Retinues in 2.5.2?
- Thread starter Kardo
- Start date
-
We have updated our Community Code of Conduct. Please read through the new rules for the forum that are an integral part of Paradox Interactive’s User Agreement.
You do need to account for cost somehow. I mean for example a Knight retinue costs 1200 and a Defense retinue costs 600 retinue cap. In an actual game the choice is not 1 Knight or 1 Defense, but 1 Knight or 2 Defense. So if you want to compare the two retinues that's what you should be comparing.
We also can't assume retinues scale linearly or quadratically, so we can't simply divide your result with the cost of a single retinue. If we want to do this properly the best approach would be to pick a given retinue cap and fill it up with retinues and then test for that. I'd recommend going for 36K retinue cap. That can't neatly be divided by all retinue costs, but it can be by most of them and the ones that don't fit neatly can still aren't too much off in most cases.
So the test would be 36K retinue cap worth of retinues vs a standard enemy, perhaps 36K of you 1-for-everything units.
As for myself, I'm currently making a test mod for retinues. It's a little rough, but so far I made all retinues and cultural buildings available to human players (and unavailable to non-human characters), retinue cap is massively increased, retinues are free (in terms of gold cost) and reinforce to max size in a few days, and it is possible to clear any constructed cultural building. What I'm still working on is a decision that spawns event troops that mimic the distribution of troops in a castle with all tech level 4 buildings constructed (except the cultural building). There will be one available that creates the exact levy mix (worth 1978 retinue cap points) and another than creates a levy mix that adds up to exactly 2000 retinue cap points. It will also be possible to create courtiers with 8, 12, and 16 Martial to command the test armies.
We also can't assume retinues scale linearly or quadratically, so we can't simply divide your result with the cost of a single retinue. If we want to do this properly the best approach would be to pick a given retinue cap and fill it up with retinues and then test for that. I'd recommend going for 36K retinue cap. That can't neatly be divided by all retinue costs, but it can be by most of them and the ones that don't fit neatly can still aren't too much off in most cases.
So the test would be 36K retinue cap worth of retinues vs a standard enemy, perhaps 36K of you 1-for-everything units.
As for myself, I'm currently making a test mod for retinues. It's a little rough, but so far I made all retinues and cultural buildings available to human players (and unavailable to non-human characters), retinue cap is massively increased, retinues are free (in terms of gold cost) and reinforce to max size in a few days, and it is possible to clear any constructed cultural building. What I'm still working on is a decision that spawns event troops that mimic the distribution of troops in a castle with all tech level 4 buildings constructed (except the cultural building). There will be one available that creates the exact levy mix (worth 1978 retinue cap points) and another than creates a levy mix that adds up to exactly 2000 retinue cap points. It will also be possible to create courtiers with 8, 12, and 16 Martial to command the test armies.
Last edited:
Valid points. Not too much work to restructure the spreadsheet for survivor ratio / cap and survivor ratio / gold. Also not too hard to change the standard opponent to light infantry so that it's a more intuitive comparison to real game situations.
I think since we convinced ourselves that morale scales linearly with kills and kills linearly with damage, we can just do the division. That said, the 36K cap test in your new test mod would be far more accurate and serve as confirmation.
What would be most helpful to my work, if there are any others still reading this thread, is if someone would compute the odds of each tactic for each pure retinue, multiply that by the bonuses to each troop type for those tactics, and add them up to give the average bonus to each troop type in each retinue. I've done the two extreme cases of no tactic and best tactic, but what really matters is the average.
I think since we convinced ourselves that morale scales linearly with kills and kills linearly with damage, we can just do the division. That said, the 36K cap test in your new test mod would be far more accurate and serve as confirmation.
What would be most helpful to my work, if there are any others still reading this thread, is if someone would compute the odds of each tactic for each pure retinue, multiply that by the bonuses to each troop type for those tactics, and add them up to give the average bonus to each troop type in each retinue. I've done the two extreme cases of no tactic and best tactic, but what really matters is the average.
Last edited:
You also have the paper rock scissors thing that some units are better against others.
I.e. if you have pikemen retinue and your opponent heavy cavalry - bad for him. If you have pikes and he heavy infantry, watch out.
I.e. if you have pikemen retinue and your opponent heavy cavalry - bad for him. If you have pikes and he heavy infantry, watch out.
That's mostly handled thru tactics and tactical affinities, and largely why I chose my generic punching bags as opponents to keep the math doable. Dragatus's testing mod will allow results against a high end levy army and the tactics it rolls, which is the real test.
Looking forward to this mod you are developing and I appreciate all the crunching you guys are doing on the numbers. Having mastered most of the game retinue is one of the more advanced things (IMO) to tackle and I've been looking into the wikipedia pages a lot more recently and planning retinue more carefully. I'm not sure I can be of any use to you guys as I'm not really familiar with the formulas yet. I've been balancing Retinue Plans for games around tactics in excel to make sure I have the right %'ages of each Retinue for optimal tactics, etc. Good fun. Anyways, keep up the good work, cheers. 
Retinue test mod is done: https://www.dropbox.com/s/tajtcbywoej8hz8/retinue_test_mod.zip?dl=0
Installation:
Unzip contents into your mod folder and enable in loader. Standard path on windows is: "My Documents/Paradox Interactive/Crusader Kings II/mod"
Features:
- Retinue cap greatly increased.
- Human players can recruit all cultural retinues.
- All retinues are free (cost no gold, have no upkeep) and will reinforce to maximum size in a few days.
- Human players can build any cultural building for low cost.
- AI can't build retinues or cultural buildings.
- Clear cultural buildings by disabling and then re-enabling cultural buildings by decision (in intrigue menu or by right-clicking your own portrait). This works by adding/removing Genius trait, when cultural buildings require characters to not have Genius in order to be built. It's messy but it seems to work (may not work if you only built 1st level of cultural building).
- Decision in intrigue menu to create levy_mix_1978. This creates event troops that perfectly matches the composition of a castle levy with all tech level 4 buildings (except cultural building). If the troops were retinues, the total retinue cap cost of the unit would be 1978.
- Decision in intrigue menu to create levy_mix_2000. Same as above, except a few extra troops added to round up retinue cap cost to 2000.
- Decision in intrigue menu to create levy_mix_36K. Same as levy_mix_2000, except 18x as many troops.
- Decision in intrigue menu to create levy_mix_12K. You get the idea.
- Decisions in intrigue menu to create commanders with 8, 12, or 16 martial rating.
- NEW (25/2/2016): Decision in intrigue menu to change your culture to one that grants special combat tactics.
Installation:
Unzip contents into your mod folder and enable in loader. Standard path on windows is: "My Documents/Paradox Interactive/Crusader Kings II/mod"
Features:
- Retinue cap greatly increased.
- Human players can recruit all cultural retinues.
- All retinues are free (cost no gold, have no upkeep) and will reinforce to maximum size in a few days.
- Human players can build any cultural building for low cost.
- AI can't build retinues or cultural buildings.
- Clear cultural buildings by disabling and then re-enabling cultural buildings by decision (in intrigue menu or by right-clicking your own portrait). This works by adding/removing Genius trait, when cultural buildings require characters to not have Genius in order to be built. It's messy but it seems to work (may not work if you only built 1st level of cultural building).
- Decision in intrigue menu to create levy_mix_1978. This creates event troops that perfectly matches the composition of a castle levy with all tech level 4 buildings (except cultural building). If the troops were retinues, the total retinue cap cost of the unit would be 1978.
- Decision in intrigue menu to create levy_mix_2000. Same as above, except a few extra troops added to round up retinue cap cost to 2000.
- Decision in intrigue menu to create levy_mix_36K. Same as levy_mix_2000, except 18x as many troops.
- Decision in intrigue menu to create levy_mix_12K. You get the idea.
- Decisions in intrigue menu to create commanders with 8, 12, or 16 martial rating.
- NEW (25/2/2016): Decision in intrigue menu to change your culture to one that grants special combat tactics.
Last edited:
Also, the compressed files are just standard ZIP files. You can open them with any common ZIP opener, altho you may need to add the .zip extension to the file name.
Unfortunately nobody ever took me up on the offer, so I haven't done any practical tests with my retinue testing mod and can't give a definitve answer to the question.
- 1
I updated the retinue test mod: https://www.dropbox.com/s/tajtcbywoej8hz8/retinue_test_mod.zip?dl=0
New version has a decision added in the intrigue menu that can change your culture to one of those that grant special combat tactics (English, Scottish, Italian, Norse, French, Mongol, Greek) or one that has no effect on combat tactics (Irish).
New version has a decision added in the intrigue menu that can change your culture to one of those that grant special combat tactics (English, Scottish, Italian, Norse, French, Mongol, Greek) or one that has no effect on combat tactics (Irish).
- 1
I decided to change my approach to retinue testing. Did some tests yesterday, courtesy of Dr Gonzo, and it turns out that setting up a multiplayer game and the armies takes more time than the actual battles themselves. So I decided I'll set up a savegame with an impending battle for each retinue to be tested and upload them. I'll also set up a google docs spreadsheet for entering results. And then anyone who wants to help can download the saves, load the save for whatever retinue they want to test, fight the battle, enter results in the spreadsheet, and then re-load and re-fight and re-enter resuts as many times as they wish. And you can do at your leisure instead of having to set up a MP game with all the extra time consuming crap that comes along with that.
As far as yesterdays' testing is concerned, the main thing for me was that we found several areas in which I can improve the test mod. But in addition to that we did get to test out 4 retinue compositions, each amounting to 36K retinue cap worth of troops (or as close as we could get): 60x Defence, 30x Knight, 45x Druzhina + 3x Shock (to trigger Shieldwall), and 40x Schiltron + 6x Defence (again so that we can trigger Shieldwall). They fought against 36K retinue cap worth of standard levy mix and everyone was supposed to have martial 13 commanders, but they kept picking up traits in battle so I can't guarantee they were martial 13 for every battle. I also forgot to swap in French commanders for the Knights, but I did remember to give Scottish commanders to the Schiltrons. All cultural retinues benefited from their respective level 3 cultural building. We recorded results for 5 battles with each retinue compositions and this are the results:
Defence composition: 91% survival rate and 29% kill rate
Knight composition: 75% survival rate and 30% kill rate
Druzhina composition: 92% survival rate 34% kill rate
Schiltron composition: 94% survival rate and 45% kill rate
Survival rate is the average percentage of troops that survived the battle. Kill rate is the average percentage of enemy troops killed in battle. Do note that with only 5 battles fought per composition the results are far from definitive. And the knights had Irish commanders instead of French ones which did reduce their performance.
EDIT: While I'm at it, let me know if there are any specific retinue compositions you want to see tested. After my test kit is ready I'll do some testing of my own, but I won't be able to thoroughly test every single retinue composition because it's simply too time consuming. So after I get the ones I'm most interested in, I'd like to go for the ones that others want tested.
EDIT #2: I'd also like to start a discussion about what the testing conditions. Specifically the martial skill of commanders and the size of the simulated retinue that would make the test most realistic. I'm currently leaning towards 12 Martial for all commanders and having 72K worth of levies vs 36K worth of retinues. And if you can think of anything else I should take into account, I'd like to hear it too.
As far as yesterdays' testing is concerned, the main thing for me was that we found several areas in which I can improve the test mod. But in addition to that we did get to test out 4 retinue compositions, each amounting to 36K retinue cap worth of troops (or as close as we could get): 60x Defence, 30x Knight, 45x Druzhina + 3x Shock (to trigger Shieldwall), and 40x Schiltron + 6x Defence (again so that we can trigger Shieldwall). They fought against 36K retinue cap worth of standard levy mix and everyone was supposed to have martial 13 commanders, but they kept picking up traits in battle so I can't guarantee they were martial 13 for every battle. I also forgot to swap in French commanders for the Knights, but I did remember to give Scottish commanders to the Schiltrons. All cultural retinues benefited from their respective level 3 cultural building. We recorded results for 5 battles with each retinue compositions and this are the results:
Defence composition: 91% survival rate and 29% kill rate
Knight composition: 75% survival rate and 30% kill rate
Druzhina composition: 92% survival rate 34% kill rate
Schiltron composition: 94% survival rate and 45% kill rate
Survival rate is the average percentage of troops that survived the battle. Kill rate is the average percentage of enemy troops killed in battle. Do note that with only 5 battles fought per composition the results are far from definitive. And the knights had Irish commanders instead of French ones which did reduce their performance.
EDIT: While I'm at it, let me know if there are any specific retinue compositions you want to see tested. After my test kit is ready I'll do some testing of my own, but I won't be able to thoroughly test every single retinue composition because it's simply too time consuming. So after I get the ones I'm most interested in, I'd like to go for the ones that others want tested.
EDIT #2: I'd also like to start a discussion about what the testing conditions. Specifically the martial skill of commanders and the size of the simulated retinue that would make the test most realistic. I'm currently leaning towards 12 Martial for all commanders and having 72K worth of levies vs 36K worth of retinues. And if you can think of anything else I should take into account, I'd like to hear it too.
Last edited:
- 1
Interesting. It seemed to me always that the impact force of knights can destroy enemies than make major destruction of knights.
If the knight retinue had French commanders they could've gotten the Couched Lance Chare tactic, which makes knights some 15-25% stronger than Awesome Charge. Taking into account that knights only represent 60% of a knight retinue and that they can still trigger Awesome Charge, I'm guessing both the survival rate and the kill rate could've been some 3-5% higher if I had done it properly and given them French commanders. The knights also only benefited from a level 3 cultural building, so their attack bonus could be further upgraded too.
Also note that the Knight retinue numbered 7500 soldiers when the equivalent Defence retinue numbered 18000 soldiers. So using Knight retinues allows you to pack more punch into the same supply limit.
On the other hand, having a better commander (16+ Martial) would've been a bigger imrpovement for the Defence retinue as it would've improved their odds of using Shieldwall in skirmish phase from around 50% to around 70%. And while their larger numbers can cause supply problems in the late game, they also make them better at sieging.
Also note that the Knight retinue numbered 7500 soldiers when the equivalent Defence retinue numbered 18000 soldiers. So using Knight retinues allows you to pack more punch into the same supply limit.
On the other hand, having a better commander (16+ Martial) would've been a bigger imrpovement for the Defence retinue as it would've improved their odds of using Shieldwall in skirmish phase from around 50% to around 70%. And while their larger numbers can cause supply problems in the late game, they also make them better at sieging.
Hello.
I am doing a game as Finnish and I was wandering whether the Baltic warriors are worth their cost, and if they are, which retinue composition I should aim at to maximize their efficiency. Not that for now I don't have the money to invest in them (but I will when I'll convert to merchant republic).
Cost effectiveness and overall effectiveness:
They seem clearly weaker than Druzhina; they got 50 more HI but are more expensive in gold and retinue cap, with weaker stats bonuses.
Are they still good compared to defense and shock retinues?
Composition (for around 1000 soldiers):
- I was thinking about 2*Baltic warriors and 2*skirmish (to get at least some damage in the skirmish phase).
overall, aiming for (3/5 to 2/3 of Baltic warriors)
With this kind of composition, am I missing some important requirement for unocking good tactics? Is there a way to build a better retinue than that?
- As a Finnish ruler, are there some compositions including only basic retinues which are more interesting than compositions built around Baltic warriors such as the one above?
I am doing a game as Finnish and I was wandering whether the Baltic warriors are worth their cost, and if they are, which retinue composition I should aim at to maximize their efficiency. Not that for now I don't have the money to invest in them (but I will when I'll convert to merchant republic).
Cost effectiveness and overall effectiveness:
They seem clearly weaker than Druzhina; they got 50 more HI but are more expensive in gold and retinue cap, with weaker stats bonuses.
Are they still good compared to defense and shock retinues?
Composition (for around 1000 soldiers):
- I was thinking about 2*Baltic warriors and 2*skirmish (to get at least some damage in the skirmish phase).
overall, aiming for (3/5 to 2/3 of Baltic warriors)
With this kind of composition, am I missing some important requirement for unocking good tactics? Is there a way to build a better retinue than that?
- As a Finnish ruler, are there some compositions including only basic retinues which are more interesting than compositions built around Baltic warriors such as the one above?