I gotta complain...

[FenrysWulf]

I know this has been said a million times, and I've seen the mathematical odds. Nevertheless, something that just happened in my game was so riduculous, I need to complain. I had a stack of tanks I had just landed outside an ottoman city, and they started their counterattack. Out of six consecutive attacks by (in order) a knight, longbowman, and four medieval infantry, I lost 5 tanks, and the other was reduced to one hitpoint. They had all been veterans at full strength. Now I know that with the odds this is technically possible, but come on! What kind of luck is that?

 

[Aramazd]

Yeah RNG is streaky

 

[Turner]

Frustrating, to be sure.

Just remember, with the RNG all things are possible!

 

[zorven]

This is where I use my imagination: the tanks landed on a beach that was surrounded by cliffs - no way out except by foot. Tank drivers vs. knights, etc stand no chance.

 

[Mongoloid Cow]

Or you could say that your tanks got stuck in quicksand.

The RNG is evil. I always end up having to take 3 to 4 times as many of my units to take a city than it seems most of the other players on these forums have to. A good idea for next time might to take a couple of naval units which can bombard and start hitting the city you landed next to. The AI then are less likely to go on the offensive against you as they should be down a couple of HP.

 

[FenrysWulf]

Well, they didn't attack from the city I was next to; they brought in units from elsewhere. With railroads they can do that. I still took the city because I had overwhelming force, but I just hated to see one after another my tanks succumbing to medieval units.

 

[Xink]

Being new to the game this causes me some difficulty when starting a military campaign.

Having bombers located to soften the target as well as naval bombardments can sometimes be enough to reduce a city to no defenders. However when moving in my modern tanks even against medieval infantry or cavalry that sometimes remains it is not a guaranteed outcome that I thought it would be.

How your resources (troops) start to feel it also if you need to leave behind numbers to deal with resistors, unless you destroy the city.

A well thought out campaign can to often run out of steam due to diminishing numbers.

Morgan

 

[budweiser]

Well, tanks dont land first in any invasion, its infantry that go in first. No one in history ever landed tanks by themselves in the first wave of an invasion.

 

[Master Shake]

RNG?

Also, has anyone done the relevant experiments with C3C just to be *sure* nothing in the combat engine is screwed up?

I've been having some awfully strange wars...

For example, the Zulu just took over half my cities. Each time he used a warrior and an archer against two spearmen, fortified inside a town. Without exception, he destroyed each defending army in one turn. The warriors were regularly defeating fully loaded spearmen! Not impossible, I know, but still... Frustrating.

-mS

 

[KingTux]

I think I know what makes the tanks lose... A fear of pointy things getting stuck down their nozzles! Just think of it, they're probably pointing their nozzles in the air and firing on themselves because they're afraid that if they get a spear, sword, or arrow down their nozzle, thet they'll blow up. So, they just fire in the air and hope they hit the enemy.

Nah, that's a silly idea. Definately the random number generator. However, I've felt like this is what they're doing when they're losing, and this is the only thing that I can think of that makes logical sense. Personally I think that one unit should represent ten or so. This way, the random number generator can be a little more useful and give better predictability to outcomes. The best part about it is that they don't have to change how it looks to us, but only change it for the inside. Then again, it could run ten or more simulations and then inflict damage based upon who won the most. This would remove some randomness, but make it a little more realistic.

 

[Xink]

Couldn't argue with you there budweiser. Then again no one in history managed to fight a tank and beat it with a spear, that I know of anyway - except in this game The attacking stats of infantry aren't as good as their defence stats so it makes sense, to me anyway, to use the tank.


Master Shake I found that my four Vetran spearmen were needed to defeat one attacking regular spearman - that RNG (Random Number Generator) has a lot to answer for I couldn’t believe it when they were killed one by one

Morgan

 

[MadScot]

Ah, the old tank vs spearman debate....

In the absence of a search function, let me assure you....

some really really keen people have run ridiculously detailed tests on the game system and both the RNG and the combat system which applies the RNG are entirely 'fair'. There is no bias against the player in the way it works.

also, the RNG is indeed 'streaky'. One way people have used to reduce the streakiness is to significantly increase the number of hit points per unit. That causes more RNG checks per combat and tends to even out the streaks.

and my own personal take on it is that they aren't really spearmen, just badly equipped modern units. Which have little chance of winning, but not NO chance. Think guerillas/mujahedin vs russian MBTs, rather than spearmen vs tanks.

 

[FenrysWulf]

The thing about the RNG is that we don't complain when it's our pikeman who defeats a tank, but we're pissed when the Ai's pikeman beats our tank. I think of it as my units falling on their swords, or running into the walls. Once or twice is lame but expected. It's just when it seems to happen too often that I really take notice.

As far as landing infantry first goes, I don't build infantry to take on attack missions. My invasion is all about mobility, so no units with only one movement are allowed.

 

[cgannon64]

The RNG sucks. Go chill out and then play some more.

 

[The Last Conformist]

That's not entirely fair, FenrysWulf. I've had some absurd streaks of luck which I've even reported on the forum; how my Elite Cav took out, without suffering any damage, a Vet Mech Inf fortified in a metropolis, or how attacked a Hill city held by four Riflemen with Medievals, and only lost two of the buggers.

The later was in game as the Spanish in which I was just invincible. Repeatedly faces with a technologically and numerically superior opponent, frontal assault tactics carried the day every time. Bizarre rarely; the only thing that actual went wrong in that game was triggering Domination when aiming for Conquest; I somehow managed to forget the existence of the former victory condition.

 

[Xink]

Originally posted by MadScot
and my own personal take on it is that they aren't really spearmen....

I think that you are in denial there

They really are spearmen , I can see that big pointy stick thing stabbing at my tank, at the very worse they should only be able to scratch the paint and even that is unlikely with bronze weapons
But I take your point on the emphasis of the word 'really' - I tend to agree - just for the sake of a different type of unit I might believe them to be armed with molotov cocktails or something. Rather than still stab you with a spear it is a pity that they can't throw something combustable at the tank to cause visible damage.

some really really keen people have run ridiculously detailed tests on the game system and both the RNG and the combat system which applies the RNG are entirely 'fair'. There is no bias against the player in the way it works.

Then I will have to think it through further about what I see happening - but thank you for that information. I appreciate what you have written as I am having some difficulty on understanding some conflict outcomes that I'm seeing.

There is no point in me listing lots and lots of instances that I think is 'strange' but just the last instance will do...

6 Vetran Infantry of mine attack a dutch city with no walls (stage 4 built on flat) guarded by 2 regular riflemen. Result is that all 7 of my vetrans are now dead. Both the defending riflemen were injured in the yellow.

I will put my neck out now on the block and suggest that the combat is weighted towards the AI as I do not see enough of those amnomolies of combat happening in my favour. Equally I do accept that I am very new to this game and time and experience might make me believe differently.


Take care

Morgan

 

[MadScot]

One explanation as to why the human tends to see biased results is that we are smarter than the Ai. Therefore:

1. Humans don't go into battle with obsolete units in general. How many humans attack tanks with spearmen? If you never do it, you can never win while doing it.

2. Humans don't make ridiculously against-the-odds attacks, usually. So again, we see the bad luck RNG cases, but never get a chance to experience the lucky streaks so often.

Also, psychologically, we are more likely to remember the streaks of bad luck than good. It's just the way we are wired.

 

[Old&Slow]

Originally posted by FenrysWulf
I know this has been said a million times, and I've seen the mathematical odds. Nevertheless, something that just happened in my game was so riduculous, I need to complain. I had a stack of tanks I had just landed outside an ottoman city, and they started their counterattack. Out of six consecutive attacks by (in order) a knight, longbowman, and four medieval infantry, I lost 5 tanks, and the other was reduced to one hitpoint. They had all been veterans at full strength. Now I know that with the odds this is technically possible, but come on! What kind of luck is that?

The odds of that are Astromical.

The AI cheats in combat, and always has no matter what " tests " have been run since the " tests" are not being run by controlling the AI.......also watch how many times the AI unit gets the first attack either in attack or defense.

The only way to balance a Game wherein Spearmen/Archers/et al fight Tanks is to allow cheats for the AI in the CRT.

Just accept it, it`s the way the Game works. ( or play Civ 2 )

 

[puglover]

Originally posted by Old&Slow


The odds of that are Astromical.

The AI cheats in combat, and always has no matter what " tests " have been run since the " tests" are not being run by controlling the AI.......also watch how many times the AI unit gets the first attack either in attack or defense.

The only way to balance a Game wherein Spearmen/Archers/et al fight Tanks is to allow cheats for the AI in the CRT.

Just accept it, it`s the way the Game works. ( or play Civ 2 )

It's random. The game doesn't "cheat". If any game cheats it's civ2, where I regularly lose stack after stack of cavalry in an attack on a phalanx in a walled city.

 

[MadScot]

Originally posted by Old&Slow


The odds of that are Astromical.

The AI cheats in combat, and always has no matter what " tests " have been run since the " tests" are not being run by controlling the AI.......also watch how many times the AI unit gets the first attack either in attack or defense.

The only way to balance a Game wherein Spearmen/Archers/et al fight Tanks is to allow cheats for the AI in the CRT.

Just accept it, it`s the way the Game works. ( or play Civ 2 )

Actually, those odds are NOT astronomical. They are high, but within the realm of experience.

Using this combat calculator from this site, I get a 14.3% chance of a longbow beating a tank. Assuming the tank was on grassland or similar. It can't be fortified, since it just landed. I assumed all units were vets.

5 in a row won, which is odds of 0.143^5 or 0.00005979. That's about 1 in 16,700. Hardly astronomic odds. The last unit lost; even had it also won, the odds would only have been 0.00000855 or 1 in 117,000.

To put that in context for the real world. When designing aircraft, equipment will fail. To ensure your safety, the FAA mandates that certain failures be "extremely improbable" - that is interpreted as occuring one time in 1,000,000,000. That is almost 10,000 times LESS likely than the event you are reporting.

Unlucky? Yes.

Evidence of "cheating"? Not by a long way.

And those "tests" you so easily dismiss have been run with AI attacks and human attacks. Both yield the same results to within statistically insignificant margins.

The game does not cheat. It gives the AI a number of advantages; some are listed in the manual and civilopedia (bonus units, cheap research etc.) Some are 'hidden' - 'know location of all units' etc. But a biased combat system is NOT part of the AI balancing.