Hello friends, I spent "some" .png" alt="smiley: " title="" />, time to make tests inspired by unanswered questions asked in https://www.games2gether.com/endless-space/forum/27-general/thread/10588-experimented-with-bombing-invasions-conquering-results

Xen0n have made a good job, but still nobody continued any further, obviously testing is very time consuming and still not everything can be resolved "on your own", so I want to share my fresh efforts with you and hear what you think and to tell if I concluded something wrong

Also I found a thread https://www.games2gether.com/endless-space/forum/29-archives/thread/13126-exp-system-invasion

It looks that not everything applies in today's version but some formulas are just fine

It's somewhat long thread so I marked sections with red so you can skip section if you wish..

I wish I could put more numbers from my test game but that would be way to long thread then.



If you guys are curious how did I perform these test here is how:

I started a game on most easy settings and over tech-ed the AI 30 times to make things as easy as possible.

after 100 turns I saved the game, and started to attack always the same system by reloading the saved game.

Few turns forth or back sometimes.

as simple as possible





Here is in short what areas have I cover:

-Industry cost of hulls with industry cost discount (discount not displayed but applied)

-Bombs

-Troops

-Raze system feature

-Troop and bomb module differences in action

-ownership gain formula after troop invading a system

-collateral damage and defense shelter

-how is dust cost of fleets and ships calculated and how and when it changes

-how does ship experience influence bomber, fighter and troop module effect







INDUSTRY COST DISCOUNT "BUG"

Many people asked that question and many of them didn't know how to answer, but here is the trick...

When you research a hull type that gives you, say, -15% ship cost on star system, you build a ship and it does not provide any bonus of -15% (or what ever bonus) at all.

so I created two ships (one hull with -15% and another without discount) with identical modules.

I put them on the queue to see the difference but both ship classes showed the same industry cost :/

The trick is that you won't see any -15% industry in the queue unless that ship 'hits' the first place on the queue, that means no other improvements or ships are being build in the queue, just the ships with -15%.

So if you put some other ship or improvement in front of your 'discount' ships then industry price will be with no -15%. (but as soon as it comes back to the first place it will show you -15% industry again)





BOMBS

In military tech tree when researching bomb modules you can hover your mouse over the bomb icons, and, you'll see following 2 descriptions:

Cluster munitions (anti personel bomb) and

Bunker busters (bomb that destroys buildings)

Now as soon as you research these two bomb techs you go to ship design to make a ship with a bomb and there you'll see somewhat different description for the 'Cluster munitions' bomb:

AP-bomb

however the other one 'bunker busters' descriptions is the same as in tech tree.

Ok, that's not such a big deal but...

I constructed a small hull ship with these 2 bomb modules and started to bomb the AI's system...

when I got the option to use the bombs it gave me 2 options to use:

cluster bombs (destroy infrastructure) and incapacitation agents (hospitalize population)

so I clicked cluster bomb field to bomb the AI's population. but that option actually destroyed buildings :/

Not such a big deal I missed once because English is not my primary language..

The point here is that I wanted to point out that my English is OK but

that in game description grammar bug is actually wrong

I hope developers will fix that some day ^^



bombing systems:

as already told in the thread (link above) to successfully bomb a system all you need is only 1 single bomb of desired type (AP or cluster).

And enough land invasion strength which is provided by bomber, fighter or even troops modules.

however I want to point out that not every type of bomber fighter is good for that, actually only one is not OK, that is the battle type of fighter because it provides 0 land invasion strength.



Only impact on the bomb destruction is LIS (land invasion power).

you need a minimum which depends on systems AA defense, any amount more than that will yield stronger result (more improvements/population destroyed, depending on bomb type)



Here is what turns out to be true for each bomb attack on population:

example from my test game:



Target system AA defense = 1660%



So what does that number mean?

it means that you need at least 1661 land invasion strength (LIS latter) to be able to drop a bomb, .png" alt="smiley: " title="" /> yes

the percentage here is cumbersome, but that's what it means.

so I sent 9 ships equipped with 200 LIS each which is 1800 LIS in total.

The bomb killed 2 out of 11 population.

I reloaded the game and attacked again with double more LIS (3600)

this time it killed 3 out 11 population.

I reloaded the game again and attacked with 7200 LIS

this time it killed 5 out of 11 population.



in percentage that is:

1. time 18%

2. time 27%

3. time 45% of population.



By some logical means it look that every LIS of amount of AA defense kills exactly 9% of population more.



And here is what turns out to be true for every bunker buster bomb:

Throwing cluster bomb on the very same system here is the same scenario as above just another bomb type:



1. time 2 out of 8 buildings (25%) (1800 LIS)

2. time 3 out of 8 building (37,5%) (3600 LIS)



The pattern is exactly the same, only difference is that bunker bombs deal greater damage in terms of percentage





TROOPS

To successfully troop invade all you need is at least one troop module in the system orbit.

more troop modules you bring to the front less you'll loose and your newly conquered system can end up with 0-100% ownership and 0-max buildings.



First we need to understand probability of success, it can be modified by LIS (land invasion power) or by number of troops deployed or combined.

It's not the same which option you choose:

by modifying with LIS (cheap way) you'll need less troop modules to take a system, and thus in the end your system will not be of much value because you'll end up with 0% ownership, which means a lot of disapproval and reduced FIDS output.

Second choice is to deploy large number of troop modules (expensive way) but you'll end up with 0-100% ownership and all what it goes with that (FIDS, less waiting time etc...)

There are tricky ways to reduce your massive troop loos by understanding collateral damage (you can read more about that in the collateral damage and defense shelter section)



Why is so important to understand how to modify success % ?

There are 3 troop module types (elite, trooper and numerous)

When % is 100% you have only two choices: elite or whatever (elite is the best choice in that case)

When % is less then 100% it's not the same which troop module you deploy in which case:

trooper is weak and you need more of them,

numerous is stronger and you need less (regardless of in game description)

elite is strongest and you'll need not so much of them. (but you'll end up with less ownership)



I ran a test to test the difference between troop types...

First let's see what's the difference between 'troopers' and 'numerous' type,

In design view the only difference is that 'numerous' type costs 1 population unit more than 'troopers' type, all other properties are exactly the same.

Now, when it came to to battlefield these troops behave differently.

'elite' troop type is stronger in terms of troop efficiency value.

There are 2 scenarios from my test:



scenario 1:

mix 'elite' with 'numerous'



scenario 2:

mix 'elite' with 'trooper'



first scenario is stronger



Here is my test result:

I deployed a fleet with 6 elite and 4 numerous against 20 population system,

2 troops survived 14 populations killed



in second scenario I switched numerous with trooper type:

0 troops survived 3 population killed.



Big difference, however I won the system in both cases, only difference is that in first case I got 5% ownership while in another 0% (more on that later)



I ran the test again but different ratio of numerous/trooper type mixed with elite type.

I also ran the test without elite type, meaning, I deployed either numerous type or trooper type, and guess what? no difference between the 2.

So by running several tests, difference can be seen only when elite type is present in the fleet.

There is also an important factor that has to be taken into account.

In both scenarios above my % probability to success was less then 100%

I tested with 100% probability, and in that case neither one of 3 troop types showed any difference except elite type.

When % probability is 100% the amount of 'elite' needed is much less then other 2 types (because they provide more LIS trough "troop efficiency value")

Logical conclusion is that different troop types (not elite one) are really different when success probability is less then 100% ^^



so 3 kinds of troop modules and each one has it's own 7 properties:

-collateral damage

-siege

-efficiency

-industry cost

-population cost

-tonnage cost

-special slot cost



the most important properties for us here are efficiency and population cost:

troopers----250 eff. (1 population)

elite--------420 eff. (1 population)

numerous---250 eff. (2 population)



troop efficiency value is actually land invasion strength!

If you deploy enough troops in terms of LIS you'll get % probability which depends on systems AA defense.

more LIS then there is AA defense will have no additional effect.

instead of troops you can deploy additional LIS with bombers or fighters of course.

What is the real purpose in that numerous cost 2 population, as far as I know from my test they turn out be more powerful then troopers when % of success is less then 100 and when combined with elite troops.



Real question here is how to figure out how many population we need in order to get 100% control?

to answer that question let's first understand that battle report will not show you how many population units we loose, it will actually show us how many troop modules we lost.

but on defenders side it will show not population modules but the actual survived system population!

That means if your 1 troops module with 2 population survived it will actually add only one population unit to the survivor list.

troops added back to the troop module will also be 1/1



Here is my test on first troop invasion:

I deployed 8 troops against 14 population system

3 troops survived on my side and 6 population survived on defenders side.

I won the system with 9 population and 15% ownership from the start...



Battle report says that 6 defending population survived and 3 troop module on our side survived, that totals in 9 population units which will be added to the system.

Reading battle report, just add up your survivors and defenders survivors and that's the new system population.



So in the invasion case above I mixed 2 elite modules and 6 numerous troop modules onto 3 ships.

I captured the system with 15% control from start, how and why 15% ?



Here is the formula:

3 troop modules survived in the battle against 20 population system

3/20 = 0,15 (15%)

that's where this 15% percent of control came from...

The control gained is so important because our goal is to troop invade the system with 100% of control which very possible as shown in my next test with troop invasion:



I deployed 40 troops modules against the same 14 population system!

I lost 8 troop modules and 6 defenders population units survived.

at the end I ended up with 32 population system and 100% ownership and no approval penalty at all.



Why 32 population?

According to above formula 32 of my survivors added with 6 defender survivors is 38 survivors.

but system only had pl

Can you sum up your findings?

Nasarog wrote:
Can you sum up your findings?

I don't understand your question,

What you mean by "sum up" my findings?

codekiddy wrote:
I don't understand your question,

What you mean by "sum up" my findings?

Basically you have a lot of data here, but it would be useful at the end of the post to show what conclusions you've reached. A reader may get lost amongst all the facts, numbers and results from your testing, so if you "sum up" your findings at the end it makes it much easier for a reader to take in what you have discovered in a short paragraph.



e.g. Having talked to 100 children I can conclude that chocolate is more popular than vegetables.



Hope that helps,

Oh, I understand..

Thread is divided into 7 red marked sections where each section tell's different testing story. Only two sections which have stuff in common are 'troops' and 'collateral damage' so this is probably most confusing. I'm sorry for that.

I'll try to put these 2 into one section and shorten up everything sometimes.

I know I'm bad writer but what can I do :/

codekiddy wrote:
I don't understand your question,

What you mean by "sum up" my findings?

You posted a "wall of text". If you want people to read what you posted, you need to quickly summarize it to get us interested.