Unit pricing

[Quinn Inuit]

I'm up late with a sick puppy, so I'm posting this.

I'm planning on cranking up my unit costs. I like the idea of a full stack being a rare occurrence in-game, as opposed to a once-a-turn mail-order surprise from some of the larger factions. My problem is that I've finally got the game balanced where factions are building realistic and balanced armies, and I don't want to break it.

So, how would you recommend increasing costs? In my research, I've noticed that the AI consistently undervalues the effect of high upkeep costs and seems to make most of its decisions based on up front cost. Is this your experience as well? If so, that's annoying, since I was hoping to make some units low cost/high upkeep to reflect that they were citizen levies.

[AngryTitusPullo]

The all time balancing nightmare.

Give too low for units, then the AI will field militias and the player elites. Gives too high only work in earlier campaign to stop players acquiring elites...

Give elites lower price than militias makes players abuse them by easily deploying elites... :hmmm:

[phoebusapollon]

how did you get the AI to produce balanced armies? its such a bother (and boring) to fight all-phalanx armies with no cavalry

and its no fun at all

[AqD]

how did you get the AI to produce balanced armies? its such a bother (and boring) to fight all-phalanx armies with no cavalry

and its no fun at all

You can modify the behavior in descr_strat.txt. However I think the problem in RTR is that cavalry cost/men ratio is really too high. I tried to lower the cost by 25%, or raise their number by 33% (same cost/men), and some AI factions, namely gauls and carthaginians, begin to recruit light cavalry, sometimes up to a quarter in an army.
________
Recall Avandia

[binTravkin]

Yes I'd like to hear that too

[Quinn Inuit]

The all time balancing nightmare.

Give too low for units, then the AI will field militias and the player elites. Gives too high only work in earlier campaign to stop players acquiring elites...

Give elites lower price than militias makes players abuse them by easily deploying elites... :hmmm:

Well, I can't stop the player from building all-elite army. I mean, if someone wants to make all-principes armies, that's the way they want to play. So I don't even worry about that. My goal is to make the AI play well, and leave the player's style up to them.

The AI is driven to deploy elites, though (anyone remember the all-hypaspistai stacks of RTR6G?), so I'm not having trouble with that. Throttling their army production is the key, IMO.

how did you get the AI to produce balanced armies? its such a bother (and boring) to fight all-phalanx armies with no cavalry

and its no fun at all

Yes I'd like to hear that too

Well, it took a lot of unit cost tweaking, limiting recruitment of all elites to the royal barracks, and, most importantly, limiting placement of the top military buildings to larger cities. That forces the AI to rely on local units, and slows its growth. (I realized that was the problem when I saw a Ptolemaic Royal Barracks in Side, which has all of 2300 people.)


Here are the results. These are the two major Macedonian frontline armies that just showed up on my doorstep:

[binTravkin]

Are your modifications a part of Extended Realism mod?
If not, do you plan to release them as a separate mod?

[Quinn Inuit]

They'll be in the 3.3 version, which is nearly complete.

[binTravkin]

Cool, gonna get that one.
Hope it works with my Platinum 1.9 + Alex.exe + 4tpy setup.

[Quinn Inuit]

It'll probably work with Alex.exe if you make the appropriate change(s). I've heard it's easy. And it incorporates the 4tpy already, so you don't need to worry about that.

[Stormwolf]

In my RTR LOGIC mod I considered each unit number by a factor of 10.
So an army of 4,000 is really 40,000 and therefore I increased the unit recruitment cost by a factor of 10. the upkeep cost is increased as well.
Furthermore all generals units have only ONE hitpoint (as do spartans, etc albeit with some increased stats).

Basically this creates the types of battles that were seen historically (one army vs one army, instead of 50 stacks vs 50 stacks) however to see true 200,000 vs 200,000 battles we must wait for RTW2 (please make the UNIT SCALE realistic like the show TIME COMMANDERS)

*this can be done my increasing the units per army to perhaps 100 instead of 20* *can modify units size in the options menu to allow slower PCs to run them?*

[Mikail Mengsk]

I didn't understood: you modified only the unit's cost?

[Quinn Inuit]

I didn't quite get that, either. Stormwolf, I encourage you to try the ExRM. I think you'll find it a solid balance of increased cost and gameplay.

[Stormwolf]

Modifications:

Unit Recruitment Price Increased by 10x
Unit Upkeep increased
Generals' Bodyguard hitpoints reduced to one

Reasons: 1)To reduce the number of battles
2)To make the unit numbers more realistic (by assuming that 4,000 soldiers represents 40,000)
3) To provide an interim sollution until Rome Total War 2 allows for full scale battles (involving hundreds of thousands of troops) as the Rome Mod. created for the TV show Time Commanders

My logic mod makes RTR much more fun to play as armies cannot be replaced quickly (an army of 20 units usually costs between 100,000 to 200,000 denarii). Therefore engaging an enemy army is a punctuated event that usually decides the strategic balance in the region (as real battles actually did).

If you want an example of what I mean you can read my AAR
"Chronicles of Pontus".

[Jamey]

At 10x unit cost and recruitment, I'm guessing that Germania can afford to garrison each settlement with one unit and send the other unit out to conquer?

Seriously, what Quinn has done with ExRM is very good. You should try it.

[Stormwolf]

Actually my mod is quite realistic. Normally germania should suffer from a chronic lack of equipment due to an absence of industry and trade. Upkeep is significantly less than 10x. It is closer to 2.5X normal cost. The recruitment is so high to avoid the typical RTW*economic rebound* of losing so many soldiers. It took Rome a full 10 years to rebuild its army after Cannae, however in RTW units are cheap.

This system works extremely well. Generals bodyguards are no longer invincible and battles are decisive. Using the 10x multiplier you can also recreate historical battles accurately.

Example: Carrhae

Parthia:
1 Unit Cataphract (800)
1 Unit General (250)
9 Units Horse Archer (9,000)

Rome:
4 Units Equites Alares (3200)
1 Unit General (200)
2 Units Funditores (3200)
18 Units Legionnares (32,000)

An accurate representation with the 10x multiplier.

The basic lesson is that rome's unit scale is far too small and the sequel should accurately present battles in the size and scope in which they occured. This can be achieved by implementing the same engine that SEGA used in the design of the TV show Time Commanders where contestants controlled a modified game of Rome Total War with much larger unit scales (I.E. 70,000 troops on the field at once)

As to the upkeep costs,
If you examine historical data you will note that a soldier was usually paid approximately 1 mina per day (if he was lucky and not fighting as a slave) which is the equivalent of $10 USD. Wages were low and equipment was expensive (even hannibal's troops confiscated the Romans' armour after Trebia).

Effectively a soldier was paid $3,000 USD per year and normally food was conficated by foraging the local terrain/populace. A Gladius, Lornica, and Helmet would easily fetch 100,000 USD. Unit recruitment and armament was far more expensive than pay/feed.

Essentially in this amazing mod. a single unit of 160 troops represent 1600. Therefore a roman legion containing one unit each of Hastati, Principes, and Triarii plus the general is approximately 5,000 troops (the exact number of troops in a roman legion). Therefore a typical roman army of two legions would contain 2 Hastati, 2 Principes, 2 Triarii, and six Units of Allied Auxilliaries for a total of 20,000 troops.

It is true that settlements are now understaffed however this is realistic.
Most settlements should only have a limited garrison as it was far too cost prohibitive to do so. Alexaner of Macedon conquered an RTR equivalent of 50 settlements, most of which were unocuppied as the persian army was destroyed in the field at Issus (Egypt was a walkover and even Babylon was practially undefended). It is the unrealistic way of packing 20 unit stacks to garrison a city that the RTR player can charge the largest possible tax.

In my campagins I usually play as bactria. They can afford basically just one or two armies with which they must defeat the Parthians and the Seleucids.
If the Bactrians lose that army, they are essentially finished as the Seleucids have the money to hire mercenaries or fresh Citizen levies while the Bactrians are essentially fried. This is more realistic than the painfull RTW attrition marathons we see where an army can be rebuilt in a single year.

I can send my data files to anyone who dares to try my Logic Realism mod

[Mikail Mengsk]

Actually my mod is quite realistic. Normally germania should suffer from a chronic lack of equipment due to an absence of industry and trade. Upkeep is significantly less than 10x. It is closer to 2.5X normal cost. The recruitment is so high to avoid the typical RTW*economic rebound* of losing so many soldiers. It took Rome a full 10 years to rebuild its army after Cannae, however in RTW units are cheap.

This system works extremely well. Generals bodyguards are no longer invincible and battles are decisive. Using the 10x multiplier you can also recreate historical battles accurately.

Example: Carrhae

Parthia:
1 Unit Cataphract (800)
1 Unit General (250)
9 Units Horse Archer (9,000)

Rome:
4 Units Equites Alares (3200)
1 Unit General (200)
2 Units Funditores (3200)
18 Units Legionnares (32,000)

An accurate representation with the 10x multiplier.

The basic lesson is that rome's unit scale is far too small and the sequel should accurately present battles in the size and scope in which they occured. This can be achieved by implementing the same engine that SEGA used in the design of the TV show Time Commanders where contestants controlled a modified game of Rome Total War with much larger unit scales (I.E. 70,000 troops on the field at once)

As to the upkeep costs,
If you examine historical data you will note that a soldier was usually paid approximately 1 mina per day (if he was lucky and not fighting as a slave) which is the equivalent of $10 USD. Wages were low and equipment was expensive (even hannibal's troops confiscated the Romans' armour after Trebia).

Effectively a soldier was paid $3,000 USD per year and normally food was conficated by foraging the local terrain/populace. A Gladius, Lornica, and Helmet would easily fetch 100,000 USD. Unit recruitment and armament was far more expensive than pay/feed.

Essentially in this amazing mod. a single unit of 160 troops represent 1600. Therefore a roman legion containing one unit each of Hastati, Principes, and Triarii plus the general is approximately 5,000 troops (the exact number of troops in a roman legion). Therefore a typical roman army of two legions would contain 2 Hastati, 2 Principes, 2 Triarii, and six Units of Allied Auxilliaries for a total of 20,000 troops.

It is true that settlements are now understaffed however this is realistic.
Most settlements should only have a limited garrison as it was far too cost prohibitive to do so. Alexaner of Macedon conquered an RTR equivalent of 50 settlements, most of which were unocuppied as the persian army was destroyed in the field at Issus (Egypt was a walkover and even Babylon was practially undefended). It is the unrealistic way of packing 20 unit stacks to garrison a city that the RTR player can charge the largest possible tax.

In my campagins I usually play as bactria. They can afford basically just one or two armies with which they must defeat the Parthians and the Seleucids.
If the Bactrians lose that army, they are essentially finished as the Seleucids have the money to hire mercenaries or fresh Citizen levies while the Bactrians are essentially fried. This is more realistic than the painfull RTW attrition marathons we see where an army can be rebuilt in a single year.

I can send my data files to anyone who dares to try my Logic Realism mod

Very intresting but i think Quinn Inuit is right: the AI factions risks to be unable to expand.

Also, what about city happiness? Many cities MUST be HEAVILY garrisoned or they will revolt, how do you avoid it?

[Stormwolf]

1) "Very intresting but i think Quinn Inuit is right: the AI factions risks to be unable to expand.

2) Also, what about city happiness? Many cities MUST be HEAVILY garrisoned or they will revolt, how do you avoid it?"

1) The AI factions battle against one another just as they normally would do, however they are more challenged by the rebel buffer factions separating them. This creates a more historical policital map with Macedon being much slower in its tsunami effect. The AI's difficulty can be increased by setting it to H/VH. On very hard with the 10,000 per turn bonus it can still recruit soldiers with a high level of expediency and the "monopoly factor" becomes far less pronounced (where the AI's diplomats bribe away everything in existence).

2) City happiness now relies on lowering taxes and creating buildings of appeasement. Even Rome with its inexorable wealth was faced with persistent revolts from its peasants/slaves/Italian cities which despised Roman occupation. One must remeber than any and all settlements were somewhat apprehensive against any foreign occupation whatsoever; a fact that is present to this day.

Therefore historically revolts were a rule and not the exception, a fact that should be represented realistically in RTR. Therefore to maintain public order one must maintain a careful monetary organization between troops, taxes, and most useful of all --> skilled/influential governors.


Above all, the traits of the managers / generals become far more important as indivudial battles will usually decide the fate of your Faction. Furthermore there is more incentive to hunt down small groups of enemy forces and battles are more a matter of fun rather than a chore (how many of us become bored of fighting stack after stack of Macedon's Phalanx armies)

Lastly, when playing against rome, Pyrrhus now poses a far more realistic threat. (since his army is accurately represented as being 25,000 strong approximately).

Rome's citizen soldiers were mostly always recruited from the city itself and thus moving the 10% training discount Retinue into Rome creates the incentive to recruit soldiers historically (since a three-unit legion now costs the player upwards of 20,000 Denarii) and the discount retinue is perhaps the most valuable starting unit for the roman faction. I suggest that everyone try this mod for a few battles and see the massive impact on accurizing the ways of RTR.

[Stormwolf]

I have provided the files to everyone who wishes to sample the Logic 1.03 Mod

Changes ~

1) Unit Recruitment Costs Increased (10x)
2) Unit Upkeep Costs Increased (2-3x)
3) Changes in Stats for all Generals Units
4) Changes in Stats for Elite Infantry Units

*suggest to always play with Unit Scale HUGE (160 troops represents 1600)*

[Quinn Inuit]

Um...I imagine that's hard for the player, but how does the AI do? At those prices, I'd be worried that the AI would be unable to expand, since it doesn't manage its economy as well as a human. Even with what I've done, some of the barb factions don't expand well (I think I've fixed that in the next patch). Increase prices more and they may not at all.

Also, bear in mind that different nations had different unit recruitment strategies. Recruiting citizen soldiers had very different costs and benefits than recruiting standing armies or mercs.


For the record, I increased unit prices by ~2x and upkeep by 1.5x.