tower 143R 1M 1S 103E 1B 1LB (2 Rnds + 6-7 Rnds (destruction) = 80-90 Seconds) and it did come out to a total of 80secs for completion so that matches your sim.

The only diff I see between this and the other combo is that in this i had elites and no xbow, in the other combo i had xbow and no elites, and plus the fact that the left tower was 2 rnds whereas the middle battle was 1 rnd.

I'm baffled at the moment but just for clarity I am providing the below link which is the Battle Report replay from the attack on the middle tower again, showing indeed 1 round which does mean that somehow somewhere the destruction part is 70 secs or 7 rnds since the total verified length was 80secs.

youtube.com/watch?v=rJZphH2NqhY

Quote:

Originally Posted by Asipak

If I simulate a battle with cannoneers only, the battle takes 6-7 rounds.

Strange. Do you guys have an idea?

Hey, you know what could help?
Giving us the exact code used to calculate it.
Just an idea :)

Asipak,

Yes we, (Lesson, Phenix0cs and I) have been researching it and I believe Lesson has come up with proof and explanations for the difference(s) seen. We'll be posting soon, so stay tuned.

Asipak,

Sanii, Phenix0cs and I have been compiling some data over the last couple days and I think we can now safely say that we have found the issue.

The issue is that within the game itself, Dead units are not participating in the Destruction time. Only unit types with a surviving member at the end of the last round get to participate in Destruction portion of the fight. It would make sense that the dev’s would attempt to echo a scenario that takes place in real life such as this.

I too had been under the belief that dead units would participate, so either something had changed semi- recently or we did not as thoroughly test that assumption in the past.

The best example we have of this is in this battle depicted in Sanii’s example here: Video

In this example, every unit type is represented (114r 1m 1s 129e 1c 1b 1lb 1x 1k), so according to the simulation, we have an expected Destruction time of 3-4 Rounds. However, every unit type except for the Elites are Dead before the end of the final round, and so we have a Destruction time of 13 rounds, (which matches your sim of 13-14 rounds if it is configured for only the Elite type), rather than the prior much shorter estimate that utilized all the original troop types for destruction.

We ran many different trials with different combinations, (we have them all in a spreadsheet), not just this one, where we varied it so that we had units dying in different rounds You’ll notice in the example above, the Cannon, which would cause the biggest difference in destruction time, didn’t die until the 2nd (last) round (see replay of battle at end of video), yet he still did not get to participate in the camp destruction.

I rewatched your Proof video in post #67. You don’t show the units alive after the battle is over, so perhaps an Elite or even a Recruit + Elite was also still alive at the end of the battle? When I ran a simulator on the configuration you used, I only saw losing both elites about 7% of the time.

Interested in your thoughts, and if you have a chance to verify with your own trials. Please contact us at your earliest convenience!

Best Regards,

Lesson, Sanii and Phenix0cs

P.S. From Sanii - Personally, I am convince based upon the results that the game itself does imitate real life action whether it was added later by dev’s or not. I believe if you reflect this “dead units tell no tales nor destroy buildings” concept, your simulator will be even more utilized by people as the best and as we spread the tale.

Quote:

Originally Posted by Lesson

Asipak,

Sanii, Phenix0cs and I have been compiling some data over the last couple days and I think we can now safely say that we have found the issue.

The issue is that within the game itself, Dead units are not participating in the Destruction time. Only unit types with a surviving member at the end of the last round get to participate in Destruction portion of the fight. It would make sense that the dev’s would attempt to echo a scenario that takes place in real life such as this.

I too had been under the belief that dead units would participate, so either something had changed semi- recently or we did not as thoroughly test that assumption in the past.

The best example we have of this is in this battle depicted in Sanii’s example here: Video

In this example, every unit type is represented (114r 1m 1s 129e 1c 1b 1lb 1x 1k), so according to the simulation, we have an expected Destruction time of 3-4 Rounds. However, every unit type except for the Elites are Dead before the end of the final round, and so we have a Destruction time of 13 rounds, (which matches your sim of 13-14 rounds if it is configured for only the Elite type), rather than the prior much shorter estimate that utilized all the original troop types for destruction.

We ran many different trials with different combinations, (we have them all in a spreadsheet), not just this one, where we varied it so that we had units dying in different rounds You’ll notice in the example above, the Cannon, which would cause the biggest difference in destruction time, didn’t die until the 2nd (last) round (see replay of battle at end of video), yet he still did not get to participate in the camp destruction.

I rewatched your Proof video in post #67. You don’t show the units alive after the battle is over, so perhaps an Elite or even a Recruit + Elite was also still alive at the end of the battle? When I ran a simulator on the configuration you used, I only saw losing both elites about 7% of the time.

Interested in your thoughts, and if you have a chance to verify with your own trials. Please contact us at your earliest convenience!

Best Regards,

Lesson, Sanii and Phenix0cs

P.S. From Sanii - Personally, I am convince based upon the results that the game itself does imitate real life action whether it was added later by dev’s or not. I believe if you reflect this “dead units tell no tales nor destroy buildings” concept, your simulator will be even more utilized by people as the best and as we spread the tale.

Asipak: Look forward to hearing from you :)

Man, calculating it was already a bitсh before...
Gj figuring it out guys.

Hi!

Thanks for your great work. I changed the settings of the calculation. Only unit types that are alive after the battle take part of the fight against the building.

This really makes more sense, perhaps BB discovered this „bug“ and fixed it. Do all your collected data match now to the results of the simulator?

Quote:

I rewatched your Proof video in post #67. You don’t show the units alive after the battle is over, so perhaps an Elite or even a Recruit + Elite was also still alive at the end of the battle? When I ran a simulator on the configuration you used, I only saw losing both elites about 7% of the time.

I don't see any elite soldier alive after this simulation. They always die: Simulation. The simulator now predicts a fight against the building about 6-7 rounds, but the video shows only 5. The only reason for this might be a fix of BB.

Regards

Ok, something is still wrong.
Look at this simulation.
Keep refreshing and checking the "fat" building destruction times.
Sometimes it goes 7-8 rounds, and sometimes 8-9.
Are you positive it really does 10,000 simulations?

Quote:

Originally Posted by Lamosos

Ok, something is still wrong.
Look at this simulation.
Keep refreshing and checking the "fat" building destruction times.
Sometimes it goes 7-8 rounds, and sometimes 8-9.
Are you positive it really does 10,000 simulations?

Asipak, would it be beneficial to perhaps add at the top of the destruction times window, the surviving troop types to help out with this at this time?

Quote:

Originally Posted by Asipak

Hi!

I don't see any elite soldier alive after this simulation. They always die: Simulation. The simulator now predicts a fight against the building about 6-7 rounds, but the video shows only 5. The only reason for this might be a fix of BB.

Regards

You are also losing a ton of Cannons, since you are looking at the config for the middle tower. The video is of the First (Left) tower. Try this Simulaton.

Quote:

Originally Posted by Lamosos

Ok, something is still wrong.
Look at this simulation.
Keep refreshing and checking the "fat" building destruction times.
Sometimes it goes 7-8 rounds, and sometimes 8-9.
Are you positive it really does 10,000 simulations?

Since the surviving troops are going to be variable (most of the time) he is going to have to run a couple variations of the building destruction times to get the proper range. Probably a set of 10000 based on the surviving troops win minimal losses, and another based on surviving troops with Maximum losses (that was still a win).

He might currently be running only the results from the last battle sim.

Just speculating.

Quote:

You are also losing a ton of Cannons, since you are looking at the config for the middle tower. The video is of the First (Left) tower.

Hmm, I'm not sure if that's right because in the video he's attacking middle tower, isn't he?

Quote:

Originally Posted by Lesson

He might currently be running only the results from the last battle sim.

That's the point. I think I have to save some more details for each battle for being able to get the proper range.

Quote:

Asipak, would it be beneficial to perhaps add at the top of the destruction times window, the surviving troop types to help out with this at this time?

Yes, I can do that, but that won't solve the problem.

Quote:

Probably a set of 10000 based on the surviving troops win minimal losses, and another based on surviving troops with Maximum losses (that was still a win).

Good idea, that should be enough.

Quote:

Originally Posted by Lamosos

In other words, "why bother when things MIGHT go wrong"?
You know you're gonna die eventually, right?
Why not just go suck on a 12 gauge right now, instead of leaving it to chance?

rofl