The Complete EDU Guide - Discussion

A new guide analysing the EDU has just been posted at:
https://forums.totalwar.org/vb/showthread.php?t=88859

This thread has been created to discuss this Guide. If you have anything to add, correct, suggest etc. then please do so. Please help make this the best, most complete and most comprehensive source of "everything EDU" so that all information regarding this is gaterred at a single place, easily accessible and found by all.

The Guide does not seek to define ways in which the information can be used, only to provide that information. So, although we might use certain aspects of it to create a counter-elephant specialisist unit (for example) or a unit of demi-gods, those applications are beyond the scope of this Guide. This file provides only the raw data (and some very quick and simple advice).

Your input is much appreciated and anticipated.

That's a great document, thank you. :2thumbsup:

Quote:

---

DLF is the difficulty lever factor. It's 0.7 for Easy, 1 for Medium, 1.5 for Hard and 2 for Very Hard. This factor is what produces the AI combat bonuses for Hard/Very Hard and the AI combat penalty for Easy.

---

That's an interesting find, I knew there was a difference but never realised it was that large.

Would it be at all possible to do an even more idiot proof explanation of the two useful formula at the end? This particular idiot isn't sure how the two constant values fit in.

And thank you, thank you, thank you, for coming up with that before I have to try and balance units for EoDII :2thumbsup:

You're very welcome Mak!

Well the melee formula gives a number, which in turn is presumably compared to random generated number that's up to 100, to see if the hit is going to kill. Without the constant, the max number one can get is 2*1.1^63, assuming Very Hard difficulty, which roughly equals 810. So you need to multiply that with a constant to make it smaller than 100, but still close to it, to make sure that the chances of spilling blood remain big. A number that could well do that is 0.095 (=0.19/2), which gives max chance to kill 77%. On the other hand, there might be no constant and the number generated is directly compared to number up to 1000, which would give a max chance to kill of 81%. Not much difference and common sense says that the second way is better, but we're talking about CA coding here and they have used the constant before in the formula for MTW.
A similar thing applies for the missile formula, but since the two are different, it's only reasonable to assume that another constant is used there, hence const2.

EDIT: Since const1, const2 are fixed, direct multiplieres and unmoddable, it really doesn't matter what their value is, they are just there to make the formula accurate.

Can you see why I left it to Aradan and Xerex now, Mak? :)

But, seriously, I hope this does help out EOD II and many other mods too.


Aradan, not the most important info in the world, but we should note that the category and class entries also affect the order in which the unit appears in the custom-battle unit-selection screen. Generally it is infantry followed by cavalry category, then declining as light-spearmen-heavy-missile class, then edu order in the case of same category/class units. This is probably not only not important but pretty useless - but it is the *complete* guide ;)

Yes, it's the completely WIP guide! :P

As I said in the dev forums...

OK; enough kidding, great work Aradan and Xerex! (I'm used to say Xerex & Aradan, most times it's just something that you get used to, the order you say the names in for example. Who ever says dad & mom, it's always mom & dad!)

So I am mom....great...:)

Thx, Pal I will update.

Great guide. Have you added it to the tutorials database?

Thx. No I haven't done so yet, as I'm unsure of how... Any pointers would be greatly appreciated! :)

this link should take you to the add link area:
https://forums.totalwar.org/vb/local...dlink&catid=71
parent categories should be under
-Rome Total War
--Modification guides and tutorials
--- --All tutorials--
and
---Units
I think, and any others that seem appropriate, if it doesn't work I'm sure Myrddraal knows how to fix (hopefully).

I could try it for you but that is meant to be something modders add their own tutorials etc to so would like to encourage you to try it out. :yes:

Done, and thank you very much!

Hi! I am one of EB's stat guys and I'd just like to say that you've done an excellent job so far. I've been experimenting with the stat_ground values and have recently been speculating that the scrub bonus could be seriously messed up - possibly giving a bonus in more terrain types (or possibly all terrain types) than it really should. I haven't done as many tests on this as I like and, as it says in the guide, it can be difficult to tell when exactly you are fighting on scrub, so it's entirely possible that I'm imagining things. Does anyone have any good data on this? I'd be very happy to know what others think about this.

You forgot screeching_women in the list of attributes.

I learned a lot of new stuff from the guide. thanks and great job!

Thanks tk-421 and Kuni. Well, you see 99% of the testing was on BI 1.6 and FATW, so no screeching_women there... I will test it on 1.5 too and add info about it once I have solid facts/numbers. Thanks for pointing it out.

Quote:

---

Originally Posted by scriptorium article

[weapon_tech] : Unit weapon's tech type. Can be: simple, other, blade, archery or siege (note that 'other' is not upgradeable).

---

Other actually is upgradeable it just doesn't show up in building browser. The note that you see that it is not upgradeable is from text entry in strat.txt

Quote:

---

{SMT_HEAVY} Heavy
{SMT_NO_UPGRADE} Cannot be upgraded

---

The 'other' units do actually upgrade, eg if you use:

Quote:

---

weapon_other 1

---

in building you get option to retrain them and their weapon changes to bronze as usual, you just don't get a "Upgrades xxx weapons" on the building description screen saying it's going to do that, which seems to be due to that part being missing from strat.txt as well.

Trying to add {SMT_CAPABILITY_WEAPON_OTHER} etc. to that file seems to have no effect - I assume the lookup for that file is hardcoded.. ?

anyway you can mod EDB so they are upgradeable. :yes:

Thx Mak, that must have been my incorrect assumption based on the lack of text.

Thank you too. Update.

Quote:

---

Originally Posted by tk-421

Hi! I am one of EB's stat guys and I'd just like to say that you've done an excellent job so far. I've been experimenting with the stat_ground values and have recently been speculating that the scrub bonus could be seriously messed up - possibly giving a bonus in more terrain types (or possibly all terrain types) than it really should. I haven't done as many tests on this as I like and, as it says in the guide, it can be difficult to tell when exactly you are fighting on scrub, so it's entirely possible that I'm imagining things. Does anyone have any good data on this? I'd be very happy to know what others think about this.

---

I was the one that did the terrain testing mostly and yes, scrub is a pain in the butt. All the other terrains work perfectly and you can clearly tell when you are in them. Scrub, however, is very sketchy. A map can be almost entirely plains with a little patch of scrub and if your unit is fighting on that piece of scrub, it will own. Moving 20 meters to the right makes your unit fight much worse. Scrub can be generally described as "rough" terrain, but pinning down exactly which terrain types are considered scrub requires that I first learn more about how the battle maps are set up which will take some time because I am quite dumb and overloaded at work.

You may also be experiencing stacking of terrain bonuses. I dont remember if terrain bonuses stacked. I think I remember testing it, but it wasn't in the guide. Let me look into it and see if terrain bonuses stack.

Didnt CA say previously that the spear attribute gave +4 attack vs cavalry? The guide says +8

Well, CA has occassionally said things that weren't 100% accurate. The spear bonus vs cavalry is 8. Perhaps Jerome said that the general bonus was +4 (+8 vs cav and -4 vs inf = 4) ?


@ tk-421 : Tested a bit more and I still get what I got in the complementary tests I did after Xerex made the 'base' ones: scrub-modifier only gives its bonus on scrub, not on any other ground-types. We're lucky to have an entirely scrub-covered map in FATW and we can know for sure what ground we fight on. As Xerex said, it's a low-grass, arid climate that is between rocky, desert and plains. Not very illuminating, but I hope it helps a bit. I'd imagine your Baktrian provinces would have it, but again I'm not a mapper.


A slight update in the guide, added "skirmish" as an option for class. It's not used at all in vanilla, but it could be useful, especially for formation guys that don't want to go the long way round (if there is anyone else except for me that's actually bothering with battlefield AI, that is).

Quote:

---

Originally Posted by Aradan

A slight update in the guide, added "skirmish" as an option for class. It's not used at all in vanilla, but it could be useful, especially for formation guys that don't want to go the long way round (if there is anyone else except for me that's actually bothering with battlefield AI, that is).

---

Firstly, let me say this is a great guide.:2thumbsup:
So skirmish is a valid option class, is that in both 1.5 and 1.6? When you say it could be useful for formation guys that don't want to go the long way around, what do you mean exactly? I'm curious as to how do you use the tag as it's relative to the priority awarded within the formation? For example let's say if missile gets .8 or .7 do you go higher or lower? or do you even play with the priority at all if your using a battle AI mod like Darth or Sinuhet; rather your just changing the unit class in the edu?

Thx.

Well the long way round is to have a block containing "missile infantry" with low priority and another one containing only "ranged_missile_infantry" with higher one, so that skirmishers will end up in the first block. It's not a perfect way, but it's acceptable. While now you can use "skirmish infantry" and it will work perfectly. I know vanilla, Sin and Darth already use it, but they always use it along with "missile infantry" for their interception-line-block and they use "ranged_missile_infantry" alone for the archer-block, so that skirmishers always end up in the first-line. I made a custom formation not a week ago and I had assigned only "skirmish infantry" in the first block with a minimum of two such units. I gave the formation a high priority (it was a defensive one, so it would be selected for deployment) and made sure all the other requirements were met, but guess what: the formation was never chosen. When I changed a unit's class from missile to skirmish, the formation was selected.

Note that perhaps the class isn't fully supported (I can't understand why it wasn't included in vanilla otherwise) and it needs further testing.

Thanks for the "long way round" explanation. That was actually a pretty dumb question on my part. For some reason i was reading something else into what you posted. I should have known immediately what you meant by it. :dizzy2:

In your custom formation; when you say you changed a unit's class from missile to skirmish and the formation was selected (defend); do you literally mean you changed 1 unit's class and the formation was selected, even though you had assigned that first block a minimum of two such units?

It wasn't dumb.

I changed a unit's (unit = EDU entry) class from 'missile infantry' to 'skirmish infantry', and then I used two 'instances' of that unit in my army. How else would the formation be selected?!

Right....of course.
Pardon, apparantly my reading and comprehension has been a bit off in this thread.:dizzy2:

The problem with using skirmish instead of missile is that the AI will recruit them as infantry and result in the AI spamming javelin units...

Thx for the feedback, mcantu. That's useful for campaign AI balancing.

One thing that I think should be included. For the mercenary_unit attribute it prevents a unit replenishing losses. I also believe that it forces the units with the attribute to use the 'merc' texture and sprite lines in DMB for all faction owners of the unit even if they have their own texture line in DMB.

Thx, Squid, added. Are you sure about the texture and sprite thingy?

Pretty sure, I was testing how to get two differently textured units for one faction from one DMB entry, so I had two EDU entries pointing to one DMB. One of the EDU entiries had the mercenary_unit attribute the other didn't. The one with the mercenary_unit attribute displayed the merc texture, the other the specific texture for the faction I was testing with.

Yep, tested it myself, didn't have the time to edit the Guide though. Thx for the confirmation!

Question for Aradan actually.....
When the developers of FATW decided to make the switch from Darth Formations to Sinuhet; did you guys have to make wholesale changes to the formation line in the EDU in terms of unit width and depth to improve the performance of Sinuhet's AI? Also did you guys have to change the amount of soldiers in the soldier line as well?

Quote:

---

Originally Posted by Squid

One thing that I think should be included. For the mercenary_unit attribute it prevents a unit replenishing losses. I also believe that it forces the units with the attribute to use the 'merc' texture and sprite lines in DMB for all faction owners of the unit even if they have their own texture line in DMB.

---

I'm pretty sure that's the merc trick we've been using in EB for years to squeeze an extra skin per model. I didn't know it wasn't common knowledge.

No, we didn't change anything in our EDU for that purpose. Sin's formations are very flexible in this regard. But that could depend also on the specifics of your game's EDU. The only suggestion I could perhaps make is to make your horse-archers formation less wide, because with Sin's formations they really tend to try and flank/envelop your infantry and being narrower makes them more versatile.


Well, Bozos, nothing remains a secret for long in a community... And now everybody knows! Or at least those that bothered to read the Guide... :book:

Quote:

---

Originally Posted by Aradan

No, we didn't change anything in our EDU for that purpose. Sin's formations are very flexible in this regard. But that could depend also on the specifics of your game's EDU. The only suggestion I could perhaps make is to make your horse-archers formation less wide, because with Sin's formations they really tend to try and flank/envelop your infantry and being narrower makes them more versatile.

---

Thanks for the tip. One more question if i could and pardon if it's a tad bit off topic. Have any FATW players reported occasions of an "idle" reinforcing army using Sin's formations?

I don't think any have, but I have seen it happening some times. But i have seen it hapenning with vanilla and Darth too. I have restricted it a bit by changing some offsets, but I don't think there's a way to solve it completely, due to the way the formations are coded - and the lack of a descrimination between normal and reinforcing armies.

another topic that might be relevant to this:

I've been having trouble with the 'Abilities at a Glance' section of the units info screen not showing up, have pinned it down to an extent as being to do with the selection of the stat_ground bonuses, problem seems to be mainly to do with having -1 and -1 for forest and snow (the last two figures) eg:
stat_ground 1, 2, -1, -1
though some other permutations seems to be doing it to me as well. The full EDU entry for the unit that definitely isn't working is:

Spoiler Alert, click show to read: 

type assakii infantry
dictionary assakii_infantry ; spear & sword
category infantry
class spearmen
voice_type Medium_1
soldier assakii_infantry, 40, 0, 1.1
officer rahala_standard
attributes sea_faring, can_sap
formation 1, 1, 2, 2, 5, square, shield_wall
stat_health 1, 0
stat_pri 8, 4, no, 0, 0, melee, blade, piercing, spear, 25 ,0.7
stat_pri_attr spear, spear_bonus_10
stat_sec 8, 4, no, 0, 0, melee, simple, piercing, sword, 25 ,0.85
stat_sec_attr no
stat_pri_armour 6, 4, 4, metal
stat_sec_armour 0, 0, flesh
stat_heat 0
stat_ground 1, 2, -1, -1
stat_mental 10, normal, trained
stat_charge_dist 30
stat_fire_delay 5000
stat_food 60, 300
stat_cost 1, 330, 90, 50, 70, 330
ownership vandals

has anyone else experienced that or got any ideas as to why that would be happening?

P.S. attributes like "can form shield wall" are showing up fine in custom battle selection they just don't appear on the unit info screen for that unit....

I don't think this has to do with the ground modifiers... Can you try to add two more attributes to the unit? Re shield-wall, all looks fine here. And I think formation abilities always show up at the top, so that can't be lack of space in the 'abilities' section... Weird.

the only thing I can spot on the go is that in vanilla Rome (guessing its the same for BI) every unit except siege-engines has the atribute;
-hide_forest, or
-hide_improved_forest

=======

**5 minutes and to custom-battles further down the line ....**

Mak you again are the finder of little oddities ..;)
see my lvl3 archer here, hadnt even noticed this before;
https://img251.imageshack.us/img251/7242/0000uj2.th.jpg

I can assure you they "can sap" as do their lvl2 brothers wich also show they can .... (similair units-stats, but +1attack, +2armour)

the stats;

Code:

---

type            roman archer auxillia
dictionary      roman_archer_auxillia      ; Archer Auxilia
category        infantry
class            missile
voice_type      Medium_1
soldier          roman_archer_auxillia, 50, 0, 0.4
officer          roman_early_standard
mount_effect    elephant +8
attributes      sea_faring, hide_improved_forest, hide_long_grass, can_sap
formation        0.8, 1.2, 1.2, 1.6, 3, square
stat_health      1, 0
stat_pri        8, 1, arrow, 170, 50, missile, archery, piercing, none, 25 ,1
stat_pri_attr    no
stat_sec        3, 1, no, 0, 0, melee, simple, piercing, knife, 25 ,1
stat_sec_attr    no
stat_pri_armour  8, 5, 0, metal
stat_sec_armour  0, 1, flesh
stat_heat        0
stat_ground      2, 0, 3, 1
stat_mental      10, disciplined, highly_trained
stat_charge_dist 15
stat_fire_delay  0
stat_food        60, 300
stat_cost        1, 260, 80, 40, 60, 260
ownership        roman


type            roman archer auxillia ii
dictionary      roman_archer_auxillia_ii      ; Archer Auxilia II
category        infantry
class            missile
voice_type      Medium_1
soldier          roman_archer_auxillia_ii, 50, 0, 0.5
officer          roman_standard
mount_effect    elephant +8
attributes      sea_faring, hide_improved_forest, hide_long_grass, can_sap, hardy
formation        0.8, 1.2, 1.2, 1.6, 3, square
stat_health      1, 0
stat_pri        9, 1, arrow, 180, 60, missile, archery, piercing, none, 25 ,1
stat_pri_attr    no
stat_sec        3, 1, no, 0, 0, melee, simple, piercing, knife, 25 ,1
stat_sec_attr    no
stat_pri_armour  10, 6, 0, metal
stat_sec_armour  0, 1, flesh
stat_heat        1
stat_ground      2, 0, 3, 1
stat_mental      12, disciplined, highly_trained
stat_charge_dist 15
stat_fire_delay  0
stat_food        60, 300
stat_cost        1, 320, 100, 50, 75, 320
ownership        roman

---

G

Red Spot, I think the dissappearence of your "can sap" there is because there are no more 'abilities' slots left. The last slot of your Legionary Archers' abilities is taken by "hardy".

reading between the lines I take it that you mean I can only assigne 4 atributes, or do you mean only 4 will show up in the "at a glance" bit??

anyway as example an "vanilla" upgraded general;

Code:

---

type            carthaginian general's cavalry
dictionary      carthaginian_generals_cavalry      ; General's Armoured Bodyguard
category        cavalry
class            heavy
voice_type      General_1
soldier          carthaginian_royal_cavalry, 12, 0, 1
mount            generals horse
mount_effect    elephant -8, camel -4
attributes      sea_faring, general_unit, hide_forest, hardy, general_unit_upgrade
formation        1.5, 4, 3, 6, 3, square
stat_health      2, 0
stat_pri        11, 6, no, 0, 0, melee, simple, piercing, spear, 25 ,1
stat_pri_attr    no
stat_sec        12, 5, no, 0, 0, melee, simple, piercing, sword, 25 ,1
stat_sec_attr    no
stat_pri_armour  11, 7, 0, metal
stat_sec_armour  0, 1, flesh
stat_heat        0
stat_ground      0, 1, -6, -1
stat_mental      10, disciplined, highly_trained
stat_charge_dist 40
stat_fire_delay  0
stat_food        60, 300
stat_cost        1, 1040, 80, 110, 160, 1040
ownership        carthage, spain, numidia

---

thats 5 atributes, though ofcourse most normally dont show up in the "at a glance" bit ...

more testing to do ....... :wall:


Edit;
still havent done the archer thing, but there are a few things I've experianced in combat based on a few simple EDU tweaks that I tought might be interesting, dont know for sure if I should post it here .... anyway ..

-setting charge distance for regular infantry to something like 15 and spear to ~5 makes them charge in a much smoother "grouped" fashion, keeping unit formation intact even in the charge, even with my formation withs (0.8,1.2inf--0.8,0.8spear)
I was actually impressed by their charge and it "seemed"(untested) to reflect in combat-results as well


G

Quote:

---

Originally Posted by Aradan

I don't think any have, but I have seen it happening some times. But i have seen it hapenning with vanilla and Darth too. I have restricted it a bit by changing some offsets, but I don't think there's a way to solve it completely, due to the way the formations are coded - and the lack of a descrimination between normal and reinforcing armies.

---

I've never actually seen it happen with Sinuhet's; I've seen it in vanilla and Darth's but only when i was tinkering with a few things and i omitted his width forcing while using huge unit sizes. I ask because it's been mentioned to me a few times. Must be just one of those weird things i haven't seen yet. Thanks for the cavalry tip...I'm gonna narrow down some of the cavalry formations in the EDU a bit and see if i like the angles they take a little better.

np Lysander. I've read the comments on your EB formations thread. If you find out what's causing this effect, please share!

@ Red Spot: Yep, small charge distance values seem to work well for infantry, but they are dependent on the ground-type speed modifiers. If they are too high some units that have slow charging animations might have some problems. Generally such things need tweaking according the specifics of the mod.

I *think* that the unit card only displays up to 7 abilities at a glance. It's not the number of attributes in the EDU, it's the actual number of displayed messages on the card.

The "can_sap" one seems to be problematic, as it seems to hide the abilities section completely if it's not coded along with at least 3 more attributes...:wall: Anyone else having this issue (besides Mak obviously)?

I haven't checked the can_sap bit, will do that tomorrow. But I can definitely fix my problem on an off with that unit I posted solely by changing the forest terrain modifier from -1 to -2 ! Thats what I meant by having pinned down to terrain modifier, though it might be an odd quirk that means that only applies combined with the other problem..... definitely :wall:

Aradan, I know(still apreciate you mentioning it), my groundtype modifiers are ~at 0.8 wich makes eg legionaries able to get trow a pilum before getting charged by cavalry (in general!), didnt post that as I figured it was relativly "general info" :D
(imo at ~0.8 is pretty good "allround" higher and certain animations are too slow to prevent AI/player imbalances, too low and animations get ugly as it looks like a unit is running its behind of while it hardly is moving (+ that it over-balances ranged units)

I agree that 7 seems to be the magic number, my lvl3(+) units never show more than a total of 7 where some should imo at least show 8 ....
Interesting though that the sapping ability is the one causing problems, I'd sooner be looking at like the turtle formation(just try to use it on other units ..:P)


G

Re: my earlier comments on the 'abilities at a glance' on the unit info screens not turning up - I haven't got a conclusive reason but can give some examples

Unit with:
attributes sea_faring, can_sap
and
stat_ground 1, 1, -1, -1
does NOT have abilities shown in screen.

Unit with:
attributes hide_forest, can_sap
and
stat_ground 1, 1, -1, -1
DOES have abilities shown in screen, Red Spot was on to something with that..

but also
Unit with:
attributes sea_faring, can_sap
and
stat_ground 1, 1, -2, -1
DOES have abilities shown in screen.

Unit with:
attributes sea_faring, druid, can_swim, can_sap
and
stat_ground 1, 1, -1, -1
does NOT have abilities shown in screen, but if you switch sea_faring to hardy it does work

but also;
Unit with:
attributes sea_faring, druid, can_swim, can_sap
and
stat_ground 1, 1, -2, -1
DOES have abilities shown in screen

also
Unit with:
attributes hardy, druid, can_sap
and
stat_ground 1, 1, -1, -1
DOES have abilities shown in screen, though have noticed that the druid attribute isn't mentioned - not sure if it ever is?

I'm a little lost as to what the linkage is between combinations that do work and ones that don't! Could anyone else have a look and see if they can confirm any of this and come up with some idea why...

From little testing I've done, with attributes sea_faring, can_sap (or only sea_faring) 'abilities at a glance' shows if I have any parameter in stat_ground 3, or -2. Though if they dont shows, they applies anyway, at least my gallic swordsman with 0,0,0,0 was clearly beaten by britons with 0,2,0,0 in desert.
So use any hide_forest or any other attributes to make abilities visible, or for non-hideable units big "-" in stat_ground...

*bump* hopefully question belongs to this topic.

I was wondering if anyone has managed to come up with any conclusive information about how unit sizes and stats affect auto-calc (other than just auto-calc sucks!) - is there anything you can do within EDU to help an AI faction with smaller unit sizes?

Guide updated with info on 'short_pike'.

Other than horse units are seriously underpowered in auto-calc and that heavy mounted missiles are terribly overpowered, nothing conclusive... A nice opportunity to do some testing though. :)

Did you figure out the can_sap/stat_ground problem? I've had no luck pinning it down.

Aradan, can you give more detailed explanation of discipline?

Quote:

---

[discipline] : Unit's discipline level, which determines the rate at which its morale is depleted or restored during melee. Can be low, normal, disciplined, impetuous or berserk. Disciplined units are harder to lose morale. Berserk units can (obviously) go berserk and impetuous units may charge without orders.

---

Does this means that impetuous is better than disciplined, or only in restoring morale?

low < normal < disciplined < impetuous < berserker

As you go towards the right, units gain morale faster and lose it slower. So disciplined is better than normal (which is better than low), but it's worse than impetuous (which is in turn worse than berserk).

Quote:

---

Originally Posted by Aradan

Other than horse units are seriously underpowered in auto-calc and that heavy mounted missiles are terribly overpowered, nothing conclusive... A nice opportunity to do some testing though. :)

Did you figure out the can_sap/stat_ground problem? I've had no luck pinning it down.

---

hmmm, I seem to be acquiring lots of opportunities to test things :laugh4:

The dissappearing description thing I didn't get any further than finding those examples of working non/working combos, couldn't figure out any particular conclusion, just fiddling the attribute or stat when it occurs!


Didn't spot you mentioning anywhere the "general_unit_upgrade"
attribute, I assume that's tied to the
upgrade_bodyguard 1
in EDB, EDB guide lists that as operating after Marian reforms. Does anyone know if you have that in EDB and reforms operating, do you need a general_unit_upgrade for all the factions?

AFAIK, general_unit_upgrade makes the game switch bodyguards to the unit with this attribute when the marian reforms hit. The upgrade_bodyguard thing in EDB has no effect in RTW 1.2+.

you know I've been meaning to test that since like I started modding the Romans ... long ago ... :embarassed:

will do that right now ...

Edit;works like a charm, that is, under BI 1.6
when the marian reforms hit all my generals instantly got the new bodyguard
Edit2;slight correction, the bonus doesnt seem to do anything, its the reforms that trigger the upgrade
setting the upgrade to 2 or giving it at 2 to a later building doesnt trigger any (next) upgrade


G

Re; general_unit attribute, I only just noticed and not sure if its mentioned somewhere but it seems to entirely stop the AI recruiting that unit. So if you have ZOR based recruitable generals bodyguards you won't get them used by the AI.

Yes, we have noted that behaviour in FATW, the AI will never use RGs. Guide will be updated, thx for poitning that out.

Thanks for that Mak. as that was an other thing I still had to test (was hoping that it was the high price and 2+ turn recruitment that stopped them from recuiting them ...)

too bad ...:(


G

Actually in RS we've had opposite effect, we have recruitable generals, and we had problems in a previous release where the AI would start spamming full stacks of general units.

We tracked the problem down to one of the AIs cost benefit analysis, where if the AI thinks that the general unit is the best unit for the cost, it will continue to make the unit until a better cost benefit unit comes along.

gah, don't understand that, I was having problem with absolutely none of the general_unit atrib cavalry turning up as bodyguard or otherwise, tried fiddling stats as I assumed that was problem but that didn't do anything, removing general_unit resulted in them getting spammed through every stack......

That and FATW are on BI were you testing on RTW exe?

I can't take credit for the solution, but what I think dvk901 did was he just kept reducing the number of men in a generals unit until they were no longer being spammed all the time.

As far as I know it happened on both exes, but I could be wrong it may only have been rtw.

Hmm, when testing I once removed all garrison from one settlement and all buildings that enabled training of other units leaving only the RG unit option to the AI. They never recruited them in a period of 20 turns. They first built barracks and stables and only after that they started traing; for the record the first unit they trained was a heavy cavaly (RG was heavy cav too) with 10 times the recruitment, support and upgrade costs of the RG, needed twice as long to recruit and had half the men. All that on BI.exe.

Quote:

---

Originally Posted by Makanyane

I was wondering if anyone has managed to come up with any conclusive information about how unit sizes and stats affect auto-calc (other than just auto-calc sucks!) - is there anything you can do within EDU to help an AI faction with smaller unit sizes?

---

With regular cavalry units you can use the second value on the stat_health line to improve auto-calc performance. It has no effect on performance in a normal battle, it might have some effect on AI recruiting preference (it looked that way to me but I didn't test it rigorously), but it definitely has an effect on auto-calc. I have used this to improve the performance of horse archer factions in XGM with great success.

That's very interesting, Dime. Thank you. Something for Aradan to check out and update ;)

you may also have noticed that even though they dont use it, vanilla does set secondary armour stats for cavalry as well ...

Code:

---

stat_sec_armour  0, 1, flesh

---

Though I dont know if it may influence anything, I just use it everywhere or nowhere at all ...


G

Yes, I have noticed that. This is definitely something for Aradan and his army of reliable, committed beta testers to look into :beam:

Here are the results of some of my recent tests. I give only the average percentages (fought some sixty-sth battles), because formatting the entire table would take a good while...
Difficulty is set to Medium-Medium, and there are no generals involved. The first row is using stat_health increase only, the second using stat-sec_armour increase only and the third is normal (both values are 0).
The first column is using 4 units of riders (2 groups of 2 couples of the same unit-type), the 2nd using just the first group, the third the other one, and the 4th and 5th using 1 unit of each troop-type.
The "*" sign means that I think the result here was so out of the 'acceptable range" that it shouldn't be taken into account as it is.

https://img119.imageshack.us/img119/...balancejk9.jpg

Increasing stat_sec_armour's second value from 0 to 2 didn't seem to have any significant result, that cannot be explained by the randomness of the results.
Increasing the second value stat_health from 0 to 2 might have some effect, but given the range of results I am not yet 100% certain. It might well be a fluke of luck, as auto-resolve seems to be very random; using the exact same armies I got results ranging from Clear Defeat to Clear Victory. Still, there is a decrease to enemy kills in most occasions that is significant and needs further research.

Other than that, the results *suggest* (though the number of tests is still quite small) that the auto-calc is using some sort of logarithmic function, that lacks balance when armies are very small of very big.

Dime, what values do u use for your horse-archers stat_health? 1,2 or way bigger?


EDIT: I tested some more using stat_health 1,10 for one unit type and 1,20 for the other. Indeed now the units did much better. Whether using all four units together, or just a couple of them or even a single unit, they always inflicted sth like 53% casualties to the enemy. It seems the auto-resolve has a cap somewhere there for this particular combination of armies. The big deal is that now my cavalry was taking far less casualties than before.

one unit 20 sec-hps: now 6% - 45% before
one unit 10 sec-hps: now 8% - 39% before
two units 20 sec-hps: now 9% - 35% before
two units 10 sec-hps: now 10% - 21% before
four units 15 sec-hps: now 5% - 15% before

So it seems that indeed stat_sec is used in auto-calc and it can make quite a difference in the amount of sustained casualties. And it actually makes no difference in real-time battle!! Now we just got to make sure that it doesn't affect AI-recruitment... :) If it doesn't then this can be a great tool for campaign balance! Thx DimeBagHo for spotting that!

EDIT 2: stat_sec_armour's second value seems to do nothing at all for horsemen and infantry. No idea why vanilla sets it as 1 for cavalry.

Previous post updated, it seems we're on a good road here.

Well done to DimeBagHo for pointing this out, Lgk for spotting it originally and for Aradan for the research/confirmation. At last, a way to have some control over that darn autoresolve!

It would be interesting to see if infantry can be affected by this... I doubt it, but who knows. If anyone has any info on that, it could save lots of time.

I can't take credit for the idea - it was suggested to me by Lgk a while back. I used quite large armies in my own tests and settled on a range of values from 1 for basic horse archers to 3 for elite.

*MAJOR UPDATE: It has been known that the second value of the stat_health attribute is not taken into account during battle for non-animal units, hence it was set to 0. However it has been discovered that this value *IS* taken into account during auto-resolve, despite making no difference on battle-map mode. This applies to ALL units (bar elephants, chariots, wardogs and pigs of course, that already make use of it) and essentially means that units can be given specific bonuses or penalties during auto-calc that will not affect their behaviour on battle-map at all; which means we may have at last found a way to partially balance (or unbalance ;) ) auto-resolve. Research and community feedback has shown that certain unit types are overpowered / underpowered, so here are some rough suggestions on how to make things more even for everybody:

• Animal units stay as they are.
• All other units are given 5 secondary hps.
• If a unit has 2 or more prim hps, it gets -1 sec hp for each extra prim hp.
• If a unit is ranged with low-medium missile attack, it gets 1 extra sec hp.
• If a unit is ranged with medium-high missile attack, it gets 2 extra sec hps.
• If a unit is mounted with low-medium charge, it gets 1 extra sec hp.
• If a unit is mounted with medium-strong charge, it gets 2 extra sec hps.

So a horse archer with 1 primary hp, strong charge and strong missile attack gets 9 sec hps, while a foot archer with 1 hp and low missile attack gets 6 hps and an infantry unit with 2 hps gets 4 sec hps and so on…

Thx to DimeBagHo for spurring me to look in that direction and to Lgk for making the first step.

Superb work on expanding this. Now the campaign as a whole will be much more balanced across the factions (player and AI!

:2thumbsup:

now all I need is to remove the autoresolve button and I can make a double balance so Romans on autoC will destroy barbarians but on the field will get destroyed, making life for the player so much more ... er ... brutal .. :evil: ..

nice find guys :2thumbsup:


G

When you get to do that, it'd be grat if you had some feedback - you usually have. ;)

guess a more realistic (read; more accepted) method would be to *somehow* reinforce the AI-army as I dont think most players would like to/appreciate completelly loose the autoC button/option ... do like the idea of double-balance but somehow without it aplying to the player ...


G

Is there a limit to the value that the mount effect can have? I mean, can a unit have mount_effect horse - 100?

How does the mount effect work? I just subtract that value from the unit's attack value to see how it does against horsed units?

If so, can a unit have a negative mount effect with an absolute value that is equal or greater than the units attack value? (e.g. mount_effect horse -12
stat_pri 10 ?)

does the charge value play any role?

Thanks

I have tested up to +-30 and it works, but I am not sure about the cap. You can experiment if you want, and let us know about your findings. The mount_effect is a direct addition/subtraction to a unit's attack when fighting against the particular mount type/class. If a negative value is greater its attack value, then I believe the overall attack will be considered 1 by the engine. No, I don't think charge is connected directly to with mount_effect (though they both affect attack while the unit is charging).

Thanks for the reply.
The thing is, I've given a unit 25 melee, and a -30 in mount effect. They still kill with ease any mounted unit. Both units have the same lethality, both have a sword skeleton and no other attributes that would affect his.

Ok, sorry, just made another test. Giving the unit a -100 against horses has quite an effect. They barely kill any mounted units, so apparently rtw doesn't just subtract the mount effect from the attack value that the unit has, and if negative or zero considers it a 1.

Of course finding an exact relation wont be easy, but if I do notice any linear behaviour while balancing the units I will post about it here.

OK, I see. An additional test would be to make some tests using a specific unit against a specific mounted enemy, find the average result of te battle after, say, 20 times, then give the unit +5 attack and a -5 mount_effect. See if you get a similar avergae to the one you had before.

Thx for the feedback!

Any chance some information regarding the ship line, which appears if category is ship, could be added to the guide? I'm interested in weather or not it is a defined list of possible ship types, and what all of them are as well as their effect on the game.