Armor Upgrades Work Properly

[SMZ]

I don't know if this debate was already settled previously, but if not - I've just done some tests which confirm that the Armor Upgrades function properly, it is merely the numbering in the unit descriptions which is incorrect.

For those unfamiliar with the question at hand. Armour is in 7 levels.
Lvl 0 = Flesh, 0 defense
Lvl 1 = Leather/Padded, 4 defense
Lvl 2 = Light Mail/Chain, 6 defense
Lvl 3 = Heavy Mail/Breastplate, 7 defense
Lvl 4 = Partial Plate, 8 defense
Lvl 5 = Full Plate, 10 defense
Lvl 6 = Advanced Plate, 11 defense
Now, this means that when you upgrade the armour on a light unit - from say lvl 0 to lvl 1, you should notice a 4 point increase in defense. However, the ingame description will only show you a 1 point increase. Thus, the question becomes, is the proper armour value being used, and not reported correctly - or is the game using the value shown and causing units to not benefit fully from armor upgrades?

In all tests I used Normal unit size, Medium difficulty, a single unit of player controlled Desert Archers firing upon a unit of computer controlled Pikemen, and the Grassy Plains map. Nothing was clicked, I de-selected Skirmish mode before starting the battle, then allowed the archers to fire upon the pikemen until melee was reached, at which point I exited battle and recorded pikemen remaining...

Test 1: Desert Archers vs. unaltered Pike Militia (Defense vs Arrows: 0)

Battle I: 21, Battle II: 31, Battle III: 33, Battle IV: 30, Battle V: 22, Average: 27.4

Test 2: Desert Archers vs. gold armour upgrade Pike Militia (Defense vs Arrows Listed: 3, Proper Defense vs Arrows: 7)

Battle I: 41, Battle II: 48, Battle III: 50, Battle IV: 45, Battle V: 41, Average: 45

Difference: 17.6

Test 3: Desert Archers vs. unaltered Tercio Pikemen (Defense vs Arrows: 0)

Battle I: 24, Battle II: 22, Battle III: 21, Battle IV: 16, Battle V: 29, Average: 22.4

Test 4: Desert Archers vs. silver armour upgrade Tercio Pikemen (Defense vs Arrows Listed: 2, Proper Defense vs Arrows: 7)

Battle I: 40, Battle II: 46, Battle III: 49, Battle IV: 40, Battle V: 47, Average: 44.4

Difference: 22

Test 5: Desert Archers vs. unit file altered Pike Militia (Defense vs. Arrows: 3)

Battle I: 37, Battle II: 36, Battle III: 38, Battle IV: 37, Battle V: 28, Average: 35.2

Difference from Test 1: 7.8
Difference from Test 2: 9.8

Adding three points to the Pike Militia directly gave results in-between the unaltered Pike Militia and the upgraded Pike Militia. Clearly then, the upgrade raised the armour of the Pike Militia by 7 points, not 3. Also, it is plain that upgrading the Tercio Pikemen resulted in an additional 7 points of defense, not 2. Two points of defense is not sufficient to explain an average difference of 22 men in the time it takes a pike unit to close to melee.

The results are clear to me. If anyone disagrees, please do your own tests and post them.

FURTHER CONFIRMATION:

1 unit of pike militia (player) vs 1 unit of peasant archers (computer)

normal unit sizes. grassy plain, 3 sets of 10 tests. archer's ammo capacity reduced to 10. pike militia taken out of spear wall and defensive stance. clear weather for all tests.

Set 1, unmodified pike militia with 3 armor upgrades. (armor stats says 3, should be 7)

Men Remaining
test 1= 57, test 2= 50, test 3=52, test 4= 50, test 5=50, test 6=53, test 7=46, test 8=54, test 9=56, test 10=51.

average men remaining= 51.9

Set 2, modified pike militia with armor stat set to 8, no armor upgrades. (armor stat 8)

test 1=55, test 2=55, test 3=61, test 4=57, test 5=56, test 6=54, test 7=54, test 8=55, test 9=57, test 10=63

average men remaining= 56.7

a difference of 4.8

Set 3, modified pike militia with armor stat set to 3, no upgrades. (armor stat 3)

test 1=47, test 2=46, test 3=46, test 4=44, test 5=45, test 6=46, test 7=44, test 8=48, test 9=46, test 10=47

average men remaining 45.8

which is 6.1 men less than the unmodded upgraded pikemen.

clearly the upgraded pikemen are performing better than their stats reflect

Alright, I've set out to settle this stupid debate. In the first test, we have longbows vs italian militia modded to have 7 armour. The second will feature italian militia (copied a second time into the file w/ a tweaked name) modified to have 0 armour and given 3 armour upgrades, which the OP has speculated should confer them the same 7 armour the first unit has. Results follow.

7 armour unit losses:
35
36
36
31
30
---
168

3 upgrade, 0 armour unit losses:
27
36
35
35
34
---
167

As anyone can readily see, this result cannot in any way be explained if the armour upgrades only grant +1 each, as the unit stat sheet says they do. The units suffered practically identical losses, which is extremely strong evidence that the OPs assertion is the correct one: the armour upgrades function exactly as they are described by type (4 for leather, etc), and not as the stat sheet reports them as simple plus 1s for any given upgrade. As it turns out this makes the leather shop a ridiculously important upgrade since it adds a full +4 to all applicable units.

All battles Very Hard difficulty, Spanish plains at midday with clear weather. All done with unmodified M2Tw with me controlling the non-missile unit. All tests done v peasant archers.

Units tested were Billmen, Berdiche Axemen, Heavy Billmen.

They have the following armour levels:

Billmen:
Flesh(0)
Leather/Padded(4)
Light Mail(5)

Berdiche Axemen:
Light Mail(5)
Heavy Mail(7)
Partial Plate(9)

Heavy Billmen
Heavy Mail(7)
Partial Plate(9)

So there is an overlap between the different armour levels to see whether armour upgrades do give more than the +1 indicated on the unit card.

Test results(number is men remaining):


Billmen:
Flesh(0) 3 8 4 6 4 Average 5
Leather/Padded(4) 22 14 15 18 26 Average 19
Light Mail(5) 34 28 41 38 36 Average 35.4

Berdiche Axemen:
Light Mail(5) 40 37 33 30 34 Average 34.8
Heavy Mail(7) 62 60 55 58 59 Average 58.8
Partial Plate(9) 74 81 85 76 79 Average 79

Heavy Billmen
Heavy Mail(7) 59 64 63 56 54 Average 59.2
Partial Plate(9) 82 78 82 85 79 Average 81.2

So as you can see Berdiche Axemen unupgraded, and Billmen with level 2 upgrade have basically the same amount of men left, which is what should happen as both have the same armour level.

Same with Berdiche Axemen level 1 upgrade and Heavy Billmen, and Berdiche Axemen level 2 upgrade and Heavy Billmen level 1 upgrade.

So armour upgrades are working properly, just an incorrect value is shown on the unit cards.

Here some test results of mine. Firing unit was peasent archers, target was scots pike militia, 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 Kills: 59

Armour Basic 5

19
14
17
13
16

Total Kills: 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.

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.

[pevergreen]

Now, this means that when you upgrade the armour on a light unit - from say lvl 0 to lvl 1, you should notice a 4 point increase in defense. However, the ingame description will only show you a 1 point increase. Thus, the question becomes, is the correct armour value being used, and not reported correctly - or is the game using the reported value and causing units to not benefit fully from armor upgrades?

If its being reported wrong, thats great. But experience from myself is that it's being reported correctly.

[SMZ]

well... the tests I just did are right there... I think the results are pretty clear - what numbers have you gotten that make you think differently?

[Carl]

Random point SMZ, Terrico pikes have 0 armour normally, so they have no missile defence without upgrades. so Silver Armour Terrico pikes should have a lower armour value than Gold Pike Militia, a much BETTER way to check would be to re-run the first Pike Militia test, but with them altered inn the EDU file to have 7 armour defence.


Lastly, your armour typ listing is wrong, it should be:

Lvl 0 = Flesh, 0 defense
Lvl 1 = Leather/Padded, 4 defense
Lvl 2 = Light Mail, 5/8 defense
Lvl 3 = Heavy Mail, 7/8 defense
Lvl 4 = Partial Plate, 7/9 defense
Lvl 5 = Full Plate, 9/10 defense
Lvl 6 = Advanced Plate, 10/11 defense

In fact I don't think their IS a standered value for ANY of the armour levels, they all have variances, just look through the western Knights at the top of the EDU file and see how many are wrong.

[SMZ]

lol, I know the txt file is very inconsistent, but that's what the values are listed as supposed to be, in my own txt file I went thru and manually set each unit to the correct numbers, although on mine I put full plate at 10 instead of 9

secondly, it's true, I forgot that defense skill doesn't count to dodging arrows, however that doesn't alter the conclusion at all... if the Tercios only received 2 points of armor from being upgraded to silver, why would they all of a sudden have 23 fewer casualties? That's not a 2 point difference... that's a 7 point difference... as it is supposed to be.

Nonetheless, I'll run the test you suggest with the Pike Militia first altered to have an armor rating of 3 and then an armor rating of 7

[Carl]

secondly, it's true, I forgot that defense skill doesn't count to dodging arrows, however that doesn't alter the conclusion at all... if the Tercios only received 2 points of armor from being upgraded to silver, why would they all of a sudden have 23 fewer casualties? That's not a 2 point difference... that's a 7 point difference... as it is supposed to be

Good, point, would still be intresting to see the result though.

lol, I know the txt file is very inconsistent, but that's what the values are listed as supposed to be, in my own txt file I went thru and manually set each unit to the correct numbers, although on mine I put full plate at 10 instead of 9

Ahh, right, fair enough.

[FactionHeir]

Arrows are a very random factor, unless you change their accuracy to near 100%. They miss more the less soldiers remain in a unit.

You probably want to run your test in a melee situation with defense values set to 0.

[SMZ]

Arrows are a very random factor

the grouping of my results says they are reliable enough... the results in the tests are pretty consistent

and I did arrows because it removes one less variable from the testing - if I do a melee contest, then the rate at which the enemy kills me also affects how quickly I kill him, the rate at which both units are swinging, the location of the generals, etc, etc, etc - all come into play.... arrows makes for a much easier measurement... but lemme go do the new manual alter tests

[SMZ]

alright, I went back and did two more tests on the upgraded Tercios to finish that set of five, and then I did 5 tests on Pike Militia which I'd altered in the unit file with 3 points of armour....

as I was hypothesizing, the results landed in between - it's clear therefore that the armor is working correctly



Actually I'm going to guess that any tests done previously probably were borked due to the shield bug... there would've been no way to tell what was going wrong until ppl knew about that - undoubtably leading to all sorts of weird results

[Carl]

Thats, most intresting, thanks for that.

[SMZ]

I just wish I knew how to make it say the right thing now...

oh, and as I was fixing the armor numbers earlier today I was thinking, I bet a lot of ppl would like having this cleaned up... but since I can't even remember all the other little tweaks I've done, I can't just upload my file for ppl to get =/

[pat the magnificent]

I just did some tests and I'm sorry to say, I'm absolutely certain your conclusion is incorrect.

here's the test...

1 unit of pike militia (player) vs 1 unit of peasant archers (computer)

normal unit sizes. grassy plain, 2 sets of 10 tests. archer's ammo capacity reduced to 10. pike militia taken out of spear wall and defensive stance. clear weather for all tests.

Set 1, unmodified pike militia with 3 armor upgrades. (armor stats says 3)

Men Remaining
test 1= 57, test 2= 50, test 3=52, test 4= 50, test 5=50, test 6=53, test 7=46, test 8=54, test 9=56, test 10=51.

average men remaining= 51.9

set 2, modified pike militia with armor stat set to 8, no armor upgrades. (armor stat says 8)

test 1=55, test 2=55, test 3=61, test 4=57, test 5=56, test 6=54, test 7=54, test 8=55, test 9=57, test 10=63

average men remaining= 56.7

a difference of 4.8

considering that the armor stat difference is only 5, I'd say thats fairly significant and isn't likely the result of a statistical anomaly.

conclusion= the armor upgrading system not only appears screw up, its also functionally screwed up. i wish it weren't so, but it is.

[Lusted]

Actually your tests confirm armour upgrades work, as a 3 armour upgrade unit of pike militia should have a defense of 7, despite only 3 being shown on the unit card. So having a difference of 4.8 between them and a unit of pike militia with their defense set to 8 is about right.

[pat the magnificent]

I'm rethinking my conclusions, i just ran a third set of tests.

modified pike militia with armor manually set to 3. no upgrades

results as follows

47, 46, 46, 44, 45, 46, 44, 48, 46, 47

average men remaining 45.8

which is 6.1 men greater than the unmodded upgraded pikemen.

clearly the upgraded pikemen are performing better than their stats reflect

[pat the magnificent]

here's another question that needs to be posed, what value does the autocalc engine use, the real defence value, or the displayed defence value?

[Foz]

here's another question that needs to be posed, what value does the autocalc engine use, the real defence value, or the displayed defence value?

We've been told that the autocalc engine uses the same mechanics as the battlefield does, so it is reasonable to assume anything you find out about the battlefield engine applies similarly to autocalc. This may not always be the case, but it's generally supposed to be so we should assume they mirror each other in the absence of contradicting evidence.

As for the find, I'm glad someone finally took the time to do it. In truth I had actually forgotten this issue was still hanging around. I would point out though, SMZ, that if you speculate that a given upgrade makes the unit have 7 defense, then the way to prove this is not really to modify it to 3 in the EDU and note that the results seem better with the upgrade - it's to set the unit to 7 in the EDU and more or less prove that you get identical results. That way you're not just generalizing by saying the upgrades do more than the sheet says, you're actually quantifying the value it does apply.

In other news, now I find myself wondering if each armor upgrade adds a base value (i.e. an armor bonus) or if it actually replaces the value the unit listed with the given value for the set of armor it's been upgraded to. That is, there are two real possibilities I see, and am curious which is used. Let's look at lvl 1, leather/padded:

1. Any affected unit has its armor value replaced by a 4 in the calculation.
2. Any affected unit has 4 added to it's armor value in the calculations, because leather has value 4 and no armor has value 0, so 4-0 is +4 armor.

The main thing that has me concerned, then, is the shield fix - if the values are replaced instead of given as a bonus, then shield fixed units actually take an armor penalty for achieving their first armor upgrade. It would probably have been easier to code as a bonus from a lookup table, but one never knows...

[Carl]

@Foz: It's got to be an addition because not all units have the same Armour value at a given Armour level in vanilla, so if it was fixed you'd get the same issue you describe possibly happening with the shield fix.

[JaM]

Guys, Darth did good research about animation files, and i can confirm that in vanila game there is nothing that links unit look to unit resistance. All units have the same values used.

Standard values are -0.090000004 0 0 -0.34999999 0.80000001 0.60000002

where

-0.090000004 = min effective pitch for left hand (animation node)

0 = max effective pitch for left hand (animation node)

0 = max effective pitch for right hand (animation node)

-0.34999999 = min effective pitch for right hand (animation node)

0.80000001 = probably works as an attack modifier for the skeleton (The bigger the more)

0.60000002 = probably works as a defense modifier for the skeleton (The bigger the less)

Same line is applied for armor upgrades. so there is no secret system that links unit look with armor protection.

[Lusted]

I cannot help but feel Darth is wrong in this matter given what the test results show.

[JaM]

Guys, its not that hard to test it... just remove shield value from spearmens and put italian militia against seargant spearmen with bronze upgrade. In your teory, they should have same protection - leather, but Italian with armor 4 will kill seargants with much lower looses than if they had armor 1 or 2 or 3 or anything...

[JaM]

As i wrote on TW, try Point Blank EDU with shieldfix, it clearly shows, that armor upgrades gave +1 point only, and unit with +1 benefit will have advantage. That EDU is the most balanced EDU for M2TW yet.

[Lusted]

Italian Militia are better in close combat then sergeant spearmen. You need to test both of them against missile troops.

[Carl]

Not just that, but their are a hell of a lot of other factors going on in melee that can bugger things up quite badly. Missile tests are the only ones where you can draw accurrate conclushions in this case.

[JaM]

Lusted, Point Blank even created his own armor classes, he has 12 upgrade levels instead of vanila 6. Burrek created lots of new units, Darth modified animation factors, with all that below, if there is any background system behind, it should definitelly screw those mods with faulty battle results.Opposite is true. I got realistic results when i used that EDU - Dismounted knights were able to do a lot of damage to low quality units,but were innefective against units in heavy armor because sword is not very good at fighting somebody in Plate... Spear units without armor upgrade were beaten by same type upgraded units ... try it! and you will see it. many mods are starting to implement that system (ask Re Berengario f.e...)

[Lusted]

And yet im not as the evidence points to armour upgrades working correctly in M2Tw without Darths fiddling around with a number no one really understands, or knows how it works.

[JaM]

Missile damage is working differently. I had looses even if i lovered missile attack to 1 and left unit to be under arrow fire...(not big thou) Shield is not working as it should, it has just 9 degree coverage. + it decrease armor value in melee.

[Lusted]

Yes shields are not working correctly. Why do you think the tests in regards to armour upgrades have been done using shieldess units?

[JaM]

If you remember, i send you an IM to ask developers about that. I posted same question on Totalwar.com forum and i was told that there is no backbround system behind armor upgrades. It was planned, but something went wrong, so they removed it.

[Lusted]

You asked me about an unused stat line in the export_descr_unit.txt file. At the time i knew people had been testing the armour upgrade system, but there had been no definitive test results due to the shield bug. Now there are definitive test results.

[JaM]

Because i did lots of tests with all kind of units. I had a big disscusion about that with TimeCommander Bob few mounths ago (at that time we didnt know anything about shield bug) he did all his tests using melee, I used missiles.His tests show that italian militia has same resistance that unit with upgrade armor 1 - leather, but shield substracted -6 from protection.

I tested pikes and billmens, and my tests show that unit with armor 1 is worse that unit in armor 2 and unit in armor 3 etc... no mather whic armor level unit had.

Units have mortality factor 0.6 right now, so even if unit has no armor at all it will not die immediatelly if hit. (it ignore 40% of hits)