battle calculator

[KGB]

LPhillips,

I don't think we would ever see KGB's worst described scenario, where one simply clicks on an enemy stack and gets a free Battle Advisor report, but it was suggested in an earlier post.

This was in fact how the War2 and DLR battle calculators worked. You clicked on a stack and it showed the potential outcome of fighting that stack. This is likely why such a suggestion was made that it work this way. This type of calculator is completely renders fight order and scouting useless. It also renders blessing, magic items and units with variable strength/hits useless as well because you the calculator does it all for you.

KGB, It is still my firm opinion that a varied dice system is not an effective solution. If we hide the unit strengths, isn't that enough?

You are right that it definitely can't be the *only* solution. It can merely be part of an overall solution to render calculators useless. But in the short term (version 0.8) it can be added with about 5 minutes of work. That's its primary advantage. Long term, it could be a game option, fixed dice or random between 80-200.

If all the numbers are hidden AND variable strength/hits units are introduced then hiding the unit strengths/hits will probably be enough. But you'd definitely need variable strength/hits units (Mercs for might always come with an extra +2 strength) as otherwise only blessings and hero values will be non-trivial to determine.

Fordus,

Biggest winning chance isn't always best, and often it's 100% regardless of the fight order.

True. But when fight order does matter in say 10-20% of the battles why should someone automatically get the best outcome due to a calculator?

Calculator can give "exact" probability as long as it knows value ranges (and preferably probabilities) of unknown variables. Or it can give range based on different combinations of unknown variables. However decision making won't differ all that much from basic 75% chance of winning, or 89% chance.

If decision making doesn't differ much between a basic 75% chance of winning and an 89% chance of winning then why do you need the calculator?

It's clear from your and other players arguments here that exact numbers do matter or players at least believe they matter. Otherwise no one would ever ask for a calculator or care if anyone else used one. So for fairness sake you either have to spoon feed everyone, dumbing down the game by giving up features or else you have to prevent everyone from getting the exact numbers thus making calculators less useful than actual player experience.

KGB

[LPhillips]

KGB,
Maybe it's paranoia but the randomized dice proposal strikes me as a huge negative. I prefer a steady general predictability, with just enough variance thrown in to make the game entertaining. The idea that odds might randomly swing between favoring one unit composition or another is nothing but negative to my mind. The dice are a fundamental balance standard which I would never want to become fluid. For me, it's as far as I can go to concede that this should be considered as a last-ditch option at the time when (if) simulator cheating becomes otherwise uncontrollable. As of right now, it's not even prevalent (or present, as far as I know).

A great deal of discussion has centered on assuming the worst scenario from the start. Doesn't it make sense that such a scenario would be unlikely to develop if some simulator capability were provided in a learning capacity? Sort of like "give the people beer so they don't brew poisonous moonshine."
This approach accomplishes two major goals:
1) It sets the boundary for simulators in such a way that there can be no mistake about what is considered correct use compared with cheating, so that only the most unscrupulous or desperate players will actively attempt to violate the boundary. These people tend to be of lesser capability than others (smart people don't cheat often), so they won't be doing anything but ruining their own experience.
2) It provides a simulator in the only capacity which is acceptable, so that major motivations for creating and maintaining one are removed. Most people with the capability to create a simulator would not do so if the legitimate needs are already supported.

[hungrytales]

I suspect English isn't your first language because it appears you didn't understand what I wrote.

Ouch. That, KGB, is hitting below the belt . So you decided to use your native English superiority as an argument in a discussion?

Yeah you got me there, English is not my first language. But for your information, I estimate my English skills at roughly CPE level (I hold CAE, but it was a long time ago and now I'm significantly better) and I assure you (you are probably aware of that, but it needs to be stressed) usually reading skills come a notch better than writing ones. It's a result of generally having more opportunities to read than write in a language you learn. Therefore, if there's some misunderstanding between us it'd be more reasonable to assume it's more on your part. From my experience when writing in a second language it's especially hard to handle emotions. Something intended as light and benign can come across as heavy and insulting :/. And I have this knack for maybe overusing emotions in my style, so I gather that'd be the case here and if it is - I'm sorry.

All in all I've been able to read English prose in original casually for some years now. For instance, right now I'm finishing the brilliant Surely you're joking, Mr. Feynman! and I'm fairly sure you'd agree that if I can follow him, I shouldn't have problems following you .

[KGB]

HungryTales

I am not trying to hit below the belt. I assure you, your English comprehension is infinitely better than my comprehension of your native language whatever it may be.

I honestly wasn't sure if you understood what I wrote since it's impossible to know how well someone understands written English (having never learned another language I always assume it's easiest to pick up the spoken part then the written part since you often don't have context like you do with speech). Especially when the discussions here often turn technical instead of just being common words/sentences.

KGB

[Negern]

I haven't read the thread, but been wondering for a while why there's so much fuzz about this calculator. It's not hard to memorize a couple of battle%-rules and be able to get even big and pretty messy battles somewhat predictable.

For not so detailed info (generally good for small battles/battles where units on both sides are pretty much as strong): count the total str of the stack and compare. Do you have the same str as opponent? 50/50. Half of opponent is half of 50/50 = 25/75. etc.
Most of the time, you have to keep over 1/4 of the str if you want to stay over the 10%.

Example for somewhat detailed info (needed for counting bigger battles, when units that differ greatly):

25 10-str units in a 0 defense city vs. 8 attackers with 50 str and 5 fear

Looks like 125 vs 400, which - I believe - would be somewhere around 11-12% if it was a small battle with pretty much as strong troops. But not this time, for when defending units are that weak (5 str) and the attacking that strong (50 str) every defending unit will have much less than 0.5% chance to win vs the 50%

In steps:
1. 50% (same strength has these odds)
2. 25% (half)
3. 12.5% (third)
4. 6.25% (quarter)
5. 3.18% (fifth)
6. 1.6% (etc)
7. 0.8% (etc)
8. 0.4% (etc)
9. 0.2% (etc)
10. 0.1% (etc)

I make it 50^-10 for every 5 str to kill a 50 str because 5 is only 1/10 of 50.

Thats why a 75-unit or two often can empty a city filled to the brim, even though the units gathered there togheter have a str many times higher than that of themselves. But even though, in raw str-points, a 75demon has a bigger lead vs a 20HI then a 50giant has vs a 5scout, the giant has a higher %-lead and therefore better chances of winning.

It's possible that my "detailed info"-equation is a little flawed and not so user friendly (try with mixed numbers, unit bonuses (e.g. vs flying), and you will see), but of the few times I've used it, most have worked fine.