View Full Version : I need to ask a favor..
Katasaki Hirojima
05-20-2001, 07:38
Ok,I stopped playing Shogun for awhile cuz I'm more in the mood for European musket battles ALA Cossacks.S'why I haven't posted in awhile.
Anyway,that favor,I need the mathmaticle system for how the computer determines weather a offensive unit makes a hit or not.Not the bonuses,just the core thing.I'm going to try and enlighten the darker world of RTS with something besides the ancient Hitpoint system..*shudders*
------------------
"I maintain none the less that Yin-Tang Dualism can bo overcome.With sufficent enlightment,we can give substance to any distinction:Mind without body,north without south,pleasure without pain.Renember,enlightment is a function of will power,not of physical strength."-Shang-ji Yang,essays on mind and matter.
Anssi Hakkinen
05-20-2001, 14:47
As I couldn't find this in any other thread, thought I'd re-post it here. This was originally posted at the Community forum by Uberuce-san, included a huge list of various bonuses which I omitted at your request, and was the subject of some debate I can post also, if you like.
Quote I've read a few variations on these, like I've read Naginata as having 1 armour rather than 6, but they don't sound outrageous.
It doesn't mention the bonus for attacking uphill or downhill, nor from improved weaponry. I assume it's the same as armour, one point for each level, so an attacker with legendary weapons has a 60% better chance of killing his opponenent.
Nor does it actually say for sure what the calculation for the attack/defence factors is. I assume it's 20% per point, so a Yari Samurai attacking a Naginata of identical honor/fatigue/armour/weapons has (0. ^6 chance of killing him. The way the numbers would have to work is for it be a successive multiplication rather than an addition/subtraction, or else a Yari Samurai has a -20% chance against a Naginata, which clearly isn't true. I've not been playing long, but I've never had Naginata come through a Yari assault unscathed, I always lose at least a couple of men.
Anyway, 0.8^6 is ~ 26%, that sound okay, anyone? That's 26% of whatever the base chance of victory is, which I assume is 50%, that is, two Samurai Archers duking it out have a 50% chance of dying. Hence, Yari Samurai vs Naginata gives a 14% chance of walking home to the boys with the spears.
That said, for say a No-Dachi attacking, you'd have a (1.2^5) chance against Samurai Archers, so 248% of 50% is 124%, which should mean zero casualties to the No-Dachi. That sounds like bollocks to me.
What sounds more feasible is the attack bonus takes 20% off the 50% base chance of not winning, which is exactly the same calculation as for defence, in that it leads to diminishing returns.
The equation for the attacker killing the defender would then be 1-{0.8^(attack factor)*0.5}
So for our No-Dachi Vs Samurai Archer showdown, we have (0.8^5 = 0.32)*.5 = 0.16, so a 84% chance of victory for the nasty bladey ones.
Let's take an extreme example: A fresh 4 honour No-Dachi from a 4 honour unit commanded by a 3 honour general, armed with legendary weapons, charges in wedge formation into the rear of a totally exhausted 0 honour Musketeer unit in wedge formation, routing on a bridge, led by Mr Bean, a -1 honour general
If you run through the list, you get an attack factor of +36, which gives a 99.9997%
chance of the No-Dachi soldier winning.[/QUOTE]Maybe the kind people at the Editing / Mods / Patches forum (http://www.totalwar.org/cgi-bin/ubb/forumdisplay.cgi?action=topics&forum=Editing+/+Mods+/+Patches&number=5) are able to give you more details.
------------------
"4. Rei: right action--a most essential quality, courtesy."
[This message has been edited by Anssi Hakkinen (edited 05-20-2001).]
Katasaki Hirojima
05-21-2001, 09:27
THank you very much,this helps lots.
------------------
"I maintain none the less that Yin-Tang Dualism can bo overcome.With sufficent enlightment,we can give substance to any distinction:Mind without body,north without south,pleasure without pain.Renember,enlightment is a function of will power,not of physical strength."-Shang-ji Yang,essays on mind and matter.
During the demo-period I think 007 (or possibly Anjin-San) made a cool little program that calculated all of that stuff + if I still have it on my hard drive I'll put it up on the .org and post the link. Also, the Prima Strategy guide has some tables which relate to various modifiers for calculations, and so on.
Okay, here's the link: STW Battlefield Calculator (http://www.totalwar.org/files/upload/shogunbattle.zip)
Someone named iAlbert made this + it may not translate to the full version as it was definitely made in the pre-release period.
(File Size=164KB)
High Voltage
05-21-2001, 11:08
Posted by Anssi Hakkinen:
"If you run through the list, you get an attack factor of +36, which gives a 99.9997%"
Cool so if my muskets are routed they have a .0003% chance of winning! That would be like rolling for a 9 on a 10,000 sided dice http://www.totalwar.org/ubb/smile.gif I would like to see that .0003% chance be calculated and actually happen. Wouldn't it be strange if the routed muskets came up victorious. Of course then the player with muskets will be accused of cheating for god knows how long.
The only formula we have is attack - defense + bonus = probability to kill, and that each point increases or decreases the probability by 20%. We don't know what the basic probability represented by 0 is. Applying the formula for honor 0 units, YS vs Nag with no bonus it's 2 - 6 = -4 for YS and 0 - 3 = -3 for Nag. At -3 the probability for the Nag to kill a YS is 60% below the base probability, and for the YS to take out a Nag it's 80% below. With what looks like an attack/defend cycle of maybe 4 sec, I don't think the base probability is 50%. If that were true, H0 SA with attack and defense = 0 would annihilate each other within 20 seconds in hth combat. I would estimate the base probability to be something like 20%.
What sets STW apart from games that represent damage by hit points is the use of statistical probability to determine individual kills within a larger unit. The reason this works is that there are a large number of individual events which smoothes out statistical error and gives reasonable results. You can still think of a unit as having 60 hit points, but the effectiveness of the unit diminshes as it looses men. In the RTS' that I've played, the effectiveness of a unit remains constant until it's destroyed. I also haven't seen fatigue and morale modelled which also sets STW apart, and you have terrain effects, weather, formations, flanking, etc.
Katasaki Hirojima
05-23-2001, 08:53
Heh,I KNOW why they don't release the base value of 0 too,one simple reason,so they can keep the idea alll to themselfs.Really though,what if Edison was the only person that chould build lightbulbs?What if China was the only nation that was allowed to use newspaper blocks??Grrr..
But anyway,before I go on a rant about my hatred of certain copyrights,I wanna say thanks for all this info.I guess it is rather important to understand how this game works.If anyone wants me to continue my rant about what I think about STW keeping its combat system to itself,say so.
BTW,check Cossakcs out,it allows you to use formations.And up to 4000 men in your army...IF you ever get that many since most games end before you have 1000.(And thats 6 hours..)They also implemented some terrain advantages and tryed to get morale in but it didn't make the cut.
What I want is for GSC(the makers of Cossacks) to adopt this combat system and make the formations and realism idea there getting at comlete.THey're a ukraineian based which is pretty cool and are really grass roots,they come to the Heaven Forum and talk to us about things and acctually implement what we want into the add on,which is cool.I don't think they've played Shogun,though.
------------------
"I maintain none the less that Yin-Tang Dualism can bo overcome.With sufficent enlightment,we can give substance to any distinction:Mind without body,north without south,pleasure without pain.Renember,enlightment is a function of will power,not of physical strength."-Shang-ji Yang,essays on mind and matter.
ShaiHulud
05-24-2001, 03:08
Hope this isn't blowing smoke, but.... I recall (can't say from where now) that 0 base is 50% and adjustments are 10% increments. So, a Nag with armor value of 5 gets a defense of 100% that is then reduced by the attack percentage of its opponent. Also, a Nag with attack of 0 is 50% to kill reduced by the armor value of its opponent.
A unit with the highest grade armor/weapons would get 30% added to their defense/attack value. Morale isn't a factor in kills, just whether or not losses cause a rout.
------------------
Wind fells blossoms, rain
fells steel,yet bamboo bends and drinks
vBulletin® v3.7.1, Copyright ©2000-2025, Jelsoft Enterprises Ltd.