Printable View
Quote:
Originally Posted by SMZ
If you look at that line you will see ; - thats mean this line is ignored! You can easilly remove it and game will work without any problems. That line was linked with that system Dev's wanted to add, but they disabled it before release as it didnt worked. If you remove that ; - game will crash...
Yes the game will, but i think the values are representative of the hidden armour system which we cannot edit.Quote:
If you look at that line you will see ; - thats mean this line is ignored! You can easilly remove it and game will work without any problem. That line was linked with that system Dev's wanted to add, but they disabled it before release as it didnt worked. If you remove that ; - game will crash...
More important is: armour_ug_levels 0, 1, 2, 3 - this line actually says which upgrades you are using + armour_ug_models Pike_Militia_ug1, Pike_Militia_ug2, Pike_Militia_ug3 - that means model Pike_Militia_ug1 will be used for upgrade 1 etc... In model file there is a line for animation capabilities (attack offset, deffence offset etc... look at some of my previous posts ) but ! all units in vanila use the same values!
Wow, excellent work, testers. :2thumbsup:
Nice to have people find that armour upgrades work the way they should. I only hope they fix the +1 representation in the patch. (Any chance of mentioning that to Palamedes, Lusted)?
Indeed. Rushes to put this info in the FAQ and tell the other players in our HRE PBM. :charge:Quote:
Originally Posted by Foz
Or system that was ignored by Dev's becouse it was not working, or they were not able to make it work, M2TW is still RTW engine after all...Quote:
Originally Posted by Lusted
Correct, those numbers, 0, 1, 2 represent what 'type' or armour the unit has each upgrade level. and despite what Darth might say he is wrong in this case, the evidence is overwhelming that armour upgrades add more than just +1.Quote:
More important is: armour_ug_levels 0, 1, 2, 3 - this line actually says which upgrades you are using + armour_ug_models Pike_Militia_ug1, Pike_Militia_ug2, Pike_Militia_ug3 - that means model Pike_Militia_ug1 will be used for upgrade 1 etc... In model file there is a line for animation capabilities (attack offset, deffence offset etc... look at some of my previous posts ) but ! all units in vanila use the same values!
Oh, and there is absolutely no documentation about what half of the numbers in the battle_modles.modelsdb file do.
Im certainly going to talk to him about him the next time i have a chat with him.Quote:
Nice to have people find that armour upgrades work the way they should. I only hope they fix the +1 representation in the patch. (Any chance of mentioning that to Palamedes, Lusted)?
Here some test results of mine. Firing unit was peasent archers with arrows modified to have 25% accurracy all the time and only 6 volleys of ammo.
Armour Silver Upgrade, Basic 0
12
10
12
13
12
Total: 59
Armour Basic 5
19
14
17
13
16
Total: 79
Whilst it tells us somthing weird, (Silver armour upgrade is more effective than 5 armour in the EDU file), Regardless it proves beyond ANY shadow of a doubt that armour upgrades ARE NOT just +1 per level.
Of course it dosen't make sense, with base, (i.e., what a unit has before any upgrades), armour values for a given level varying by 2 or 3 points for almost all values and some tests relating to EDU's modified with the sheild fix it's quite clear that each level of upgrade ADDS to the existing Armour value. All they have to do their is specifiy how much each level of upgrade adds to a units base value. So level 1 would be +4 for example.Quote:
Just imagine - game has specific file for units, there are values for protection of unit,but it is not used, instead of this there is some other table where upgrade 1 means armor 4, upgrade 2 armor 5 etc... That dont make sense.
What was the unit in question?
Scots Pike Militia.
WOOPS, sorry about that missing peice of data.
Quote:
Originally Posted by Carl
Again: thats not a proof! What was arrow attack value? what were resoults against unit with base armor 2?
I'm sorry I lied earlier, but this is really the last time I repeat that:Quote:
More important is: armour_ug_levels 0, 1, 2, 3 - this line actually says which upgrades you are using + armour_ug_models Pike_Militia_ug1, Pike_Militia_ug2, Pike_Militia_ug3 - that means model Pike_Militia_ug1 will be used for upgrade 1 etc... In model file there is a line for animation capabilities (attack offset, deffence offset etc... look at some of my previous posts ) but ! all units in vanila use the same values!
armour 7 flesh is the same thing as upgrade 3 breastplate
I'll admit without being prodded that I know very little about modding, coding, etc - all I know is the results... I don't know how it works - I just know it works... ppl can argue about HOW the game does it til they're blue in the face, and I'll have to just keep referencing them back to the first post which now has results from 4 people and something like 100 test runs all leading to one inevitable conclusion... it's not an anomaly, it's not an isolated case - it's a simple test which can be repeated by anyone over and over and over, and they will get results consistent with those already shown
Good to hear.Quote:
Of course it dosen't make sense, with base, (i.e., what a unit has before any upgrades), armour values for a given level varying by 2 or 3 points for almost all values and some tests relating to EDU's modified with the sheild fix it's quite clear that each level of upgrade ADDS to the existing Armour value. All they have to do their is specifiy how much each level of upgrade adds to a units base value. So level 1 would be +4 for example.
And JaM, just read through this thread gain, more than enough evidence has been posted by me, SMZ, Carl and Foz. If that is not enough to convince you, nothing is.
I didn't do the test but I don't need to. The Scots Pike Militia have a LOWER loss rate with silver armour than with 5 armour, Higher armour=better missile defence, their no WAY the Scots Pike Militia could suffer less dead from the attack AND have a lower armour value at the same time, it's IMPOSSIBBILE, unless armour dosen't ork right against missile fire, and a low armour value actually provides the best missile defence.Quote:
Again: thats not a proof! What was arrow attack value? what were resoults against unit with base armor 2?
Lastkly, having a Higher attack than your opponnents defence DOES NOT instantly give a 100% chance of a kill. According to others in the know, the values range from 0-64 to +64. So peasent archers will not have a high enoughj attack (at 5), to get the maximum possibbile value,. which is where 100% kill chances per hit come in.
oh yeah, and if you're doing any tests, make sure you clean up the file for at least that particular info first
given how inconsistent the file was, I can't remember if scots pike had a 5 a 6 or a 7 for the second upgrade number, before I cleaned mine up - did you already check that beforehand Carl? it might explain why they did better than 5
and as I said, I don't know much about coding - but if the ';' character says a line is to be ignored, then how does the last line: ';unit_info' still work? if I change a value there it will show up in game in the descriptions... so it would seem to me that lines prefaced by ';' do still have an effect =/
Wait, what? Changing the unit info will result in changes ingame?Quote:
and as I said, I don't know much about coding - but if the ';' character says a line is to be ignored, then how does the last line: ';unit_info' still work? if I change a value there it will show up in game in the descriptions... so it would seem to me that lines prefaced by ';' do still have an effect =/
Guys, i will post my tests soon. My point is: Unit with armor 0 will take much higher looses than unit with armor 1. its because any attack value will overmatch armor 0 by extreme margin in mathematical formula. It is not that big margin against armor 1 or 2 or3 etc... Simple mathematic.
All you need to test, how resistant will be unit with armor 1 leather in comparation to armor 4 leather.
it changes what is displayed... maybe I'm wrong, I'll check - I was certain a second ago, now you've got me wondering - but I was positive that the unit_info line determines what text is displayed for the units numbers
Quote:
Originally Posted by SMZ
;unit_info dont work too, they wanted to implement different system, but didnt worked. same like upgrades.
Your wrong, i just tried it myself. I was worried for a second as a ; before a line should mean it is not read.Quote:
it changes what is displayed... maybe I'm wrong, I'll check - I was certain a second ago, now you've got me wondering - but I was positive that the unit_info line determines what text is displayed for the units numbers
That test DOES NOT MATTER AT ALL.Quote:
My point is: Unit with armor 0 will take much higher looses than unit with armor 1. its because any attack value will overmatch armor 0 by extreme margin in mathematical formula. It is not that big margin against armor 1 or 2 or3 etc... Simple mathematic.
All you need to test, how resistant will be unit with armor 1 leather in comparation to armor 4 leather.
It is IMPOSSSIBILE for a unit with less than 5 armour to suffer LESS kills than the same unit modified to have 5 armour. That what my tests show, it dosen't matter what the diffrance between 0 and 1 is. what matters is the diffrance between 5 armour and 2 silver on Scots Pike Militia, the results CLEARLY show that 2 silver is BETTER than 5 armour. T
This CLEARLY shows that Silver Armour MUST give Scots Pike militia a BETTER armour value after the upgrade than 5.
Not just that, but the diffrance in overmatching is a mere 1 point diffrance betwen 1 and 0. the maximum overmatch is 64, so clearly 4 and 5 respectivly can't be that far apart or the uper end of the 64 scale would be meaningless. This clearly isn't the case.
apologies JaM & Lusted, I was wrong about the ';' lines - just checked, I was confused on that
Do armor upgrades work correctly in conjunction with base armor values though?
All the testing has been armor X vs armor 0 with upgrade Y but nothing like armor X vs armor Z with upgrade Y where X should equal Y+Z.
It could well be that the armour upgrade could override the base armor value given for example instead of adding onto it.
Afterall, in late game battles where units may have same stats but only 1 upgrade difference, the kill difference isn't all that large.
No carl did the tests. A unit has the armour upgrade for each level added onto their base.Quote:
It could well be that the armour upgrade could override the base armor value given for example instead of adding onto it.
Thats exactly what i did in my tests. Billmen with 2 armour upgrades should have an armor rating of 5. Berdiche Axemen unupgraded have a defence of 5. They both get the same results v peasant archers so both must have armor values of 5, despite the billmen unit card showing armor of 2.Quote:
but nothing like armor X vs armor Z with upgrade Y where X should equal Y+Z
Ok, i finished with my tests. What i found? There IS something behind armor upgrades. I can definitelly confirm, that armor upgrades adds a bonus to the upgraded unit. To me, it looks like faulty programing. (why they created two systems for upgrade? to make players more confused?) Results of my tests are clear. There is no armor upgrade table, where upgrade 1 = armor4, leather.
In my tests i used PointBlank EDU, because he uses more armor upgrade levels than vanila (12).
I got exactly same results when i set armour_ug_levels 0, 1, 2, 3 and armour_ug_levels 7, 8, 9, 10
jumps in bonus between upgrades were - 20% (average casaulties)
Ah, I didn't see your post on page 2.
About your tests though, how come armor 4 takes significantly more casualties than armor 5 in comparison to a difference of 2 points between say 5 and 7 or 7 and 9?
Also, for some of the tests your variance is quite high....
Its good to see that they also work together with an armour set value though.
My guess is the level 1 armour type upgrade does not provide as big a armour value boost as level 2 armour type upgrade.Quote:
About your tests though, how come armor 4 takes significantly more casualties than armor 5 in comparison to a difference of 2 points between say 5 and 7 or 7 and 9?
To me it looks that armor upgrade 3 = armor upgrade 2 + armor upgrade 1
Armor upgrade 2 = armor upgrade 1 + armor upgrade 1
Here is my table:
Armor upgrade levels 7, 8, 9, 10 - Point Blank - Realcombat mod EDU
armor upgrade 7 kills :49,59,58,48,51 - 53 average
armor upgrade 8 kills: 46,48,36,46,39 - 43 average
armor upgrade 9 kills: 34,33,29,35,33 - 33 average
armor upgrade 10 kills: 23,22,18,26,25 - 23 average
Same results when i used Armor upgrade levels 0, 1, 2, 3
i'm starting to think i bought this game too early... They told us, this game will be much more moddable than RTW. instead of that, wee have a lot of backgroung hard coding that prevents any own modifications...
Well I'll copy and paste my observations from my post at TWC so to maximize the probability that people more expert than me can give some hints.
I presume maybe there's another formula which take in account something that we don't actually know that modifies armor AFTER upgrades are calculated and that new value isn't the one shown.
To be clearer:
int base_armor;
int armor_upgrade_level;
int upgraded_armor;
int unknown_factor;
int definitive_armor;
Formula 1
upgraded_armor = base_armor + armor_upgrade_level; THIS IS THE VALUE SHOWN IN THE UNIT INFO
Formula 2
definitive_armor = upgraded_armor + (or * or / or whatever) unknown_factor;
THIS IS THE VALUE USED FOR COMBAT CALCULATIONS
This "unknown value" can be linked to anything. It could be weather, it could be linked to the era from which the unit comes, making old units less lethal when confronting units of following eras (this because otherway low lethality would lead to looooong and boring battles early in the game). It could be a fixed factor linked to the fact that the unit is player controlled or AI, it could be a mix of everything with maybe also some randomization inside just to give different battle outcomes everytime a battle is played.
And if we really have a fixed 40% chance of survivability to successfull hits, it seems to me that it is a huge value of variance in statistical terms considering that fights are really shorts in time terms. But I'd let this to say to statistical expert as I'm not one.
All considered we really DON'T KNOW which value is used in combat unless some nice CA programmer will reveal us the arcane matter.
Note entirely true, if the "unkown factor" I was talking about is used for every combat then it will be a sort of "fixed variance" while we can play with the rest of the stats to modify the final outcome.Quote:
Originally Posted by JaM
Of course if the weight of the modificable stats is not the most significative in combat calculations then the game is unmoddable, the armor importance is completely depleted and all that counts is the number of the troops.