Quick Answers / 'Newbie' Questions

I am not a nerd either and I hate Windows but have you 'updated' (= upgraded) to Windows 10 because I'm not sure the CD works on that platform? Others, vastly more knowledgeable than me, will doubtless chip in.

 

http://forums.civfanatics.com/showpost.php?p=13958362&postcount=6

The update deaktivated safedisc, a driver to support copy protection. As many old games require that copy protection they will no longer run. Microsoft included that disadvantage from Windows 10 to older operating systems. Congratulation are in order.

So you have three options:

1. Use Windows update to deinstall KB3086255 and shortly later put it on the list of ignored updates. Then it will no longer install itself and you no longer suffer from the disadvantages of that update.

2. Aktivate safedisk manually, the post i linked to does describe how that is achieved.

3. Use the NO-CD-Patch. That will help only for existing installations.

 

Unless something has changed recently this won't work with Windows 10, since Microsoft has enabled forced updates. There's a tool to permanently uninstall/block some unwanted updates, but security updates, like this one, even if uninstalled will automatically be re-installed by Microsoft.

 

No-CD patches…? I'm not sure there's one for v 1.22.

If you want to skip Windows updates, then maybe you could use separate accounts and never use the administrator one…?

 

Just out of curiosity, what is the probability of:

Veteran Swordsman defeating a Fortified (in a 13+ Metro) Veteran Infantry (Metro built on normal Desert, no Civil Defence or etc'ers). Swordsman left with 1 Hit Point.

Can someone write it out for me, like a step-by-step "What exactly happened".

Thanks

 

Very low. I don't have the math on me (I think it was on a thread by Steph?) but it's rather difficult.

You're attacking with 3 against a unit defending with 10… and that unit has a defensive bonus from the metro. You'd have to score four hits without ever losing… so, no, sorry, your Swordsman is almost certainly dead. Still, you can try it for the fun of it.

 

Yes, there is. I haven't played without it for years. I am always surprised when someone does not use it.

desert gives +10%, fortified gives +25%, Metropolis gives +100%, so the Infantry has a defence value of 10 x (1+0.1+0.25+1) = 23.5.

Now for each combat turn the probability is 3/(3+23.5) = 11.32% for the swordsman to win the turn, for the infantry it is 23.5/(3+23.5) = 88.68%. As no retreat can occur the battle will continue for as many turns are needed for one unit to be defeated.

I am not in the mood to calculate the rest manually, but the CivIII Combat Calculator says the attackers wins in 0.433% of cases.

 

That is so illegible, lol. Give me dice rolls over whatever system that is any day of the week.

20 sided = Swordsman 4-23 vs 24.5-43.5 - Swordsman cannot dent Infantry.

40 sided = Swordsman 4-43 vs 24.5-63.5 - Swordsman has a decent chance.

60 sided = Swordsman 4-63 vs 24.5-83.5 - It's just a crap shoot.

I also had some other ideas, such as Upgraded Units change their animation avatar, but don't actually change their original stats, this might explain a lot. Another idea is that during the testing phase they found defence Units too overpowered and just dropped a basic x2 or x3 secret bonus into attack rolls.

I say this because I really didn't do much combat in this game, I played a very tight game for experimental reasons, so each combat I did notice quite a lot. I lost an Elite Numidian Spearman that was attacking a Barbarian Warrior, for example, and, even stranger, a Fortified Elite Numidian Spearman was reduced to 2 Hit Points by an attacking Warrior.

I've long since stopped bothering to complain about such trivialities, they are just absurdities that don't mean much in the grand scheme of high Production outputs, but it does make me very curious about the math. The formula you posted just doesn't seem 'right' if you know what I mean. Is the formula an officially disclosed one or is it something someone made up from observation?

 

As the formula works fine i never cared to find out. I would not be surprised if there is a threat dedicated to the matter.

 

I don't see how you can claim it works fine. You might as well say "it's just a roulette wheel" and then be as comfortable with the results, happy in the knowledge that anything can happen, each upgrade just marginally shifting the odds in your favour, but not by anything quantifiable in small quantities, only quantifiable in large quantities, like 100 attempts at roulette instead of 10.

If anyone knows the thread where the origins of the formula started or how the formula was determined I'd be grateful for the nosey.

 

More than one, in fact. This has long since been thoroughly researched. And here seems to be the most useful/definitive info that I found, via a thread I found using the search string "civ3 combat results calculator".

And at risk of pointing out the obvious (Buttercup), Civ3 is indeed just rolling a die: a pRNG-outcome is simply a die-roll made via a computer processor, except that the computer can use whatever sized 'die' is needed to make the numbers come out, e.g.

Spoiler :

When the attacker has a 10% chance of winning (based on relative A/D stats and bonuses), the computer is effectively rolling a 10-sided die for each combat-round and saying "Roll 1 = defender loses 1HP, roll 2-10 = attacker loses 1HP". Or if you prefer, for each combat-round, the attacker gets 1 raffle ticket vs. 9 tickets for the defender, so the attacker has only a 10% chance of the pRNG drawing his lucky number out of its hat.

Which is not to say that the attacker can't win, just that for any given single trial that likelihood decreases as the A-D disparity increases. OTOH, the more often you repeat the trial, the more inevitable it becomes that a 'freak' result eventually occurs, and at least one freak result becomes 100% certain to occur over an infinite number of trials.

So if you're actually quoting a real battle result in your post there, then yeah: very unlikely but not impossible.

 

Is there an impossible outcome? Is it impossible for a regular warrior, say, to defeat a mechanised infantry unit, fortified in a megalopolis, situated behind a river on a hill?

 

It works very well in big numbers. If you do test runs in multiples of 10000 units, than the results are what they are to be expected. Due to the nature of the formula upgrades from say swordsmen to medieval infantry have significant effect. In the case you have given that upgrade would increase the chance of a victory from 0.433% to 1.083%. That is factor of 2.5. It still will be very little, but that is to be expected against a highly superior enemy.

 

I'm aware that a formula exists and the philosophy behind it, I'm just curious as to how the formula was created, and, again, that thread just works from the assumption that the formula is correct, with people suggesting it's taken from watching sample battles in quantity, a solution that works but might not be the actual formula. If we don't know the official formula, then how do we know if there are hidden tweaks to the formula dependent on circumstances we don't know about - such as vulnerability and risk or even drama. If their formula allows for any variable, there's no way you could 'investigate' hidden tweaks because everything could be argued to be possible, even though many results appear as if something more is happening.

The widest variable is a stack of Barbarian Warriors attacking a the Modern Infantry as you describe. I prefer to speak from experience rather than philosophy and I have experienced a hoard of Barbarian Horsemen lay siege to a Modern Armour Fortified in a regular size 6 town, and it did lose Hit Points. But Modern Armour aren't the best defenders, that would be Mech Infantry, but that just nitpicking the question, to which the answer is, the only guaranteed result is the very best 99.999999% likely to defeat the very worst in the game, everything inbetween is 'open to the Gods'.

What do you think to the idea that possibly there's a % chance that each time you create a Unit the Unit could be either Hardy, Normal or Soft? for an example of a potential hidden tweak that would be hidden in large numbers and possibly create an illusion of a formula that looks right in large numbers, but doesn't in fact give a true understanding of how combat works.

 

Oi Buttercup! I said 'mech infantry' in the first place.

I did relate on another thread the extraordinary (to me) feat of an infantry army in destroying about 35 infantry units for the loss of a couple of hit points while fortified in a city. Possibly behind a river.

It will come as news to nobody that I do not know how to calculate combat outcomes and that I play entirely by feel where such things are concerned.

 

You did indeed, lol

 

I really doubt it. It is unnecassary complex and it violate the observations made without saved random seed aktivated and deactivated aswell.

 

Well, I'm sorry, but I don't really know what the parameters are for testing with 10,000 Units vs 10,000 Units. Were they all individually produced? Were the successful Units then used a second time? Did someone reload a save 10,000 times or uniquely produce 10,000 individual Units. etc etc

 

AFAIK, the various CFCers who tested this, set up combat-testing scenarios using the Editor, ran the battles, then varied the unit parameters one by one, rerunning the same tests each time, to see how (if at all) those parameter-changes altered the battle outcomes.

Using the Editor, you can easily set up a test where each side starts with, say, 10000 preplaced Warriors (with zero unit-maintenance cost so there are no funding-difficulties), and then pit them against each other on various different terrain types and/or city-sizes (1000 battles per parameter-test should be sufficient). This would only take one 'turn' -- and you could use the results to count not just outright victories, but also HP remaining for the victorious units, promotions awarded, etc. Then for the next test you can give each side e.g. 10000 Archers and do the same thing again (or e.g. pit Archers vs. Spears). If you comb through the Civ3 utilities section of CFC, (I believe that) there is actually such a combat-testing scenario available to download -- I've read about it somewhere, but I couldn't find it quickly myself.

Given that the Civ3 program is a fixed (non-evolving) system, and given that the testers knew what (most of) the inputs were (A/D stats, hitpoints, terrain defence bonuses, etc.), and what outputs they'd measured, deriving a formula(e) which would produce the observed outputs from those inputs, would be relatively straightforward for anyone with a sufficient knowledge of calculus and statistics. Having derived the formula(e), predictions can then be made for as-yet untested parameter values, and appropriate tests set up and run accordingly. If the predicted results are actually observed in the tests (within acceptable error-limits), it is reasonable to assume that the formula(e) were/are correct.

Yes, doing all this would have required lots of time/ patience on the testers' part -- but having done so, I think their results can be considered to be reliable.

 

You don't seem to grasp what I'm saying tsj. I fully understand how averages are calculated. If, for example, there were Hardy, Normal and Soft units, then the average would still be the same as if they were all Normal. You would end-up with a probability formula which represents the average outcomes of 'normality', but it wouldn't register the fact that combat isn't entirely 'just' percentages chances.

I believe there is 'more' to it than just percentages, and showing someone a flat average of an unknown system doesn't actually tell you anything, unless the game is using those exact methods and only those methods to conduct combat.