Battles aren't random, they're scripted

[Cartouche Bee]

phizuol, but even in the earliest fortran languages they used formulas for random number generation. Lists would have been way to expensive to house and retreive data from. I'm pretty sure that if someone looked at the source for that fortran90 function, no look up list is involved at all.

 

[phizuol]

Yeah, that's why I said I didn't know of any lists that had been used on a computer, only that the lists do exist. Why bother when you have the computing power? hehe.

 

[Hygro]

Grey Fox that is what I kept trying to say, but everytime I thought of it I didnt write it

 

[zeeter]

Wait. The point of this thread is that we're discussing whether or not the RNG is actually a sequence rather than a random number. It seems to me that there is an easy way to find out. With preserve random seed on, save at the begining of a turn when at war. Run through a series of five battles or so. Reload without saving. If the results are always the same, then it is a sequence. If they are ever different then there is not a sequence.

 

[Tacit_Exit]

My two cents worth:
Nothing is truly random.
If you can precisely identify all of the factors influencing an event and quantify those influences then you can tell exactly what the 'event' will be.
Even Chaos is not random,just a standard limit function with an infinite string of limits.
Things just appear random if you don't fully understand the influencing factors.
If you got anything out of this please send 2 cents to . . .

 

[sumthinelse]

Originally posted by Grey Fox
You will often see the same patterns if you are often faced with battles between the same type of units on the same type of terrain.

"Often" ??? Is there any way you can express your idea better, such as " Vet Knight vs. reg spearman in such-and such a situation produces the following result 50% of the time in 300 trials: ...."

 

[Hurricane]

Originally posted by Tacit_Exit
My two cents worth:
Nothing is truly random.
If you can precisely identify all of the factors influencing an event and quantify those influences then you can tell exactly what the 'event' will be.
Even Chaos is not random,just a standard limit function with an infinite string of limits.
Things just appear random if you don't fully understand the influencing factors.
If you got anything out of this please send 2 cents to . . .

Guys, have you any idea what you are talking about? You obviously know nothing about real randomness, of which there are plenty in the real world. Quantum theory, for example.

 

[Hurricane]

Originally posted by zeeter
Wait. The point of this thread is that we're discussing whether or not the RNG is actually a sequence rather than a random number. It seems to me that there is an easy way to find out. With preserve random seed on, save at the begining of a turn when at war. Run through a series of five battles or so. Reload without saving. If the results are always the same, then it is a sequence. If they are ever different then there is not a sequence.

If that is your definition, it is definitely a sequence. But this sequence is just as random as a game of Black Jack, where the shoe is filled with 100 million decks of cards, very thouroughly shuffled. Now tell me any gambler who would claim they can predict the next card by looking at the previous.

 

[Tacit_Exit]

Originally posted by Hurricane


Guys, have you any idea what you are talking about? You obviously know nothing about real randomness, of which there are plenty in the real world. Quantum theory, for example.

Your reply makes it obvious it is YOU who doesn't understand what I'M talking about.
Of course randomness as a concept exists, or there wouldn't be a word for it!
btw I'm just one person, you can address me directly if you wish to quote my posts.

 

[Grille]

Originally posted by Cartouche Bee
phizuol, but even in the earliest fortran languages they used formulas for random number generation. Lists would have been way to expensive to house and retreive data from. I'm pretty sure that if someone looked at the source for that fortran90 function, no look up list is involved at all.

AFAIK, that function is a list (I want to believe...). Due to memory, a list
is kind of expensive, but saves cpu time.
But also I have to admit that there are indeed formulars for a "random" number
calculation. I found one where the currently drawn number is dependent on the
seed, pre-drawn number and a certain range. To put a line back on topic,
such dependencies would be basic necessities for patterns or scripts.
But I'm still thinking civ3's RNG is a "true" one (I want to believe...).

 

[TheNiceOne]

Originally posted by zeeter
Wait. The point of this thread is that we're discussing whether or not the RNG is actually a sequence rather than a random number. It seems to me that there is an easy way to find out. With preserve random seed on, save at the begining of a turn when at war. Run through a series of five battles or so. Reload without saving. If the results are always the same, then it is a sequence. If they are ever different then there is not a sequence.

But it is a sequence. Noone who knows a bit about computer-generated pseudo-random numbers questions that. A sequence that are not pre-made, but a sequence where the next number is calculated from the previous when you need it. There are no random elements in the calculation. So with the same random seed, you will get exactly the same next number each time.

The point is that the formula which calculates the next number of this sequence is so good that for all practical matters, it is a random number.

 

[TheNiceOne]

Originally posted by Dementia
Well I know something dodgy is certainly going on in my current game:

I was fighting is archers or something, both me and my enemy each had 4 health bars each.

I Hit
I Hit
I Hit
Get Hit
Get Hit
Get Hit
Get Hit
Die

I'm not exagerating when I say this exact sequence happened 8 times in a row, and overall in the battle about 14 out of 20 times.

Now THAT isn't random

The same was happening when they attacked me, I successfully defended 3 times then got hit 4 times and died. Lost my entire fookin force and he (AI) lost about 4/5 units. - Only on regent difficulty too.

From what I could see everytime I got the first 2 hits it followed the sequence above. However my other games are fine, which suggests there may be a bug somewhere or something due to the seed value for that game?

Do you have a save? (no - I didn't think so)

I believe you're exaggerating. The results civ3 produced went through a filter which is called the human brain - in this case your brain. The human brain is made to recognice patterns - this is simply one of its functions. By doing this, the brain also filters out (or forgets) things that don't fit into the pattern it expects.

In your case, you started to expect this pattern after two or three suspicious results, and after that I believe you tend to filter out the results that don't fit in your pattern and exagerate those that almost do.

Do you disagree? Then show me a save.

 

[Grille]

I totally agree.

 

[Grey Fox]

Originally posted by sumthinelse


"Often" ??? Is there any way you can express your idea better, such as " Vet Knight vs. reg spearman in such-and such a situation produces the following result 50% of the time in 300 trials: ...."

Ok I first have to say that I in no way believe in this "script"-thinghy. I know how the game works, and what I wanted to say with this:

You will often see the same patterns if you are often faced with battles between the same type of units on the same type of terrain.

...is that there aren't so many possible type of results you can get in a similar battle. If you almost always battle with Swordsmen against fortified Spearmen in towns you will most of the time end up killing the spearman, and ending up with 2 or 3 Hp left (if they were both Veteran), most of the time.

 

[TheNiceOne]

Grey Fox and sumthinelse: I don't think you two disagree.

I'd like to give an example:

Assume that two players each attack three spearmen with three swordmen and both looses all three battles so that the spearmen have one HP left.

Player A gets the result l-w-l-w-l-w-l for all three swordmen while player B gets the result w-w-w-l-l-l-l.

Now quite a lot of players will think A's result seems plausible, while B's result seems strange and non-random, while the fact is that those two sequences has exactly the same probability.

 

[Plexus]

Originally posted by Zachriel


Gamblers' Fallacy
http://info-pollution.com/evidence.htm

The human mind sees patterns where none exist. Just look at the clouds. Do you see the man with the big nose?

Ever see the movie Pi, Zachriel? Good, weird movie.

Patterns are not in the Civ3 battle system.

 

[ukrneal]

There was an article in the New York Times around Aug 10, 2002 (I could be off some days) that touched upon this very subject. One part I liked was, "We believe in such things as hot hands and arthritic forecasting and predestined blind dates because we notice only the winning streaks, only the chance meetings that lead to romance, only the days that Grandma's hands ache before it rains. ''We forget all the times that nothing happens.''

This is so true. Even if we suspend disbeleif and believe the same event happened 8 times in a row, why is this inherently non-random? IS it becuase you lost and you sought an answer to explain that (couldn't be luck or poor strategy, I'm sure) or is it because it didn't fit your "definition" of random. Plus, then we pesonalize it and it takes on a whole new meaning.

It could happen 100 times in a row and still be random.

As to predicting - that's great luck. Is it knowledge or coincidence? Again, it is selective memory. How often are you wrong? I'll bet that in reality, you are wrong more often than you say you are. I don't say this to say you're liar. Quite the opposite, you simply remember the times you were right more clearly and say how good you are at predicting. We all do it.

I remember one line from the article that stuck with me. There are approximately 280,000,000 US citizens. That means there will be 280 million to one shots every day! Just think about that!

 

[sumthinelse]

Originally posted by TheNiceOne
Grey Fox and sumthinelse: I don't think you two disagree.

OK, sorry Grey Fox, I thought you meant that you were seeing patterns but now I see what you mean. Since you are one of the most thoughtful of our posters, I thought you might have found some "non-random" patterns.

I am a professional software engineer, and I know that programs have bugs. Possible bugs would be in thr RNG itself, or in the use of random numbers. I have not seen any evidence that such bugs exist in Civ3, but I try to keep an open mind.

 

[Grey Fox]

Originally posted by sumthinelse
OK, sorry Grey Fox, I thought you meant that you were seeing patterns but now I see what you mean. Since you are one of the most thoughtful of our posters, I thought you might have found some "non-random" patterns.

Maybe I should have written it in a better way the first time , but I was a little tired...

 

[sumthinelse]

Originally posted by Grey Fox
Maybe I should have written it in a better way the first time , but I was a little tired...

And my reply to your post might have been stated a little more tactfully.

Anyway, I think you and I and others who have thought about this issue a little deeper agree that it's just human nature to be tempted to misinterpret random events. At least in Civ3 the victims of these delusions don't lose their life savings, as they would in Vegas or Monte Carlo.