|
|
Post by Daimaoh34 on Jul 7, 2004 20:58:00 GMT -5
OK, this is coming from the same person that felt like the computer didn't want me to win on Cratos 1.
Anyway, ever since that match, I've only lost once to the CPU in story mode. So I've been wondering, has the computer been taking it easy on me?
I only have a little of 300 different cards in my collection, and I have been playing an OK green-red deck, but I don't understand how I got started on a 9 game winning streak against CPU characters in Story Mode.
After reading the thread taking about how Zeneth gets really lucky, I was dreading facing him. I gave it a go, and beat him in 3 straight games to unlock Najaran (sp?). Then I go onto challenge Najaran and beat him in around 30 rounds as well...
It seems like the computer gets "lucky" rolls in the first 15 rounds or so of a match, and then all of a sudden they start getting really bad rolls and landing on all of my high level lands. So I usually come from behind in all of these matches, and somehow end up winning...
I wonder if CPU "luck" increases/decreases depending on how large your card collection is? Or maybe it's because I had a large amount of losses vs. CPU in story mode?
Anyone else had a streak of luck like this?
|
|
|
|
Post by Thumpasaurus on Jul 7, 2004 22:06:38 GMT -5
I'll reiterate, I am firmly of the belief that the CPU does NOT cheat in this game. Its decks are shuffled and randomized every time you play, with no ability to draw the right cards at the right time. Its die rolls are completely random every time as well.
I've played this game so much that I've seen just about every degree of luck for myself and the CPU. Some games the luck goes all one way. Other games, it goes completely opposite. Some matches start out lucky for one and not the other and vice versa. It's all LUCK!
The CPU did not take pity on you. And when you feel like everything is going against you, it's just bad luck on your part. The same rules apply to the CPU and human players. Once you acquire good cards and learn good strategy and how to assemble a good deck against your opponent, pity/luck won't have a whole lot to do with winning or losing. It eventually becomes pretty easy to consistently beat the CPU due to AI weaknesses and how easy it is to create counter-decks when you know what your opponent uses.
|
|
s6
Baron
i know what i like
Posts: 88
|
Post by s6 on Jul 8, 2004 9:36:40 GMT -5
Could be that the computer does not actually "cheat" but the deals and dice rolls are not exactly randomized, either. That's why I'm looking for certain patterns, as in programming because there is no such thing as "luck."
|
|
|
|
Post by Thumpasaurus on Jul 8, 2004 12:29:18 GMT -5
Of course there's such a thing as luck. By luck, I don't mean some kind of supernatural influence helping or hurting you. Luck is just how to describe drawing favorable cards or rolling favorably by means of pure random chance. If a game is going your way or going the CPU's way, it's not because of a set pattern but because of how much luck you had in the generation of random numbers. It's like if you flip a coin 10 times, you might guess how it will come up 10 times out of 10 or maybe 0 times out of 10, or somewhere in between--that's just luck, not a pattern. I am almost certain that the game uses pure randomization in deck shuffling and dice rolling. Search all you want, but I don't think you will ever find any patterns.
|
|
|
|
Post by Korvan on Jul 8, 2004 13:35:31 GMT -5
Computer generated are not exactly random. They use a formula that recyles the last last number genenerated as a seed for the formula. This esentially results in a large "virtual" list that gets walked through one number at a time. if you initialize his list with the same seed, then you will experience the exact same sequence of random numbers as you did before.
The trick to make it seem totally random, is that the cpu initializes the random list with the current clock time (measured in milliseconds).
|
|
|
|
Post by Thumpasaurus on Jul 8, 2004 19:00:24 GMT -5
The seed patterns you're talking about are very simplistic and old methods of randomization that were common when games had to fit in small amounts of data, or where sophisticated randomization isn't very important.
In this day and age, any professional computer programmer worth his salt can create a random number generator that is in all practical sense, just that: random. If done correctly, you are not going to ever find any pattern or some way to predict actual results while playing. If you think that you are going to find seed patterns in this game, I think you'll be sorely mistaken. It's true that the CPU isn't flipping coins in there, but for all PRACTICAL purposes it might as well be. You're talking technicalities and semantics which are completely pointless and relevant only to extreme technopile snobs. I mean seriously, what is the difference if it is random in the truest sense, based on milliseconds on the internal clock, or some other method, as long as the RESULT is a number that the player cannot possibly predict during a game? You've got to be seriously nitpicky to care...
The POINT is you aren't going to be able to discern a pattern, because the game is intentionally designed to be random (again, the method of producing random numbers doesn't matter, so long as it is impossible for the player to predict it).
|
|
KL
Knight
Posts: 29
|
Post by KL on Aug 28, 2004 22:57:39 GMT -5
While the number itself is probably very random, I do believe the CPU knows what number it will get before they roll it.
I've seen too many cases where the CPU will use a card like Peace on a high toll X number spaces away and then roll X on the die.
So X was probably randomly generated just like the human player rolls are, I do believe the CPU knows what number they will get. It's just a matter of whether or not they can do something about it with the cards in their hand and the situation the human/other AI players created.
|
|
|
|
Post by Computer Pity ha on Sept 21, 2004 12:16:17 GMT -5
If you are within movement range of a particularly nasty land (high toll / strong Defense) and you do something to give yourself a better chance of missing that land or improving you chances of beating it, you would say it is good play.
But if the CPU does it, it's cheating?
|
|
|
|
Post by WhoseYerLilith on Sept 21, 2004 12:33:16 GMT -5
That was a pretty awful comparison.
|
|
|
|
Post by Daimaoh34 on Sept 25, 2004 1:40:52 GMT -5
Not only that, but it was out of context.  I looked through the thread and didn't see anywhere where it was mentioned that the computer making a move to avoid a high level land cheating. The argument was on things that aren't controllable, such as rolls (when you don't have any roll-influencing cards) and card draws. Considering that I use some tricks that I learn from the computer, I would never consider a good move by the computer cheating. But when they consistently avoid rows of high-toll land on regular rolls, it makes you think. |
|
|
|
Post by Shadowshear on Jun 21, 2005 20:58:08 GMT -5
THATS WHAT IT IS!
I've been thinking for a while and even when contemplating a CPU's seemingly dumb moves (such as casting peace on a Lv.5 territory right in front of them) I think "Why are they doing that? The chances of them rolling a 1 on a 1-9 die (side note" I'll give a million bucks to the first bloke who can create a nine-sided polyhedral die.) are quite slim. then I remember: "Its because they know that they are going to land there."
then they roll a 1.
|
|
