| Author |
|
Innoble 
Wordsmith
Joined: 06 Dec 2011
Status: Offline
Points: 141
|
 Posted: 03 May 2012 at 21:29 |
Oh, while we're on the discussion of rounding and the changes. I ran into a really annoying thing with my large forest report. I saw a legion of rats in the woods, scouted it, sent the armies and ran the report through my calculator. It predicted I should be losing 1013 charioteers. Part of my calculator copy-pasted down below:
| Type |
Amount |
bonus(%) |
Casualties |
| Cavalry |
1015 |
15 |
1013 |
| 0 |
|
|
0 |
But unfortunately: This really annoys me. I cannot explain the 1 extra survivor. I know noone probably cares if you can get that close, but I kind of want to know the exact calculations they use. This must have something to do with rounding, because my calculator predicts 1013,03 casualties. If it had been 1012,99 my calculator would have said 1012 too. So somewhere in some step in the calculation, the devs have set it to round in some way, up or down, left or right, who knows. The 0,04 difference you can actually achieve with a tenth of a ratling... thats how close it is.
It is impossible to explain this deviation with the terrain-modifiers as any change to those has a 10 to 100 times greater effect. You can't make it right that way. I have other large forest reports, also on the order of 500 troops sent against a legion and those did predict exactly, so it is a rare deviation only happening when things are *just* right for it. That's typical of rounding issues. |
 |
Mara Zira
Forum Warrior
Joined: 14 Aug 2011
Location: Arkansas, USA
Status: Offline
Points: 223
|
Posted: 03 May 2012 at 20:53 |
 Innoble wrote:
Also, cavalry defending could be wrong I guess. I have it within 1-2% accuracy. |
I'm guessing that some of my results could be 1% off, too. I just thought it would be polite to let you know where we did get different results.
Innoble wrote:
You have to be careful testing though. There are tons of little things that only give a very small error such as commander division modifier and terrain modifier. You can multiply those modifiers (1,15 * 1,30) or add them (1 + 15/100 + 30/100). The difference won't be very large, but it *will* give you varying results that may lead to modifiers that are slightly off. Not saying that that's the case, but be careful.
An easy solution is to use vanilla commanders. If that's what you're doing there should be no problem. |
I'm using 0 level commanders of the same type as the troop they're leading so that as few factors as possible are influencing the tests. I think some of the changes to the combat system may be making terrain testing more difficult as sometimes the only explanation I have for the results of several tests is that casualties occasionally round up now, even for single-type troops attacking single-type troops. Perhaps I'm wrong, though, and I'm overlooking something. I didn't think that the round-up effect was going to effect single-type-troop armies. I need to re-read that dev post. 
Edited by Mara Zira - 03 May 2012 at 20:55
|
|
lokifeyson
Forum Warrior
Joined: 29 Jul 2010
Status: Offline
Points: 211
|
Posted: 03 May 2012 at 02:39 |
|
Innoble is god of the calculations, he already predicted the amount of views this thread would get ;)
|
|
|
|
Drydenn
New Poster
Joined: 04 Nov 2011
Location: The Dirty Jerz
Status: Offline
Points: 17
|
Posted: 03 May 2012 at 01:53 |
TY Innoble & Hendel! 
I shall now proceed to try staunching the upcoming flow of blood from my ears as I try to digest this.
|
|
Innoble
Wordsmith
Joined: 06 Dec 2011
Status: Offline
Points: 141
|
Posted: 02 May 2012 at 18:44 |
Spear attacking may be wrong, you're right. Thats indeed why its yellow. I havent even tried that one so it's a total guess.
Also, cavalry defending could be wrong I guess. I have it within 1-2% accuracy.
I had a hard time testing that and chose to hit some hits on rhino packs. I did notice that the defending bonus is slightly smaller than the offensive one. If you get 26% or 24% instead of my 25% that is totally possible. 25% seemed likely since I could fit most, if not all, of the bonuses in 5% increments.
You have to be careful testing though. There are tons of little things that only give a very small error such as commander division modifier and terrain modifier. You can multiply those modifiers (1,15 * 1,30) or add them (1 + 15/100 + 30/100). The difference won't be very large, but it *will* give you varying results that may lead to modifiers that are slightly off. Not saying that that's the case, but be careful.
An easy solution is to use vanilla commanders. If that's what you're doing there should be no problem.
Also, I am pretty happy that 3 of those results agree on mountains as I have done very little testing on them! That means I'm on track.
If you don't mind, share your results with me. Maybe I have some pointers or maybe you can help me out by finishing it up :)
|
|
Mara Zira
Forum Warrior
Joined: 14 Aug 2011
Location: Arkansas, USA
Status: Offline
Points: 223
|
Posted: 02 May 2012 at 18:26 |
I've been some terrain testing for the last few weeks. I haven't gotten to forests or buildings yet, and I'm still trying to refine my calculations on large hills, but otherwise I'm pretty certain of the accuracy of my results.
For plains, I have a different result than you for spear attacking (which, yes, I see you have as yellow). My results for cav defending on plains is also 1% different than yours. I did several tests, and I'm pretty certain my results are correct, but I guess I'll check them again.
For attacking on large or small mountains, only 3 of my results agree with yours.
|
|
Cerex Flikex
Forum Warrior
Joined: 11 Apr 2012
Location: BC
Status: Offline
Points: 211
|
Posted: 02 May 2012 at 17:23 |
|
Awesome, I'll have to check this out. :D
|
|
Captain Ganoes Paran
Wordsmith
Joined: 23 Jan 2012
Status: Offline
Points: 158
|
Posted: 02 May 2012 at 16:35 |
|
Great work Innoble :) thanks
|
|
Chadgod, Norman
New Poster
Joined: 10 Nov 2011
Status: Offline
Points: 13
|
Posted: 02 May 2012 at 15:08 |
HendelTheDwarf wrote:
Here's a link to a pdf version of the file - the link should work, message me if you have any problems.
http://db.tt/MRNI5xM3 |
Yup, this works.
Thanks.
|
|
Gasterix
New Poster
Joined: 30 Apr 2012
Status: Offline
Points: 12
|
Posted: 02 May 2012 at 12:54 |
|
I just learned a lot. Thanks for this.
|
|
Tweet
Advanced Attacking: a work in progress
Topic Options
