The Cossack: The Russian Panzer

[barbeerhj]

l O l

Maybe, one could make a factory of gamers who are to do nothing but see the one HP warrior beat the 4 unit MA army, like you said, it could happen.

[Kuciwalker]

Actually, assuming you lived 100 years and did such a battle once per second for your entire life, it would take roughly forty hundred thousand trillion trillion trillion lives

[barbeerhj]

Wow

But Indeed, I would live to see the day 4 MA's inside an army got routed by a barbarian with 1 HP.

Another thing, how did you calculate that?

[Kuciwalker]

I took 1 divided by the probabity of it occuring (giving the average number of trials you'd have to make for it to occur) and then divided that by the number of seconds in 100 years

[joncnunn]

That's if true randomness was used instead of the RNG.

RNG's are notorious for being the least accuracute for both very rare and very common results: (0% < x% < 5% or 95% < x% < 100%)

Originally posted by Kuciwalker

If you play long enough, you'll see an unfortified barbarian 1hp warrior on grassland defeat a full hp 4-MA army bonused by the Mil. Academy while playing chieftain. The chances, however, are merely 1.3*10^-51.

[joncnunn]

There's not that many numbers in the RNG pool.

Originally posted by Kuciwalker
I took the chance of the Warrior taking off one hitpoint from the army, and raised it to the 20th power.

[barbeerhj]

Joncnunn

How do you know?

-Ron

[Kuciwalker]

Originally posted by joncnunn
There's not that many numbers in the RNG pool.

Depends on what they're using to store the random number. If they're using 20+ bytes, then yes, it is

[Petrus2]

A few thoughts...

1. If you DID experience such an occurance, or any of the unlikely events, it could benefit you, don't forget that.

2. It's a game, what is really the big deal??

3. Have you never, never ever, not ever, reloaded when you didn't get the combat result you wanted? If you have, then how big a deal is the oddball result, since you are likely to reload then anyway. (or get the goody hut result you want, map start you want, etc)

4. If your war plan is devastated by an oddball combat result, then your warplan was doomed to failure in the first place.

Just my thoughts on a topic where it seems that players sometimes get unnecessarily upset when things don't go along with what they expected.

[Theseus]

You guys have gone waaaay into the deep end of the pool of silliness.

Uhhh, back on-topic:

In AU 601, I just got put down by Nathan (nbarclay) coming at me with 50+ Cavs... and that's with the AU Mod, where we have notched them down to 5a.

Damn... if they had blitz too???

[Solomwi]

I haven't reviewed Nathan's battle logs with this in mind, but we've got a ready made source of information as to just how many of his cavs would have been up for a second attack.

[Kuciwalker]

[QUOTE] Originally posted by Theseus
You guys have gone waaaay into the deep end of the pool of silliness. /QUOTE]

I actually came up with that example a long time ago in the C3-general when someone was *****ing about a spear vs tank-type scenario. It's a useful example.

[joncnunn]

Originally posted by barbeerhj
Joncnunn

How do you know?

-Ron

I'm a programmer by trade.

RNG algorithms don't actually store a set of longs, but it instead yields which bit pattern of a specific length to grab. (Neither does Free Cell actually store in memory those thousands (or hundreds of thousands with XP) of starting hands, the number entered is enough to compute the starting hand.)

There is direct programming support in some langagues for 32 bit and others for 64 bit, and so one of those numbers is most likely the size of the RNG pattern. RNG isn't important enough to spend the effort to write a software implemention of a 256 bit RNG, let alone a 32 byte long one, and in addition, the larger the set, the slower it's going to take, and it's more important that an RNG take almost no time to run than keep probababilties correct.

[joncnunn]

Originally posted by Kuciwalker


Depends on what they're using to store the random number. If they're using 20+ bytes, then yes, it is

They definately aren't using 20 bytes. It's going to be an even power of 2. Most likely 32 or 64 bits. Come to think of it, I wouldn't completely rule out Civ III using a mere 16 bits given a report on how civ III came to be, which is probably the length it was in Civ II and SMAC. Civ IV is most likely to have a 64 bit RNG.

[barbeerhj]

Back to the Cossack

I believe that the strength of the blitz attack is more in counter-attacks then it is in offensive assualts against well defended positions.

Seeing as how a cavalry with 6 attack has about a 44% chance of defeating a fortified musketman inside a town. Also, he has a 35% of defeating a fortified rifleman inside a city (32% if inside a metro).

With such amazingly low odds of winning againsts defenders, it would seem unlikely that the blitz move could really come in handy unless you have the advantage of going against pikeman or even spearman in the mid-late middle ages.

But, when you consider the ease at which you can decimate enemy field troops, especially their cavalry and musketmen, the blitz becomes a more attractive offer.
I mean, being able to hit up an enemy cavalry group 3 times with on unit can do some seriose pwning unless the lovely RNG gives you a stinker.

-Ron