JoC/MWY Taxonomy of units

[Nemesis2345]

Sharing a bit of my opinion on unit pricing, as I'm quite confused (097 test)

Just look at these prices (all from 1 city in England)

Norman Serjeants -- 2304 florins (!) ~ how about 1200?
Hobilars ------------- 1896 florins (!) ~ 800?
Archers -------------- 1338 florins (!) ~ 600?
Levy Archers -------- 433 florins (ok)
Spear Militia --------- 318 florins (ok)
Miles/Miles Pedites -- 0 florins (ok, feudal)

Spoiler Alert, click show to read: 

I'm confused because I don't understand such high prices on some units. They ain't supermen, their units size aren't numerous, yet cost as much as later full-plate professionals. As a player, I would expect pricing to hint me quality of a unit. It might not be uber historical, but gameplay-wise it's so comforting.

If I were to evaluate current pricing guidelines, it would look somewhat close to this:


0 florins
Feudal units
Upkeep: very high, 1000+ (just like already implemented)

100-300
Poor levy/Ad hoc units
Upkeep: half (no less, lots of men need to eat)

300-500
Units requiring some training (archers, specialized militia)
Upkeep: half

500-800
Cheap professionals, cheap cav, good militia
Upkeep: 3/4 melee and cav, half for missile units

800-1200
Decent professionals, ok cav, excellent militia
Upkeep: 3/4 melee, full melee cav, half missile

1200-1800
Good professionals and cav
Upkeep: 3/4, half for missile

1800-2200
Absolute best extraordinary professionals
Upkeep: half, 3/4 cav


Just tossing an idea.

Yeah , i agree, i also adressed this to JoC and as ive read through the last posts , someone made a program in order to use the EDU file more efficent ,so he might be planning for a full rework of the units.
I already started editing the EDU myself , but i only did it simple by making tiers instead of historical accuracy ( for example Mailed Knights > Feudals > Man at Arms > Kings Men (for England) or Imperial Knights (hre) or French Noble Knights (France)

[Jurand of Cracow]

Yeah, this is something to be discussed indeed. Belovese programe will be very helpful since the moder who was responsible i.a. for the units' prices has left the team years ago and he didn't leave "his" file. The new prices are his last modification I trust(ed) him to be precise. But now I see they're probably not so. Thus Belovese work is vital. Also my (and his) taxonomy in this thread should help to find the "right" prices, ie those based on a precise methodology. Eg. EBII has such, see the link in the first post. JLMP is another one, although it's not derived from the stats and quite subjective in many aspect.

have a look at the explantations at the beginning of the thread - quite much written about the basis for the prices.

Besides, I'm still thinking about the classes of the units. Eg. mercenaries. Maybe the whole class should be dismantled, all the units would be available to all or certain factions, and some of the units would be made available as mercenaries but they would differ only by the recruitment prices, eg. 2x the normal one?
This is already a case, eg. with Woodsmen: they're theoretically mercenary, but they are on of the two main units available for Poland from the barrarcks.

furthermore, I'd add at least two classes: cannons and ships. They should have different set of rules concerning all their characteristics

and the prices mentioned by JLMP are the recrutiment prices. start from the upkeep prices - these matter actually matter as they're payed by the number of turns. The recruitment are actually only the outcome of the 2,66 multiplier.

[Just let me post]

You did a good job JoC.

I checked out the pricing link from the first post and tbh it seems awfully complicated and not very transparent/clear to me. I'm gonna use a simple statement here, risking embarrassment but here it goes: the simplier solution the better.

What do I mean:

1) As you plunged through EDU multiple times and tested the units, you surely have an idea of what units are worth, just by looking at their stats.

2) All else that is needed would be to follow some simple guidelines to stay on track (maybe make a simple poll here?) and just have at it.

3) What's your take on this?



Regarding mercenaries: why get rid of a complety fine feature? Imo mercenaries are awesome for roleplaying, if anything I would kick out mercenaries from city barracks And currently their prices are perfect, I wouldn't touch them.

Same with cannons and ships: if it ain't broken, don't fix it imo.

[Jurand of Cracow]

I might have been imprecise before. To clear things up:
ad 1. not really. The numbers are MWY's (the previous modder). I have some formulas to count the worth of the units but I'll make a regular file with Belovese tool. Then I'll be able to come up with a kind of objective price.
ad.2 - in this respect I don't believe in polls as people usually see only a part of the whole picture. I like discussions but then rules (like those in the table at the beginning of this thread) should dictate prices or other aspects.
ad. cannons and ships - they are not removed, only the type has been clearly stated in the EDU: ship, artillery (before it might have been "feudal levy" or nothing)
ad. mercenaries: the goal is not to remove the feature. Actually, it is rather to extend the feature. I mean: the units that are now in type "mercenary" will still be in-game, they still will be available as mercenaries, but most of them will be at the same time available for certain factions or in certain regions with the regular recruitment (barracks, stables etc.). The pricing will change (if a mercenary unit is included in militia, recruitment price = upkeep, if feudal levy: recruitment price = 2,66 x upkeep, etc.). The recruitment price will be valid if recruited from the regular buildigns. If recruited like mercenaries, the prices will be 16/3, as the regular pricing in the eg. EB II. But this is a tentative idea.

[Just let me post]

Ad. 1 Ok, I'm curious how this will turn out then

Ad. 2 Understandable, considering the above

Ad. mercs
I don't know, on one hand there is no con (other than filling up more recruitment blocks in cities so you'd need to scroll down even more frequently later in game + more voices for me to add heh) but... I'm not sure even why? I mean yes, militias are sustainable, while mercs are greedy bastards, but I think we have enough units already for good variety.
It might be discussed later, surely.

[paleologos]

Will you take into account the faction strengths when you assign recruitment costs and upkeep?
For example, will the Turks and other steppe factions be able to get rather good horse archers at a lower price, or will their "special strength" units have more soldiers in them?

[Jurand of Cracow]

Will you take into account the faction strengths when you assign recruitment costs and upkeep?
For example, will the Turks and other steppe factions be able to get rather good horse archers at a lower price, or will their "special strength" units have more soldiers in them?

Yes, this is my intention.
I'll probably play with formula (it's rather complicated, eg. the impact of morale is =POWER(LOG(Q8;10);4)*10 - I even don't remember why and how I got to it...), then apply some subjective coefficients etc.
The number of soldiers will rather be standarized:

Standardization of the sizes of the units (multiply by 2.5 if you play on huge):
- bodyguards 10
- Feudal Knights: 24
- Dismounted Feudal Knights: 40
- Mounted Crossbomen: 28
- various specialized cavalry: 30/32
- Archers & Crossbowmen: 48
- specialized troops, esp. with axes: 52
- Spear Militias and Speamen: 60
- special Militias (Italian, Lithuanian): 80

Anyway, the prices for 097 won't be the definitive ones, I hope to gather opinions from various people to adjust them.

[XXZit]

Jurand, I think the the changes you outline here are incredibly well thought out and I wholeheartedly support them. The numbers are just details, what matters is the map you've established.
I've implemented a new feature in my program to help you with the unit cost changes. Let me explain how:
You can already establish conditions to selectively apply different modifiers to the base upkeep cost of all units. Alternatively, you could scrap the current upkeep cost of all units, and set the cost to 0 or a base minimum value in a single program run, then use conditions to selectively change this base price. Depending on the parameters you look at to reach the final upkeep price, i.e if they are all present EDU records, you could do this automatically with the program. You may have to run it a few times with different conditions, but it can do it. Now, with the new feature I added, once that base upkeep price of all units is established, you can automatically set the recruitment cost of all the units in the EDU in a single pass of the program.

What the new feature does, it allows you to modify a record based on the value of another record and a value you provide. So you can tell the program to look at the upkeep cost and then have it divide or multiply that cost, whatever it may be, by 16/3 or whatever value you decide to use, and save the result of that operation in the recruitment cost field.

EDIT: I just re-read your OP and realized you plan on using, different ratios for different units, in that case it can't be done a single pass, you'll have to run the program multiple times with differing conditions or filters for each group that uses a unique ratio.

I hope this helps. If you have a request on additional functionality that can help you automate this task or any other, just ask. The same goes for anyone here. I've been adding things that I myself find useful or need, but my aim is to release something that can benefit the community as a whole.




PS: Now that I see you mention formulas and numbers, I had much success modifying and standardizing weapon stats when I was playing the Witcher 3 by using a quadratic equation of the form: ax + bx + c. Perhaps you should give it a try; you can use an equation graph visualizer to see the curve and how different values affect it.

[Nemesis2345]

I like the idea presented by JoC as well , its simple yet very necessary right now as troop costs and upkeep are all over the place.


There is only 1 problem right now........ i cant wait!!!!

[Jurand of Cracow]

@XXZit:
- actually the taxonomy is not so well thought out for the late game. I'm not sure how the gunpowder units fit in. But it's to be corrected on the way.
- I'll have a look at your program along the the Belovese's one. I'll give you feedback.
- yep, the ratios will be different for different types of units, and after there may be some changes due to local circumstances (eg. English Longbows will perhaps be made cheaper than the crude comparison of the stats is concerned. Also various types of the Lithuanian and Scottish troops).
- the problem with formula it that diffent stats should have different weights, and then their value rise at different speed. It's why I'm inclined to use logarithms as a tool for normalization. Maybe I'll also come up with quadratic equations. Actually, knowing your savvy with the numbers, I'll expect a perfect scrutiny from your side ;-)

[k/t]

The way I ended up implementing Feudal units was making them recruitable from a special building called Fiefs. That building gives a big penalty to trade income, since the ruler is no longer receiving as much revenue from the province, as lands have been granted to nobles. They have free_upkeep, since you don't have to pay them when they're staying at home, and they only cost 100 dollars to recruit, since they have to supply their own equipment, but they have pretty high upkeep, so you don't want to lug them around too long (to simulate the fact that their duty ended quickly and you'd have to pay them to stay longer (but they could refuse).

In my BC submod, I remember tying the Ghulams and Mamluks to the slave market buildings as well as to the barracks.

Take at look at my unit pricing ideas, maybe something in there sparks your imagination.

Who's MWY?

[kostic]

MWY is a former SSHIP modder who we haven't heard from for a long time.

[Jurand of Cracow]

The way I ended up implementing Feudal units was making them recruitable from a special building called Fiefs. That building gives a big penalty to trade income, since the ruler is no longer receiving as much revenue from the province, as lands have been granted to nobles. They have free_upkeep, since you don't have to pay them when they're staying at home, and they only cost 100 dollars to recruit, since they have to supply their own equipment, but they have pretty high upkeep, so you don't want to lug them around too long (to simulate the fact that their duty ended quickly and you'd have to pay them to stay longer (but they could refuse).

In my BC submod, I remember tying the Ghulams and Mamluks to the slave market buildings as well as to the barracks.

Take at look at my unit pricing ideas, maybe something in there sparks your imagination.

Who's MWY?

MWY was an ingenious moder, a lead developer of the SSHIP in the past, along Fair Prince and Lifthrasir.

We do have a kind of Fiefs building in the SSHIP. It's called Local Guards / Provincial Council and provide the Feudal units. Available only in the cities because castle provide knights with other buildings. Big penalty to trade income is a good idea indeed. I may actually implement it, as a trade off. Recruitment cost - yes, I'm thinking about re-introduction, mainly for the AI to treat them as a more normal unit.

Your submod will modify SS6.4?

[k/t]

It was for BC.

I vaguely remembered MWY but I also thought it was short for something.

[Jurand of Cracow]

In the SSHIP v. 0.98 there'll be once change: all the knights units (feudal/iqta/pronoia etc.), including the religious, units will have a recruitment price of 50%. This is both because of historicity (an initial payment was always required) but also for the gameplay/engine - the AI is confused about 0 price.

[Nemesis2345]

In the SSHIP v. 0.98 there'll be once change: all the knights units (feudal/iqta/pronoia etc.), including the religious, units will have a recruitment price of 50%. This is both because of historicity (an initial payment was always required) but also for the gameplay/engine - the AI is confused about 0 price.

I was going to suggest this after i noticed the AI throwing stacks of Cavalry over and over and over at me. It's unfun.

[QKuhlmann]

Thank you, JoC, for the quick answer and for putting me to this thread. My save game, Novgorod in the year of 1210, is here: https://drive.google.com/file/d/1spo...usp=share_link

As for my claim, that there are too many elite units in the game: I was talking about the Landowners units, cavalry or infantry, that initially are quite cheap but have high upkeep. They are not exactly elite, as I understand now. I think, there are too many of them in the KI-armies, maybe because of the low initial price and the relatively fast replenishment. As I remember from playing earlier versions of SSHIP, the problem was even bigger, when the units were free to recruit, but it is still there now and leads to armies, that are not very diverse. For players armies the feature is not so problematic, because the player can plan better and will not recruit as many of them, although they are available, because he fears the high upkeep. As player, I often recruit them in case of emergency. I understand the concept behind it, that a governor can call those smaller landlords and their entourage to arms very quick, but they cost him, if he keeps them in arms for longer. But I think it is difficult to balance it with the M2TW-Engine. My impression ist, that the KI often thinks: "O well, such a decent unit, and so cheap! Let's recruit them instead of the other 12 available kinds of guys, and don't think about the upkeep, that they will cost tomorrow." Of course, this is only my opinion, and a decision, to make them cost 100% would change the whole game-balance and could lead to new problems.
Best regards QK

[Macaras]

As for my claim, that there are too many elite units in the game: I was talking about the Landowners units, cavalry or infantry, that initially are quite cheap but have high upkeep. They are not exactly elite, as I understand now. I think, there are too many of them in the KI-armies, maybe because of the low initial price and the relatively fast replenishment. As I remember from playing earlier versions of SSHIP, the problem was even bigger, when the units were free to recruit, but it is still there now and leads to armies, that are not very diverse. For players armies the feature is not so problematic, because the player can plan better and will not recruit as many of them, although they are available, because he fears the high upkeep. As player, I often recruit them in case of emergency. I understand the concept behind it, that a governor can call those smaller landlords and their entourage to arms very quick, but they cost him, if he keeps them in arms for longer.

We actually modified the recruitment offset (armoured sergants, cavalry etc) and the availability of the landowner units. It influences visibly the composition of AI armies, it will be available in the next version. I think also you are right that AI considers as a big factor the cost of recruitment, so that would be a possible way to lower the numbers of the knight units, but so far it was not necessary. I tested the last varsions and it really is noticable - the lower number of landowner and cavalry units. Its all coming soon

[QKuhlmann]

Thank you very much, Macaras! This is very good news! I am looking forward to the next version.

[j.a.luna]

We actually modified the recruitment offset (armoured sergants, cavalry etc) and the availability of the landowner units. It influences visibly the composition of AI armies, it will be available in the next version. I think also you are right that AI considers as a big factor the cost of recruitment, so that would be a possible way to lower the numbers of the knight units, but so far it was not necessary. I tested the last varsions and it really is noticable - the lower number of landowner and cavalry units. Its all coming soon

Perfect @Macaras! I have the same situation regarding the landowners/noble units, in my kievan rus campaign the cumans are destroying my kingdom and it is difficult to fight against them and with my militia troops... these cumans have quite a lot of Beck Druzhina units (landowners) they are very strong against my troops, I currently only have one Druzhina unit!!! when they in an army of 20 units can have up to 10 or more landowner units... I think the knights/noble units should be less than 1/3 of the composition of an army, except for some exception. They should be mainly militia/lev units and according to their wealth increase the quality of the troops and mercenaries...