« Back to All Topics
Invasion Research
Invasion Research
Invasion Research
Over the past two days I've done about 6k clicks worth of "invasion" research on TA.  Throughout my study I ensured I was there all by myself.  Because of the time and clicks involved, this is not a strictly scientific study, but I believe it to be a great start to understanding what is happening on TA.


WHEN INVASIONS HAPPEN

Invasions happen on a strict schedule, currently approximately 25 seconds before the market prices fall.  For example, on my PC clock, market prices fall at 15 seconds into every minute and invasions happen at 55 seconds into every minute.  Invasions, therefore can happen a maximum of once per minute, and they sometimes don't happen for as many as 10 consecutive minutes.  They seem to happen every 4-5 minutes on average, or 12-15 times per hour.

I suspect this is a misconception, but it seemed like the more rapidly I patrolled the front-lines, the fewer invasions there were.  This is probably just the "watched pot doesn't boil" phenomenon.  If someone wants to check it out, try this on a high speed internet connection: patrol all the front line blocks as quickly as you can, repeatedly visiting each block on a pre-defined route until you encounter an invasion, write down the time, repeat a bunch of times.


NUMBER OF INVADERS

The number of invaders per invasion seems to be randomly chosen between 2 and 6, weighted towards the middle, meaning 4 seems like the most likely number.


WHERE THEY INVADE

The invaders seem to only hit one block at a time.  The only exception I noticed to this is when a group of invaders try to invade a block that can't hold the entire group without exceeding it's maximum monster capacity.  In this case, the remaining invaders "spill over" onto another randomly chosen block.  In one example I observed, the invasion group had 5 monsters and they maxed out 4 blocks (that were all at max-1) and dropped the last man on a previously clear block.


WHAT THIS MEANS IF IT'S TRUE

- The "invasion trap" strategy of leaving blocks at max-1 is a waste of time and clicks.
- The number of cleared or non-cleared blocks on the front line makes no difference to the size and frequency of the invasions, although it may make a difference to other aspects of how we use TA.
- Letting invaders pile up on a block makes no difference to the size and frequence of invasions.
- Approximately 45-60 monsters invade TA every hour, or 1080-1440 per day.

This seems to make the strategy clear for me.  Simply put: first kill any monsters that are "in your way" for whatever reason (construction, monstology, land bridge, curiosity, meaningless preference, superstition, etc), then go kill all the rest of them.

Sandy

Sandy,

Well done. But one thing I am curious about.

We have cleared several squares since the invasion started, and I have not noticed a difference in the quantity or frequency of invaders. It appears to be as maddeningly random as when this all started. I have noticed large numbers of invaders, but this has been attributed to the lack of clearing activities.

We can assume that either:
A) A predetermined quantity of invaders are dispensed randomly among invadable squares.
B) A random number of invaders (2-6) are dumped on random invadable squares at regular intervals.

If A is true - then we will always have a similar number of invaders regrardless of how many occupied blocks remain. If this were true, then as we clear more squares the number of invaders would increase. This would also mean that when we are clearing the last block, 1080-1440 invaders would be dumped onto the few neighboring squares. 

If B is true - then decreasing the number of invadable squares will have an effect on the number of invaders we are forced to clear.

-B
I don't understand the distinction you are making between A and B.  It seems to me that in both A and B, the same number of monsters get added to TA, just on different blocks.

> We have cleared several squares since the invasion started, and I have not noticed a difference in the quantity or frequency of invaders.

This is exactly what I believe my research has shown.  There won't 
be
 a difference unless there is some event triggers or army thresholds built into the game at which the invasion rate changes or stops.  

> A) A predetermined quantity of invaders are dispensed randomly among invadable squares.

Do you mean predetermined as in "1000 per day", or something else?  Personally, I'm with Occam's Razor on this one.  The following are invader counts for 23 sample invasions that I observed during my research.  The exclaimation mark indicates they maxed-out a block.

2, 3, 5, 4, 6, 1!+1, 3!+1!, 2, 3, 4, 4, 2+1!, 4, 4, 5+1!, 5, 4+1!, 1!+5, 2, 6, 4, 1!+1!+1!+1!+1, 4

Size 2: 4 times
Size 3: 3 times
Size 4: 8 times
Size 5: 4 times
Size 6: 4 times

If it's predetermined, it's needlessly complicated, cause it sure seems random to me!

> If A is true - then we will always have a similar number of invaders regrardless of how many occupied blocks remain. If this were true, then as we clear more squares the number of invaders would increase.

Is the second sentence supposed to read "the number of invaders PER block would increase"?  Otherwise, it seems to contradict your first sentence. 

> B) A random number of invaders (2-6) are dumped on random invadable squares at regular intervals.
> If B is true - then decreasing the number of invadable squares will have an effect on the number of invaders we are forced to clear.

I don't understand this conclusion either.  Do you mean "per day" again, or something?  The fewer invadable blocks there are, the more monsters each block will get, on average, over time.

In my opinion, this is how it works:

1. The computer decides if an invasion will happen this minute (which seems to be about a 20-25% chance).
2. Once an invasion IS happening, it chooses the number of invaders (approximately 2-6, usually 4).
3. Then it chooses a block and adds the invaders up to the max capacity of the block.
4. If there are still invaders remaining, go back to step 3.

If we have any provable theories to try out (ie. patrolling the front lines decreases the number of invasions, or clearing the 500k blocks may trigger something good), I suggest we formulate them and prove or disprove them to keep our footing on solid ground as much as possible.  Personally, I won't be blindly working away based on assumptions any longer.  While I don't LIKE what my study is suggesting (that nothing we do is going to make any difference to the invasion rate), I certainly prefer making my decisions based on conclusions drawn from carefully observed phenomena as distinct from conjecture and superstition!

Sandy

Difference between A and B

A - predefined total number of monsters are populated on TA regardless of populated/invadable squares

clearing one block over another will have no effect on our progress

B - predefined number of monsters are populated based ont he number of invadable squares.

reducing invadable blocks will have a positive effect on our progress

-B
Okay, I think I get it.  When you use "predefined", it makes me think the exact number of invaders in a given time period is set, and is slowly decreasing as they are doled out.  This isn't exactly what you mean.

I think it's safe to say that 
if
 the number of invaders is related to how many blocks are left uncleared, it's only at the most basic level... similar to how the number of oreputty a digger digs up is related to his luck.  Each luck point makes a difference in your oreputty take, but it takes a LOT of progress to make any noticable difference.

Even if this is the case... what good does it do us?  No one is suggesting we slow down our conquesting.  We're making progress in the only way we know how: killing millions of monsters.  We can't really improve much on the rate we're killing them at.  Our only hopes for improving our situation is to either somehow reduce the invasions or spend more clicks killing them, ideally by adding more characters.

Has anyone done an "estimated completion date" recently?

Sandy

Some theories that deserve consideration of our focused attention to prove or disprove them:

- THEORY: The larger "mystery" blocks, including the block north of the lake that suddenly started expanding past it's apparent max, are special.  Clearing these blocks could unlock something that could help us.

- THEORY: Frequently patroling the front lines could decrease the number of invasions/invaders.

Others?

Sandy
While clearing away those pesky invaders in the northern city: 35 towards west, splatted them down to 1, then 11 monsters invade and force me away.  

11 seems a rather high number for a one time invasion from what I've seen, and from other studies.  Just thought I'd pass it along.
Practical experiment:
Let's allow monsters conquer some area on southern part without allowing them ruin our buildings, and after it happens - cut 1 block with monsters apart leaving it near the seaside from the other TA army, then clear all others blocks that were conquered earlier. So we will get 3 groups of monsters: Eastern (the largest), Western (left from coridor) and almost empty near seaside - 1 separated from others block with 2 or 6 monsters, or whatever... The it will be easier to study monster's evasions using this one ceparated block: 
1. We will see - is number of monsters on TA constant, or is it always increasing
2. How many monsters repopulate and where, also remembering the time between 1st and 2nd repop.
3. (maybe) - do they have some goal which they want to reach - city center, for example, or back to their army?
Let me know what you think...
P.S. I also understand that EVERYONE who trips on TA should know about this experiment - otherwise person not involved may kill that experimental block.
Were you clicking constantly through the time when this invasion happened?  I mean, was it one click they weren't there, then suddenly one second later they WERE there?  Or did a number of minutes pass by meanwhile?  If the assumptive conclusions of my research are correct, it would take a minimum of 61 seconds and 2 invasions to get to 11 invaders on a block.

Sandy
I've been splatting so much recently, I can't remember the exact details of that particular incident.  I'm usually fairly consistant while splatting, but it's always possible that I got distracted by chat or the forum or my cat stepping on my computer and refreshing the game...

Maybe I did just get lucky with two invasions to the same square within two minutes. 

TCA - I'm pretty sure someone tried the isolated square test before with inconclusive results, though I'm still confused on how that test will prove anything.  Were you able to pull any results from it the other day?
I've managed to leave a separated group twice: in both tries there was no move of these groups during 2 hours (groups of 4; then 18 monsters).
Does it means that there is some center where monsters get reinforcements from? That means cleaning the coridor, separating W army from E army is useful - W army will get no additional forces while we will be trying to clear it, unless there is some center on the W bank, too.
Now I will try to leave 2 blocks separated: will monsters from 1st move to 2nd or will they repop then? Wait for results :D
It may sound crazy and I don't believe it, too. But - when I was clearing the corrior square W from the S  square of lake, I killed a blue cityguard, started to gather hitpoints again, and when attacked - just went through it! I saw no players around and I am sure there were 6 monsters to kill yet.
Did someone notice something alike at least once?
TRAiNOR
So I came on about three hours after sandy's post that 10448 was clear.  Sad times.

I cleared over 600 invaders from TA.  That's a lot of invaders for 3 hours, assuming someone kept them clear while splatting on the target.  If you guys didn't bother clearing north city before focusing on the target, that could explain the large number of invaders I found, but if someone did, this might show that invaders are partially dependent on how much we splat (I really hope not, as if invasion forces weren't evil enough).
I have been doing an invasion study by recording target blocks monster populations and total splats by using top conquest list everyday around the same time.  
Here are the daily invasions and splats at target blocks between February 4 and March 22:

-----------
	start date	---	end date	---	tot. splats	-	invs	-	b.10205	-	b.10348

between--	04.02.2013	---	05.02.2013	---	1738	---	588	---	150	---	1000
between--	05.02.2013	---	06.02.2013	---	1437	---	627	---	110	----	700
between--	06.02.2013	---	07.02.2013	---	1665	---	665	---	400	----	600
between--	07.02.2013	---	08.02.2013	---	2820	---	520	---	100	---	2200
between--	08.02.2013	---	09.02.2013	---	3467	---	587	---	600	---	2280
between--	09.02.2013	---	10.02.2013	---	3307	---	697	---	190	---	2420
between--	10.02.2013	---	11.02.2013	---	2337	---	727	---	310	---	1300
between--	11.02.2013	---	12.02.2013	---	2908	---	598	---	410	---	1900
between--	12.02.2013	---	13.02.2013	---	2281	---	626	---	103	---	1550
between--	13.02.2013	---	14.02.2013	---	2281	---	626	---	103	---	1550
between--	14.02.2013	---	15.02.2013	---	2945	---	585	---	260	---	2100
between--	15.02.2013	---	16.02.2013	---	2146	---	476	---	270	---	1400
between--	16.02.2013	---	17.02.2013	---	2060	---	690	----	70	---	1300
between--	17.02.2013	---	18.02.2013	---	1920	---	540	---	180	---	1200
between--	18.02.2013	---	19.02.2013	---	2201	---	651	---	150	---	1400
between--	19.02.2013	---	20.02.2013	---	2187	---	575	---	212	---	1400
between--	20.02.2013	---	21.02.2013	---	1892	---	664	---	128	---	1100
between--	21.02.2013	---	22.02.2013	---	2544	---	719	---	425	---	1400
between--	22.02.2013	---	23.02.2013	---	3296	---	611	---	185	---	2500
between--	23.02.2013	---	24.02.2013	---	1873	---	573	---	200	---	1100
between--	24.02.2013	---	25.02.2013	---	1906	---	506	---	200	---	1200
between--	25.02.2013	---	26.02.2013	---	1911	---	611	---	100	---	1200
between--	26.02.2013	---	27.02.2013	---	2584	---	544	---	640	---	1400
between--	27.02.2013	---	28.02.2013	---	2286	---	486	---	400	---	1400
between--	28.02.2013	---	01.03.2013	---	2260	---	640	---	220	---	1400
between--	01.03.2013	---	02.03.2013	---	2413	---	673	---	340	---	1400
between--	02.03.2013	---	03.03.2013	---	2179	---	699	----	80	---	1400
between--	03.03.2013	---	04.03.2013	---	2025	---	425	---	200	---	1400
between--	04.03.2013	---	05.03.2013	---	2614	---	654	---	560	---	1400
between--	05.03.2013	---	06.03.2013	---	2215	---	635	---	180	---	1400
between--	06.03.2013	---	07.03.2013	---	2064	---	624	---	40	---	1400
between--	07.03.2013	---	08.03.2013	---	2008	---	538	----	70	---	1400
between--	08.03.2013	---	09.03.2013	---	2193	---	633	---	160	---	1400
between--	09.03.2013	---	10.03.2013	---	2135	---	476	---	259	---	1400
between--	10.03.2013	---	11.03.2013	---	2047	---	576	----	71	---	1400
between--	11.03.2013	---	12.03.2013	---	2157	---	607	----	50	---	1500
between--	12.03.2013	---	13.03.2013	---	2180	---	567	----	13	---	1600
between--	13.03.2013	---	14.03.2013	---	2378	---	537	---	141	---	1700
between--	14.03.2013	---	15.03.2013	---	2499	---	619	----	80	---	1800
between--	15.03.2013	---	16.03.2013	---	2303	---	582	----	21	---	1700
between--	16.03.2013	---	17.03.2013	---	2242	---	444	----	98	---	1700
between--	17.03.2013	---	18.03.2013	---	1420	---	413	-----	7	---	1000
between--	18.03.2013	---	19.03.2013	---	2414	---	648	----	66	---	1700
between--	19.03.2013	---	20.03.2013	---	2008	---	529	----	29	---	1450
between--	20.03.2013	---	21.03.2013	---	2074	---	703	---	121	---	1250
between--	21.03.2013	---	22.03.2013	---	1618	---	540	----	78	---	1000

average invasions: 593/day

-kristal
CALCULATION OF TOTAL NUMBER OF INVASIONS FROM BEGINNING TO TODAY:

Invasions started on Jan 16, 2009
ref: 
http://www.flyordie.com/games/deploy/larkinor/h/en/help/news.html?R=090109


Number of invasions at the beginning calculated by SandyT: between 1080-1440  (average=(1080+1440)/2=1260/day)
ref: research of SandyT first post of this topic

Invasions decreased on Aug 11, 2010
ref: 
http://forum.flyordie.com/thread.jsp?forum=13&thread=52537
 ( post by lilgyp)

number of days of between Aug 11, 2010 and Jan 16, 2009 = 572

number of days of between March 22, 2013 and Aug 11, 2010 = 954

total number invasions from beginning to today = 1260*572+593*954=1,286,442

-kristal
..no wonder block-clearing is getting along so slowly!
INVASION RESEARCH AFTER CLEARING 1 MILLION SQUARE


It took my attention something about block 10383 (south of 1 million block).  Its monster population stayed as 499992 for two days.  So decided to make an experiment;

1)monster population of b.10383 droped to 499990, b.10312 (300k block) dropped to 299990.

2)after one hour there were 3 invasions at 1mil block. No change of monster populations of b.10383 and b.10312

3)after 4 hours there were 10 invasions at 1mil block, b.10312 was at 300k, b.10383 was still 499990.

Conclusion: 
Effect of invasions on this block slower.
The reason of slow effect could be not having a full block between b.10383 and RC. 

If this is the case, this knowledge could help us on decision of next target.

What do you think?

INVASION STUDY AFTER CLEARING 1 MILLION SQUARE


Measurements of invasions between May 17 and June 5
609	599	609	620	657	568	591	605	628	589	523	578	569	537	701	688	460	624	688	520	


Average = 598 invasions per day


Monster population of the blocks does not effect the number of invasions.
ESTIMATION OF TOTAL NUMBER OF INVASIONS FROM BEGINNING TO TODAY:

Invasions started on Jan 16, 2009
ref: 
http://www.flyordie.com/games/deploy/larkinor/h/en/help/news.html?R=090109


Number of invasions at the beginning calculated by SandyT: between 1080-1440 (average=(1080+1440)/2=1260/day)
ref: research of SandyT first post of this topic

Invasions decreased on Aug 11, 2010
ref: http://forum. flyordie.com/thread.jsp?forum=13&thread=52537 ( post by lilgyp)

number of days of between Aug 11, 2010 and Jan 16, 2009 = 572

number of days of between June 5, 2014 and Aug 11, 2010 = 1394

total number invasions from beginning to today = 1260*572+598*1394=1,554,332

-kristal