PDA

View Full Version : Investigation of Bribery Costs



therother
11-12-2004, 01:06
Some previous work that may be of interest to this topic:

Preliminary 1.1 bribing research (https://forums.totalwar.org/vb/showpost.php?p=626490&postcount=9)
Public Order and Bribe Cost (https://forums.totalwar.org/vb/showthread.php?t=39681)
The Influence of Influence for diplomats (https://forums.totalwar.org/vb/showthread.php?t=39677)

QwertyMIDX
03-01-2005, 07:19
Has anyone done any serious research to determine where the costs come from? I was wondering if the price of the units being bribed was a factor in determining the costs, for example, if I were to increase the price of Iberian Infantry to 500/250 would the cost of bribing an army with units of Iberian Infantry increase? I will check into this on my own and post anything I find, but any input would be much appreciated.

hung41584
03-01-2005, 08:35
seems you have to pay a base price for bribes and then the rest of the units are moderately more expensive than usual. I'd say about 2-3 times their value. Plus the base. Its actually better to bribe larger armies than smaller ones, which is quite the opposite in reality.

khelvan
03-02-2005, 07:55
Bump - anyone have information on this?

Pode
03-02-2005, 16:29
In one of the homebrew mods that was trying to force better recruiting, the guy reversed the cost ratio of units, such that peasants cost much more than cohorts. He threw in an offhand comment that bribing peasant rebels became ludicrously expensive.

QwertyMIDX
03-03-2005, 00:25
Ok, I did some testing, I bribed a rebel army with 2 units of peasents, 1 iberian infantry, and 1 belric slingers. With vanilla unit costs it cost 8640. I then increased the cost of iberian infantry to 1000 (roughly 4x) and the cost of bribing the army increased to 15040.

Simetrical
03-03-2005, 04:39
Interesting. Vanilla recruitment costs would put the cost for that army at 100 + 100 + 240 + 470 = 910 denarii, and bribing it cost 8640 (or about 9.5 times the cost). Changing that to 100 + 100 + 1000 + 470 = 1670 denarii, the bribe cost became 15040, or about 9 times the cost. This does indeed suggest a reduction in cost for more expensive armies.

Now, let's assume the bribe cost function is linear. That would mean that it could be expressed as y = mx + b. The two data points give us 8640 = 910m + b and 15040 = 1670m + b. Subtract the former from the latter, and we know that 6400 = 760m, or m ≈ 8.42. Then we can solve for b: 8640 ≈ 7663.16 + b, which means that b ≈ 976.84. The final function, therefore, should be y ≈ 8.42x + 976.84. Get another data point, and we can figure out if this is right. (I'd get the point myself, but I have to go to bed soon, so I don't have time.)

Edit: Or, of course, it could be based on custom battle recruitment cost. This would make more sense, since it would take into account two-turn training times, higher-level barracks, and similar cost reducers.

-Simetrical

Fisherking
03-04-2005, 17:42
Hay, that is just ridiculous! :furious3: While the costs in the original may have been a bit low, now they are just stupid, especially sense almost everyone goes home.

There needed to be other considerations made in the bribery costs to actually reduce them... like the likelihood that they were going to die anyway & there is nothing wrong with bribing away faction hairs or bribing them back....who was the Greek General that switched sides so many times & was finally assassinated?

If you are playing the game properly you are not going to have enough cash on hand to pay that much for some insignificant group like that anyway. You would just kill them for the experience for your units.

It is not a straight forward calculation for all units either, you know. Different factions have different calculations for them... Look at the cost to bribe an egyptian vs. say a greek. I know it was that way in the original & assume that it remains so in 1.2 :dizzy2:

Simetrical
03-04-2005, 19:45
Different factions have different calculations for them... Look at the cost to bribe an egyptian vs. say a greek.Have you determined that in a controlled experiment? That is, have you tried to bribe an Egyptian unit, and then an exactly identical Greek unit?

-Simetrical

Quietus
03-05-2005, 08:02
I think bribery cost is dependent on average settlement income as well (other factors include influence of the diplomats and possibly unit cost).

I noticed when I get extremely rich, the bribes skyrocket (although I don't usually bribe but do it just to test). Here's what you do.

1) Destroy all your ports (or other moneymaking buildings). Click end-turn (it doesn't work without clicking).
2) Bribe any unit(s).
3) Reload and bribe the same unit(s).
4) Compare the costs. The former is cheaper. (Don't do #3 first since it might change your diplomat's influence).

I'll have to do more tests to be completely sure. Next, I'll give away my richest provinces to lower the province count and the average income to see any difference. ~:)

therother
03-05-2005, 13:43
There are a few threads/posts scattered around with some interest to this topic:

Preliminary 1.1 bribing research (https://forums.totalwar.org/vb/showpost.php?p=626490&postcount=9)
Public Order and Bribe Cost (https://forums.totalwar.org/vb/showthread.php?t=39681)
The Influence of Influence for diplomats (https://forums.totalwar.org/vb/showthread.php?t=39677)

Quietus
03-12-2005, 00:23
Here's some update:

1) Bribe value is NOT fixed. It is changed every turn.
2) There is definitely a base value that is being factored. Currently the base value is tied to either Gross or Net income (rather it is a function of). The factoring is FIXED but the base values changes per turn.
3) Upgrades to a unit seem to be a big factor (shield, weapon, or experience).
4) Generals seem to be a good factor as well.
5) Cost seem to be a factor as well.

So the current proposal is:
Bribe value = Base value x Factors.
Base value = the function of income. Higher income means higher base value resulting in higher bribery costs. I still don't have an exact equation for this but the trend supports this.

Does anybody know where to find the prices (and unit strength at huge setting) for each unit? I can't always see the prices and the units in the test are not always full strength.

TIA.

Edit: Here's some primary data where I based my proposal (reference: old Julii game, VH/VH from 1.0 loaded to 1.2; p=projected).


3^Condition|Provinces|2 Peasants|Thracian stack|pGross Income|pNet Income
1^Normal + 0 turn|35|3986 d|197898 d|99212 d|45467 d
1^Normal + 1 turn|35|3695 d|185013 d|91614 d|38665 d
1^Destroyed Ports+ 1 turn|35|3379 d|169172 d|77208 d|27787 d
1^-6 Provinces + 1 turn |29|2872 d|143827 d|67426 d|22311 d


Now divide the "2 Peasants" by the "Thracian stack"
197898/3986 = ~49.65
185013/3695 = ~50.07
169172/3379 = ~50.07
143827/2872 = ~50.08

Now, the ratio is FIXED, but the actual prices are changed! This tells you that the factors are the same but a base value is being changed!!

now for the "Normal + 1 turn", I bribed a diplomat
diplomat cost = 2463 denari
2 peasant cost = 3695 denari
2463 denari/2 = 1231.5 denari
1231.5 x 3 = 3694.6 denari

There's a base value that is being factored!! Of course there are a lot of factors as I've outlined earlier. That's why I need the prices since not all the units in the "Thracian stacks" are full.

It is possible that a unit is divided into base values and a set of those values constitute a unit (along with other factors). And this is calculated for the bribe price. Look, as the income goes down so does the bribe prices. However, like I've said, there's still no direct or exact correlation that can be derived. ~:)

Simetrical
03-13-2005, 03:03
Does anybody know where to find the prices (and unit strength at huge setting) for each unit?Try looking in export_descr_unit.txt. The section at the top explains the notation below.

-Simetrical

Quietus
03-13-2005, 09:23
Thx. (Ahh, I "hate" looking at actual stat sheets. :dizzy2: )

Quietus
03-21-2005, 07:58
Update: Here's the new equation for the bribes:

1) This works for a regular bribe with full strength units w/o chevrons.
2) Unit Chevron/Experience is a factor.
3) Armor and Weapon upgrade doesn't seem to be a factor
4) The diplomat's influence doesn't seem to be a factor for units.

New equation:
Bribe value = (Unit Recruitment Cost/10)(Base Value)(Factors)
Where Base Value is the function of Income. (Higher Income means higher Base Value).

Old Julii game, VH/VH, from 1.0 loaded to 1.2, Large units.

A=Armor
W=Weapon
C=Calculated
ELC=Early Legionary Cohort
LC=Legionary Cohort
We also asume the Factor here is 1 so the equation works.


1^Bribe Value|For what?|Per Unit|Recr. Cost|Recr.Cost/10|Upgrades|C.BaseValue
3^15513 d|3 Warbands|5171 d|420 d|42 denari |None|123.11904
3^3693 d|2 Peasants|1846 d|150 d|15 denari |None|123.06666
3^6279 d|1 C.Axemen|6279 d|510 d| 51 denari|A1&W1|123.110765
3^16621d|1LC+1LC|------|740 d & 610 d|74 d+61 d|None|123.11851


5171 d/42 d= 123.11904
1846 d/15 d=123.06666
6279 d/51 d=123.110765
16621d/(75 d + 61 d = 135 d)=123.11851

When the unit has a chevron, it doesn't work. When a unit is not full strength, it doesn't work either. So those are two factors.

Bribing diplomats seemed to use a different set of a factors but the base value likewise is being used and computed

BD=Bribed Diplomat
AgeD=Age of Diplomat

1^Bribe Value|Diplomat vs. BD Influence|AgeD|AgeBD|Bribe Value/Base Value
3^10835 d|8 vs. 4|52|46|88
3^2462 d|7 vs. 0|55|59|20
3^9357 d|1 vs. 3|23|50|76
3^5417 d|1 vs. 4|33|59|44
3^3201 d|3 vs. 1|35|59|26

10835 d/123.1 = 88.02
2462 d/123.1 = 20.00 etc. for other values.

Note, 1 vs. 4 is lower than 1 vs. 3 (traits and ancillaries a factor?)

I'm also trying to tie in the Base Value to income but no solid results yet. For this turn: Projected Net Income/Base Value=?
37052 d/123.1=300.99=301.0
Will do more tests.

~:)

Avicenna
02-26-2006, 15:17
....who was the Greek General that switched sides so many times & was finally assassinated?

Alcibiades? The Athenian general who helped the Spartans against his own state, then went back to Athens, where he was killed by the Spartans as they burned the Athenian walls.

MajorFreak
05-22-2011, 03:09
hmmm...i was wondering why i was able to bribe so cheaply. And here i'd thought it was because i hadn't patched yet. lol. 'Base Value' being revenue, huh? I guess there's something to be said about trying to keep <50k in the treasury. The usual charge for a 25yr general these days is about 3500gp...i'm just lucky the enemy doesn't get slaphappy with bribes against my troops. lol...speaking of which, a captain with a stack usually accept bribes of <1000gp. seriously!

EDIT: belay that...i'm playing unpatched
D. Diplomacy: Diplomacy is practically nil after patch 1.2. Map Selling is curtailed and Bribery cost was increased. However, the protectorate was fixed (research pending).