Shogun 2 Realism + | Main Thread: Feedback & Discussion

[GShock]

You have some good thoughts and ideas. Some are not easy to achieve ;)

Undoubtedly.
I was an official beta from STW 1.0 to MI final... I've read something like 10 books on japanese history and warfare. I know what I am talking about. :)

As for the exp theme i indeed would prefer a file that can limit the exp gaining in the whole, so units can ie. get 5 or 6 points maximum. I'll search for a file. Otherwise if this is not possible, i think about it to limit the exp via buildings by 1 point for all building upgrades and technology. Also i might like to reduce the exp gaining via battle.

You should tune up the number of kills required to achieve +1xp, increase the threshold at which units are autodisbanded after battle because of their losses (66% losses is a good figure imo) and work on a chain of buildings where the maximum starting xp level of a unit is a mere 2. The losses and slow replenishment will make sure high xp is very hard to achieve.
If you consider there's a cap in unit numbers, you can't really make do WITHOUT a depleted unit for long so you will undoubtedly risk them and, eventually, lose them.

The difficulty in reaching high xp for a unit is because when a crack unit loses men it's hard to replenish it with men of the same skill coming from other units. The CTL-M is mandatory for the player in your excellent mod but it doesn't take into account the XP of the units you are merging (so it mixes green with veterans). This however, still ensures at least one of the depleted units remains depleted and it, eventually will just die out sooner than the elite unit.

Be advised in your system the cavalry being so buffed has an easy time, especially because of easy routing troops, in gaining high xp. While battles last longer, I think the losses are extremely high because units are resilient in combat and stay on the field without routing for far too long. That's ALSO extra unneeded kills. Troops should rout sooner (or general call the retreat) and live to fight another day. Routing speed must be increased, that's a must so they can have a better chance to escape.

Well, 15 years is after all 60 turns, a lot happens there on such a small designed Japan map. I also would prefer a forth and back war of regions lost and won. That's a very hard part for a balancing job, you know.

That happens because the AI doesn't really garrison its castles so it's relatively easy to conquer many many provinces with just one army even when it's not replenishing. Possible solutions are 1 extra garrison unit (but do reduce the numbers of the garrison bowmen because they are more numerous than field units), further reduction of land movement (slower movement means more time to defend but I don't really like this compromised alternative) or... just find a way to make the AI protect its borders and make more use of alliances. As shimazu I'm not going to attack Sagara if it's backed up by a strong clan like Mori that would destroy my trade routes with its navy.
Perhaps a good compromised may be achieved if the happiness level of a newly conquered region are low. This would force the AI to garrison new conquests much better which in turn makes it harder to lose the province. Shoni for example doesn't garrison Tsukushima, Hizen and Buzen at all. I actually almost never catch Sagara from Bungo... it's 10 times easier to conquer these 3 as they are undefended. If these problems are solved it will take much much longer than 15 yrs to have 5 or 6 clans on the map. ;)

Now as i gave the major clans Oda, Shimazu (and also Mori) extraordinary support to keep them in the game, besides that other major clans have some extras as well, it is unlikely that minor factions can survive for longer than 15 years, i think so, even if diplomacy and CAI modding is done.

Diplomacy is key here. minor clans should do the impossible to align to their more powerful and troublesome neighbors. The penalties for attacking an ally are extreme. This doesn't happen unfortunately. More vassals and more allies = longer campaigning.

What i did also via CAI modding is, that minor factions are more passive now ... more on the defense side of AI behaviour, while the playable major factions remain active but also i gave them more caution for defense. The non-playable "minor" factions are divided in two branches, the real "peaceful" factions and the in-between factions, the latter ones are still between passive and active. Actually it was my goal to get the playable major factions out of the looser-street, i believe i achieved that now.

The situation with Oda/Tokugawa is probably the hardest one. However what we know now is the history that those winning factions did and they did it on the field of battle. A minor faction in this problematic period was impossible to be kept alive without being allied or vassalized by a stronger neighbor for their own interest. You never know when you can take your revenge against your lord or ally later on so it makes sense.

I think as majors prioritize defense so you can make minors prioritize defence through alliances. As I mentioned, I am not going to attack Sagara even if I am 3 times stronger if it's allied to Mori. Sagara instead plays so peacefully the BEST I've seen it do is take Tsukushima (only to lose it a couple of rounds later). It's too static, impossible to survive like that and even a peaceful daimyo must have known there's no room for peace in the age of the country at war. Can you do something about minors allying? To the extent of being vassalized is not a bad idea.

As i'm a history addict to some degree, for me it is actually very important to have historical major clans in the game, which get the upperhand, in the best case it is an Oda-led "east alliance" vs. a Shimazu-led "west-alliance". But i also would not like it, if Oda and Shimazu can steamroll the map too easy, too quick ... 15 years with about 5-6 clans, well, at this year it should become slowly hard to have enough units for a faction to control the regions, as i limited the units quite strict, except for Yari Ashigaru, but also they will be empty some day. This is done by intention, so at one point in the game the remaining factions have issues to keep their regions properly controlled - i'm interested to see playtests, when this becomes relevant, maybe i shall limit Yari Ashigaru more?

With a static samurai cap number that's correct, you can basically field 3 maybe 4 armies but of course the more you have the less powerful they are so the bigger the reign the harder it is to properly defend it but there are seldom problems once you conquer a region. As I said in the previous feedback, Ashigaru are hands stripped off the economy. You don't need limiters here if the upkeep simulates the fact that men can't work the land because they are in the army, having many Ashigaru will cost you a fortune (and force you to loot more often or disband units). It's the ashigaru upkeep that must rise.

Also i might like to include some tweaks which can decrease the repression rates (what i have done already here and there, maybe i need to increase late unrest factors), and rebellions occur eventually in the later game?

Aggressive AI doesn't defend castles and often insurrections arise but according to me this parameter should rise a bit to contain fast expansions from the beginning to the end not just in the late game. The minors are so inactive that when a region revolts they don't even try to recapture it... and that's an easy one. I hope that changes... it takes but 2 regions to turn a minor into a major clan after all.

However, i'll tweak some codes, ie. i think i can nerf Oda and Shimazu and also Ikko Ikki unit abilities a bit now, have to look then that they still play well in the campaign.
Also i'll try to increase the help ally percentage, in the hope factions won't let their main settlement(s) alone to help an ally. The care for protecting settlements and regions is already tweaked a lot, more is hardly possible because it would go on the cost of army activity (attacking other regions) and i guess the AI behaviour is also somehow hardcoded in this regard, ie. i saw Shimazu very often letting their home capital unprotected (while facing Ito as neighbor with quite some units) although they should have enough money to recruit lots of units and keeping them in the castle, they didn't and lost it ... recently, i just found a workaround which ensures that Shimazu (and Oda) expand and can't loose their starting region in the 1st turns vs. AI enemies ... lol... i have beaten the damn S2 engine, or in other words the odd CA start-situation balance.

Diplomacy again is the key. You go vs Shimazu and you'll be at war with its allies. One region and even one clan eliminated as a threat don't ensure the final survival in the war. :)

All right, keep up. I'll be waiting for your next versions. :)

[DaVinci]

Now, v1.96.5 uploaded ("1.96.5", because i'll need more update steps than the predicted remaining 5 from 1.95 to 2.0).
This is merely an internal .org version, and i hope i'll get again campaign playtest feedback from you, GShock :) (maybe also other step by).

I tested one in-between version but not the very last one and won't have time over the week.

What i need is a deep play into the campaign, and this with Chosokabe on H/H with ultra unit size, long campaign, and the unit speed reduction mod should be applied, no other gameplay mods atm., observing the AI, esp. Shimazu, Mori and Oda, ie. if they die soon again, then you can quit and report, because this would be a no go for me (and i have to adjust it back to higher stability for them, while Mori is a quite "whacky"-faction, due to their "hardcoded" naval focus, and it might be inevitable that they suffer from Amako or another one, and have to leave the theater, because they didn't focus enough on land).


GShock, there are of course a lot things in all your comments which are known by me (i'm balancing tw games since 2005), and a lot things could be replied by me, regarding aspects which are relevant (that would lead to an extreme detail discussion which i can't afford timewise) - but really, i wanna thank you for your contribution here, it's very very appreciated, you put some light on things which i wouldn't have had considered in such a depth without your focus on them, maybe i even would have ignored them esp. your expressed desire to put a lot focus on alliance-diplomacy-behaviour, i agree with you now, you convinced me, it's a major factor (for this theme i'll relate at least 2 update steps, it's not in this 1.96.5, here only slightly strived, and i hope i can do something in this regard, which significantly changes the pace of the AI behaviour, while the diplo-behaviour in all TW games is a very critical one, and merely impossible to achieve a proper change of the campaign course, which we would like to see) - i'll mention you soon in the credits section, as undoubtly your input influences my modding.

[GShock]

Since you aim for realism and I join realism-oriented projects in all games I have given my contribution to, I am here to help you.

Beginning tests on 1.96.5 immediately.

I recommend you download and study the AI files of the Darth Mod because he's totally focused atm on forcing the AI inside castles when they can't win by sallying out and by NOT storming structures when they can't win the offensive siege due to low numbers. I am sure his files will help.

Again, if the AI can be stopped from playing stupid (which is EXCLUSIVELY attacking when weak) this game will dramatically change for the better. AI needs to properly garrison towns because they are mostly left undefended. I do not know whether it's best to get there by increasing the riot factors (which would force a stronger garrison) or the garrisons themselves (which however would risk leading the offensive AI to crash and die even sooner with wrong siege attempts with weak numbers) or, reducing the movement speeds (more time taken in travels means more time to reinforce for the defender). That's up to you and what you can do. I am a tester, not a modder. :)

TTYL on tests but I can't afford ultra-ultra. I'll play hard Shimazu on medium sized.

[GShock]

Testing on the AI behavior is rather easy. All you need is to play Shimazu and see how clans Ito and Sagara react to your moves.

The 1.96.5 showed there's no change in the attitude of Ito. You can conquer Osumi at round 1, 2 or 3 and the fact Satsuma may recruit 3 ashigaru units in just one turn makes me think and I don't think it's a good idea. Yes, the number of recruiting slots is too big for just a very small castle and then of course Ito has no chance in defending Osumi in the long term.

In the short term, Osumi, just like 99% of all castles I have seen in S2TW (all mods or versions) is the same: totally ungarrisoned.
I observed a single unit popping out and going actually back towards Hyuga many many times where the Ito Daimyo is and he's coming with an army. That leaves the already weak Osumi undefended 100% of times. If you want you can take Osumi WITH that unit inside or not. The intervening garrison is too weak anyway.

What the Ito daimyo does is simply attack you if you are laying siege to Osumi (which would be correct if he had the numbers, unfortunately, due to the 3 ashi recruiting slots the AI DOESN'T have the numbers), otherwise it goes through the woods towards Satsuma. It is an interesting diversion but with an easy countermove. Once Osumi is captured, he doesn't have the movement points to attack Satsuma right away, so you can basically go between the swordsmith and the farms of Satsuma, eventually cutting out of taxes Osumi (to keep it from rebelling) and the Ito Daimyo will either lose or fall back.

this move can be repeated many times until you are strong enough to defend Satsuma/Osumi and while the Ito Daimyo is playing club and carrot with you, you can take Hyuga... which ends the Ito days for good.

On one side I am against the spawning of garrisons at all. This way you can take Osumi but with a high riot factor (fixing the recruiting slots @ Satsuma) you will take losses and be unable to defend Osumi from the upcoming Ito counterstrike (which you survive just because of the garrison which should NOT be there since you just conquered an enemy region). On the other side, if the AI doesn't garrison its castles, making the garrisons stronger (3 times stronger but the units must field less men than their corresponding field unit, i.e. bow ashi is 70 in garrison but the bow ashi unit is 60 men, so the garrison should be 30 men and the field unit 60 men but with this idea, the garrison should have 1 retainer, 2 ashi and 2 bow ashi).

If my idea works, and the stronger garrisons (the bigger the castle, the more numerous the garrison units) then the AI may move relatively freely. Perhaps in order to slow down steamrolling, optionizing that the repairing of a castle takes 2 turns instead of 1 is a better idea (s you see I am listing the options). If the caslte is not repaired it does not REPRESS. Hence if you don't conquer the region with appropriate forces it will most certainly riot.

At this point, give the rebel armies STRONG UNITS and not only ashigaru (i.e. MANY ronin in their army) and you'll see how things change.

Summary:

Garrison units should be more numerous but with HALF the numbers in each unit for the corresponding field unit (bow ashi garrison is 70 should be 30 when field unit bow ashi is 60 man). An increment in unit numbers means lev1 structure gives 1 retainer and TRIPLE units (3 bow ashi, 1 retainer, 3 ashi infantry).

Castle repair (this idea should be extended to all dmgd structures because it's too cheap to repair them right now) taking more turns.

Stronger rebellions.

Anyway at present campaign I adopted the same strategy as always: osumi, hyuga, bungo, buzen, tsukushima, hizen, bingo. This time though, MORI declared war on me and everything changed (though I managed to bring in the war the Chosokabe). The only missing thing is clan Sagara which saw me conquer the whole Kyushu without making a single move against me (not tried to ally, not declared war, did not shift from friendly to unfriendly... did not attack me in the back. This all made things easy for me, it always does).

Perhaps a stronger decline of attitude when a daimyo expands could help sway the tide of attitudes. This is ALSO an option to slow down steamrolling.

I suggest you focus on the strategic side of the problem with the AI, it's most certainly the greatest weakness of ALL TW games.

[DaVinci]

Damn, i wrote a longer reply, it's lost somehow.

Summary, steps for the next version (test):
- Upped garrison along your model (with decreased garrison unit number).
- Minor and full minor AI factions more militant again.

Prior to this, i need a 1.96.5 test as Chosokabe, if Shimazu is AI (the course of Kyushu is different if Shimazu is AI steered, believe me).

Still a short hint: All factions have one extra recruitment slot in their home-region-capital.
Do you think it is senseful to reduce the default slots per castle tier by 1?

[GShock]

I am starting a Chosokabe campaign.

About castles my idea is
lev1/lev2 -> 1 unit
lev3/lev4 -> 2 units

We need more units in garrisons to match this criteria so that the bigger the castle the more units in the garrison. Again, is it possible to increase the cost (if not the time) to repair dmg structures? A dmg castle still gives the garrison... :(

Yesterday I got jammed in Hyuga and attacked by Shoni. I had the time to produce the naginata samurai. Somehow at their first battle they already had XP2. That's not good. Meanwhile Shoni had 2 bow, 2 muskets and the rest were all yashi with 2 generals. Possible solution, unless you want to change the system with one building type producing one unit type (as opposed to one recruiting center able to produce all unit types), make sure the AI builds one military building before building 1 market, otherwise obviously it lacks the structures to produce samurai units.
Some sort of building limitation might be useful: I.e. you can build the sword dojo in Satsuma but not in Hizen. This means in Hizen I'll build a bow dojo, hence armies will be more balanced. That's for the AI of course, we already build different dojos in different places to have balanced armies but I still see mostly yashi armies for the AI.

[GShock]

All right, you're on spot, Shimazu doesn't survive when you play as Chosokabe.
I've got to say a few things right now having seen the recruitment tree of this clan but first let's look at how Shimazu loses.

This time instead of going to Osumi, Shimazu actually conquered Higo and destroyed Sagara. A few turns later, Higo rioted. Clan Ouchi and Shoni now have Kyushu split exactly along the middle line, blue on left, grey on right. I was surprised, I thought the AI had the mission to conquer Osumi but it didn't and this is very bad because I had in my mind to use these missions to influence the AI choice-making process so we could keep things perfectly smooth at start.

My personal point of view is of total "what if" with this game, regardless of the S2R+ mod. So I totally disagree on giving each faction a separate tech tree and unit advantages or disadvantages. It's pretty simple. Oda wrote history with the ashigaru troops but this history was written on the field of battle, the way they were employed, trained, recruited, not by pre-existent bonuses. The same goes with the Takeda cavalry... it became famous because of how it was used not because of the better horses of the Shinano province. ;)
The same goes with Shimazu. This clan starts with one province, Satsuma, so it can't be said to be a major clan. It's HISTORY that made him major the way things went in battle for Shimazu but that's in the "future" not in the "starting points" of the clan. So all clans should be treated the same, at least for testing purposes and then we may move a step ahead once things work correctly in adding flavor and different daimyo personalities.
The fact Chosokabe can't recruit yashi but only bashi is a minus. All clans should have the same choices at hand bc I don't think the AI actually is able to take the right choice for each clan. It will take the same choice, always, and so with some clan it will pay with some other, it won't.

So, I suggest we start working from scratch, giving all clans the same unit tech tree, unit costs, unit limiters, etc (and testing at normal difficulty level without AI bonuses or handicaps). This will help balance things and reinstate your additions (and others) one by one on a SOLID base.

Let's say that the problem with the AI is that for example, I conquered Iyo and stayed in the castle, I was stormed at first sight by a 300 man army while inside I had 1200 men including the garrison. This attitude is suicidal that's why Shimazu loses. the AI doesn't seem to keep anything safe including the capital but if you look at Sagara, it only attacks Tsukushima (1.96.5 as Shimazu) when Higo has 20 units inside. In 2 turns, that army is back in Higo, Tsukushima is undefended again and obviously lost to Shoni or Ouchi.

It might be that reducing the repression factors of castles might help keep the AI inside with a garrison, otherwise it's a matter of daimyo's attitude/personality. Peaceful clans act like Sagara, they stay put in great numbers. Others go out and move as far as they can even after conquering a province without giving ANY garrison to the newly conquered garrisons. In other words the AI will keep moving regardless of how safe things are left behind, this is the reason why ALL clans die and no effort on rebalancing can help. It takes but a catastrophe with your single army the stupid way the AI attacks and since there's nobody at home, the AI faction is defenseless.

I m thinking right now that we could have multiple tier structures (i.e. bow dojo, bow dojo lev 2, bow dojo lev 3, bow dojo lev 4 and so on, according to the castle level) and each structure produce still just 1 recruiting slot but a different unit.
Say, lev1 castle produces 30 man yashi unit, the sword dojo lev 1 can produce 30 man sword ashi unit while lev 2 castle can produce 50 man yashi unit and allows sword dojo lev 2 which can produce 50 man sword ashi (and 0xp 30 man no-dachi). Lev 3 castle can produce 60 man yashi unit, allowing sword dojo lev 3 which can produce 60 man sword ashi and 0xp 60 man no-dachi. Of course all these units must have a max number capping. In the long run and especially if you can find the parameter to disband a unit if 66% of men are lost, with the passing time, you have stronger units without messing around with the xp... it's the evolution of warfare, structures able to recruit and train more men but of course, only combat gives the experience.
I suggest to disable bushido art as requirement for yashi (that's supposedly the main bulk unit of the game for all clans) and allow the special ashi type at spear art (so we can drop that 1xp bonus), the second spear/bow/sword is ok with +1xp. As I said, 1XP is the MAXIMUM xp level we should allow for a starting unit, which is already modified by the command, traits and retainers of its leader. The inability to build ANY structure in ANY place should ALSO be a very good factor as said above: we can't let AI armies be built unbalanced and if the AI can't build a dojo here it will build some other dojo and that helps. Finally it is ****MANDATORY**** all provinces can recruit RONIN.

Perhaps the solution to the AI being so reckless is a reduction of its movement points. If it were possible to also increase the "red circle" of influence of structures, ships and armies, the AI would be compelled, geographically to take different choices because it can't move further in that same turn. Essentially, this should help. In this case I'd make armies move 1 (just a bit less than they now do), agents 2 and ships 3 (bigger ship movement helps with naval invasion option).

Again, I strongly urge you to at least give a look at the Darth Mod files.

[GShock]

Mmm some more thoughts: It's likely that the production of units is affected by the insane choice of the AI to build so many markets instead of multiple military buildings. If that is so, maybe it's a good idea to make the lev 1 castle only be able to build 1 military building, hence all other civil buildings be available at lev2 castle where we already have military capability in that region.

Again, lev1 castle with any building should be able to produce ronin. Heaviest upkeep should go on ronin, lowest on samurai. Of course samurai are limited both in number of max units and in manpower in each unit. I like the idea of having more men produced with higher tier buildings instead of more xp, it's much better but this requires multiple units to be created though it should be a piece of cake, easy to do.

Maybe the best solution is to have castle lev1 produce yashi and ybow (+ronin), lev2 castle yashi (more men) and ybow (more men) (+ ronin more men) + nagi ashi and sword ashi. Lev3 castle would allow a higher number of all ashi units + higher tier of military building (i.e. if u have yari dojo built at castle lev 2 you could recruit ysam and with lev3 castle you can build nag dojo which gives more men to ysam + nagi sam, then with an extra yari dojo you have more men at yari sam and more men at nagi sam which should still be fewer than yari sam in any case all at lev0 xp).

Just extra thoughts.

[DaVinci]

Huh, lots of valueable stuff what you provide again, awesome, really i'm impressed by your descriptive capability. The problem now: You would like to see a whole new mod from scratch with total what-if base? I'm a historical-realism-man, means, the total war (vanilla) what-if approach is not my cup of tea, even if realism contents/aspects are modded into a game, i can get only immersion and a flavour feeling if historically successful factions are no weak ducks and make a good solid play. [I'm modding TW since Rome, there i participated in the old TFT mod (today known under FRRE mods) and realised parallel my own proper Roman faction mod, based on a lot of historical stuff/background, this modding style went on with several other mods on RTW (for example ChivalryTW) and then M2TW, then ETW and NTW].

Actually i'm working since a huge set of S2R+ versions on the key-factions (if AI steered) Oda*, Shimazu (and recently also Mori), getting support to reflect their historical path to some degree, in the sense that they don't get wiped out easily or at all by "minor" factions AI. Hope you understand that your start-anew suggestions are a no-go for me. I see you have a very good conceptual approach for modding TW games, like a developer, maybe you can find someone who likes to modify what you provide as conception, then you have your own mod as project director.

* as compromise (i know in reality Tokugawa went with Oda to success)

Edit: Maybe it sounds harsh, it is not meant that way, just only clear words for a clarification of the S2R+ mod-design ;)

Also i'm working on a v1.97 which shows very well results in every respect, have playtested (always as Chosokabe) current dev-version until 1556 or so, and everything looks good, except that Shimazu got overwhelmed recently by Shoni due to their odd behaviour to attack Sagara which has a whole full stack in their castle (as you have described above) so Shimazu lost their veteran army and Shoni stepped in to take Sagara and bit later Simazu- i'll try to "correct" that as far it is possible. Else the map still looks good, many factions alive and kicking, including that some rebellions occur and change regions-course for a while, from the major factions, only Tokugawa and Imagawa are erased (by Oda of course), and then later Uesegi by Takeda, which got parallel big issues with Ikko Ikki, Ikko Ikki also attacked recently Oda, who fought them back (1 lost region at least), however now Oda is at war with Takeda, Hojo and Ikko Ikki, what stops him significantly. Many, really many "minor" factions are alive and make a good play, campaign-wise it's a good step further now, along my campaign-balance-goals.
And I have lots of ideas (got them while playtesting) in my mind which i wanna put into the upcoming release version to complement the course of the campaign-balance-style, and many of your thoughts went into that design recently, i would be happy if you still wanna playtest and suggest to the S2R+ approach, your contribution is highly appreciated, although S2R+ is never designed to support total what-if, i would feel honored if you stay aside (the what-if happens anyway if a player chooses a certain faction, and plays just as he wants, ie. complete ahistorical, ie. destroys Oda as soon as he can - but i'm making my mods for myself in the first place for my personal playstyle, for the kind of immersion and flavour which i need).

[GShock]

Huh, lots of valueable stuff what you provide again, awesome, really i'm impressed by your descriptive capability. The problem now: You would like to see a whole new mod from scratch with total what-if base? I'm a historical-realism-man, means, the total war (vanilla) what-if approach is not my cup of tea, even if realism contents/aspects are modded into a game, i can get only immersion and a flavour feeling if historically successful factions are no weak ducks and make a good solid play.

Believe it or not, you have no choice, my friend. One thing is to balance (melee 4 + armor 5) vs (melee 4 + armor 5) and a totally different thing is to balance (melee 4 + armor 5 + x + y + z) vs (melee 4 + armor 5 + a + b + c). Start from the man: a soldier dies if he can't help it.
The AI will take the most logical choice according to its logics and I don't really think it takes into account 50 variables but just a few. The more stuff you add to differentiate one faction from the other, the hardest (beyond impossible) it becomes to figure out how the mechanics really work. We have a fine example with the daimyos attitudes. You might spend a few decades to balance the mere island of kyushu with just 5 clans.
What made the Shimazu ksamurai good was the way they were used (successfully) not a starting advantage and there's no guarantee the AI will favor ksamurai and actually employ them correctly in battle (it doesn't in fact). I am on the what if because I get Shimazu and make its daimyo become Shogun. If he can become shogun, then the same can be said for any other clan. As I said, Shimazu starts with just ONE province and you can't be less minor than that! :)

Actually i'm working since a huge set of S2R+ versions on the key-factions (if AI steered) Oda*, Shimazu (and recently also Mori), getting support to reflect their historical path to some degree, in the sense that they don't get wiped out easily or at all by "minor" factions AI. Hope you understand that your start-anew suggestions are a no-go for me. I see you have a very good conceptual approach for modding TW games, like a developer, maybe you can find someone who likes to modify what you provide as conception, then you have your own mod as project director.

Mmm no not really, not again please. :)
Well I was merely suggesting the "scratch" approach for the sake of simplicity. It's simple to balance (a + b) vs (a + b) than balance 20 different variables but I understand your approach and respect it, I will help you all the same in your quest. :)
The problem of your approach is that if a faction always behaves the same, there's really little room for "surprises" the same strategies will fail and the same strategies will succeed. While this might be good for the AI (minor clan will always be subdued) that might not be so much fun for the player but hey, after all you might lose a key battle and all the strategy goes to hell so ... you never can tell.

Also i'm working on a v1.97 which shows very well results in every respect, have playtested (always as Chosokabe) current dev-version until 1556 or so, and everything looks good, except that Shimazu got overwhelmed recently by Shoni due to their odd behaviour to attack Sagara which has a whole full stack in their castle (as you have described above) so Shimazu lost their veteran army and Shoni stepped in to take Sagara and bit later Simazu- i'll try to "correct" that as far it is possible.

I don't know how he did it but I played 5 mins DMS mod and he managed actually to prevent the AI from attacking without a credible chance of success. Note that DMS has a vastly humongous "zone control" so it's impossible for an army to bypass a structure or an enemy army. Also, the mov allowance is huge. Where ITO fails to reach Satsuma in SR2+ it CAN reach Satsuma in DMS while you are in Osumi. That changes everything for Shimazu.

Else the map still looks good, many factions alive and kicking, including that some rebellions occur and change regions-course for a while, from the major factions, only Tokugawa and Imagawa are erased (by Oda of course), and then later Uesegi by Takeda, which got parallel big issues with Ikko Ikki, Ikko Ikki also attacked recently Oda, who fought them back (1 lost region at least), however now Oda is at war with Takeda, Hojo and Ikko Ikki, what stops him significantly. Many, really many "minor" factions are alive and make a good play, campaign-wise it's a good step further now, along my campaign-balance-goals.

Did the alliance war support pay? Did any AI vassalize another AI minor?

And I have lots of ideas (got them while playtesting) in my mind which i wanna put into the upcoming release version to complement the course of the campaign-balance-style, and many of your thoughts went into that design recently, i would be happy if you still wanna playtest and suggest to the S2R+ approach, your contribution is highly appreciated, although S2R+ is never designed to support total what-if, i would feel honored if you stay aside (the what-if happens anyway if a player chooses a certain faction, and plays just as he wants, ie. complete ahistorical, ie. destroys Oda as soon as he can - but i'm making my mods for myself in the first place for my personal playstyle, for the kind of immersion and flavour which i need).

Of course. I'll help you in any way I can. :)

[DaVinci]

Of course. I'll help you in any way I can. :)

Thank you :)

I'm still in favour for the mod, also as player if i choose a faction like Oda or Shimazu (who are buffed), because i might be able to reach the victory conditions in time ;) ... a RealmDivide mod, or extended campaign victory mod etc. (workarounds for the win conditions via vanilla) might be superflouos then, seriously i had this in my mind all the way - as i started to play this game, it was clear the vanilla difficulty approach is near insane regarding the game victory, you would have to rush all the way, in principle recruit recruit recruit conquer conquer conquer every turn to reach the victory conditions, this is not my cup of coffee for a historical shaped game, because this i would understand as sports-festival where i don't get any immersion etc..
On the contrary side, if i choose not one of the buffed factions, i'll have a harder challenge than with vanilla vs. those factions, unless i manage it to throw them out initially. A challenge can be found always in this mod in the one or other way, here it is for example or especially also the campaign play, which is far more complex in S2R+ with increased realism backgrounds, this provides me fun, and i mod/play for that kind of fun.

---
As for some of your points regarding AI. I have actually other experiences, the TW AI is able to act strategically/tactically along their capabilities also if the mod is complex. The problem is the modding, a small turned screw into the wrong direction can have "fatal" impacts, therefor balancing a TW game can be a years-long job. S2 balancing is in principle quite easy as for the map and factions and unit structures etc.. A point to clear is the goal, the approach for the mod. I think i declared it, and ie. if Shimazu (the only key-faction which has still real issues to expand and hold it stable) is able to hold its stand and even expand, then i'll be happy, and the mod is finished, while one can of course adjust and add stuff etc. without end. Edit: ... maybe i give Shimazu finally another AI character ;)
Second, the "alliance" is misunderstood, i meant west- and east alliances led by Shimazu and Oda, this means not that they vassalize minor factions, but conquer them ;) ..how would you else be able to get the victory conditions? Also the AI's take this victory condition into account. Such an approach of collecting vassals would need a modding of the victory conditions, the reduction of win-region numbers in significant number, ie. halfed or much more reduced ... would be a nice mod as well with this design, but also much much more harder to gain, and it would need really weak minor factions who see absolutely no sense to fight vs. you, that's ... hmm ... not a good idea, ie. you come with your 20 unit stack and the threatened AI gives up instantly, because it sees it cannot win the siege-battle, or won't ever attack your region, because it sees your heavy garrisoned castle with your superior units...then again, the human player will always play smarter than the AI... and in such a design, the human player steers the whole campaign. I prefer then a slight stupid AI which defends and attacks even if they can't win, just fighting for their honor along the way of the bushido ;) ... this means not that i want a complete stupid AI, no way, i'm always working on AI support in my mods.
Last but not least, as you say the outcome is predictable, no, it is not. There is enough random factor in the game/mod, and it will always play different along the chosen faction by the human player, it is more different in my mod here, as in vanilla, i think, just due to the complexity of this mod.

[DaVinci]

V 1.97 uploaded. See changelog.

[GShock]

Noted, unfortunately I can't even see it before this week end. Extremely busy at work. :(

I liked the notes about buffind one faction or another because they are surrounded by enemies... it's the age of the country at war, they are ALL surrounded by enemies. :)
Damn that's why we need a strong diplomacy rework...
About the recruiting, did you consider the idea of the "troops in evolution" ? (start with poor stats, then new building gives bit better stats and increased numbers)
I also like the idea of lowering some stats. In S2R+ I think troops stand too long. Battles are overly bloody and there's never a mass rout. Most casualties are in actual combat.
What do you mean with experience threshold heavily upped? (Must be silly Qs but since I don't have time to test before the week end, I am too curious to wait)

Probably a major feature to test will be the "radar" and new movement speeds. Can't wait to see it. :)

[DaVinci]

Damn that's why we need a strong diplomacy rework...
About the recruiting, did you consider the idea of the "troops in evolution" ? (start with poor stats, then new building gives bit better stats and increased numbers)
I also like the idea of lowering some stats. In S2R+ I think troops stand too long. Battles are overly bloody and there's never a mass rout. Most casualties are in actual combat.
What do you mean with experience threshold heavily upped?

Strong diplo: Not sure, if this is really needed. Ie. if factions primarily don't fight each other but build alliances, to what will lead this? AI is dependent on the codes, human player not, such a design in the end will mere help the human player, not the AI.
Also, if a faction has more than 1-2 allies, it will go on the costs of other factions who have not the luck to get allies, in result i think there is no positive effect, other thant that it is more historical reflected if there are bigger alliances, the gameplay might not be the winner.
Evolution recruitment troop quality: No, quite some work that would be, needs time i normally have not. Also, this is reflected to a degree already, ie. the "late" yari ashigaru (calling Bushi Yari) are way more capable than the default starting Ashigaru, besides this, experience ups via building/researchment offers this to a degree. But recently i followed your thought to decrease/cap strictly the exp points via buildings, they are all even now, max 1 point. Look, exp points primarily just up morale, melee and defence skill and accuracy and reload, what you would do else via different troop quality (difference ie mere then only unit number if this is different). But now, as recruitment brings out the same quality, battle exp is more important, while the gaining of exp levels and its bonus is harder to get now, so a Samurai will have always relative more capability, even if a good Ashigaru with some exp comes, but the best Ashis will be a danger for Samurai, as for numbers and gained skill. I consider this indeed more balanced in regard of chances for the factions if we bring out even skilled units per buildings without big differences. If we make an evolution of different troop quality per building tiers, then a faction with less money who can't construct it, can't recruit such units, and has never a chance to win a (mid-late) battle (here always viewed for the AI).
Lowering stats, routing: That might be apoint to revise for the combat model, yes.
Exp threshold: A file manages this, there are 9 xp levels (so a unit can get max level 9 xp, i think), every level has a threshold value, just higher per level, which i understand as delay-sum for the xp levels. I upped the values heavily from on xp level 3, real high exp units should be near impossible now. While auto-resolve battle always gives more xp automatically, so a cheat could be to autocalc battles, but where the fun then? Anyway, me personally, if i playtest i autocalc, but only as for time anf to see the default battle outcome, if i play a real campaign, i play every battle.

Still to the recruitment theme: Now since the 1.9x versions i added units to lower tier buildings. I still play with the thought to put the yari and sword ronin into the castle tiers (now it requires the sake den building) what you suggested, ie. indeed from on fort, but latest from on stronghold, also money plays a higher role then. My sorrow then is Shimazu, that recruitment design could kill them with certainty. However we can test this. Or another way: Two S2R+ variants, just one would offer a design that increases the challenge for a Shimazu player or Oda etc., the other (default as of yet) is a mere AI balancing in favour of the major key factions.

Edit: On a second thought, i guess the two variants model is a good idea, and quick done. The S2R+.pack files will be different named then, the "default" one is then "_Key-Majors" (reflects the whole mod-design as of yet to support the major key factions Oda, Shimazu and also Mori). The other variant then"_Key-Non-Majors", that means if you play Shimazu or Oda or whomever, the challenge is increased, while the probability that ie. Oda-AI or Shimazu-AI (ie. you play Date or Chosokabe) to survive/expand is clearly decreased.

[GShock]

Is there also a way to increase the factor for disbanding units with heavy losses? Sometimes I get to very low numbers but the unit is not disbanded. By using CTLM the game bypasses the fact different experience can't be mixed in the same unit and so you are basically able to replenish good troops with bad ones. I said 66% losses are a good compromise so that if a unit of 99 base man loses 66, those 33 remaining have got to be disbanded.
Do you think this is possible?

As of the alliances it's how it really happened, they were super important and many clans were doomed for betrayal, so many it's useless to recount them. You can't really fight the whole Japan single handedly but if you can keep good allies from the start when you are about to become Shogun they will not attack you (realm divided). I think if the allies are kept strong with the ally help factor, it would be very interesting and it would keep many clans active for longer. Meanwhile I have read that Darth enabled a much more active Ashikaga shogunate which now behaves as a real faction, that would also be nice in S2R+.

Finally, you said so yourself, according to the player's choice of clan there might have to be different files. That's because you picked the path of history against the what if which is easier. :) :) :)

I dled the update but had no time to play, will do the possible this week end.

[DaVinci]

A next update 1.97.5 will be released today, probably. I've got too good ideas in the meantime to hold them back.

Do you think this is possible?

Iirc. it is possible in ETW and NTW, need to find the file and column. Edit: Can't find it, but if, then i'm with you, the suggested percentages are good, if Samurai, the left 33% will make seppuku ;) and if peasant Ashis, go home.

Meanwhile I have read that Darth enabled a much more active Ashikaga shogunate which now behaves as a real faction, that would also be nice in S2R+.

Interesting, but hmm, will delay the course of the campaign-play, i guess, and more a total what-if, maybe some day in future as extra content, i'll do such a thing.

[GShock]

I really urge you to study the DMS files. A guy in the DMS feedback thread has found a way without boosting to save the Oda Clan. He basically managed to make Tokugawa go on one side and Oda on the other as per historical paths. This for you who are a history fan might very well be the cherry on the cake because if this can be done with Oda it can be done with anyone (and that's probably what he's going to do with Ashikaga who, by v2.35 are supposed to expand but don't... it is IIRC a small mod that is compatible with dms, it's all in the feedback thread of DMS and it would save you a lot of time, besides Darth is a guru in modding... there's always something to learn from him). You could for example tell Shimazu to fight Ito first and not Sagara, tell Sagara to fight Otomo first and so on. Naturally... this logic doesn't ensure the clan will actually make it but ... at least diplomatically it will help keeping the AI alive (i.e. shimazu won't die so soon).

Anyway, I really really hope you can find that file. (Neither DMS nor SR2+ disband low number units but IIRC vanilla does). There's something amiss.

Please, include in the next update ALL the files you are using for SR2+ and not just the SR2+ pack because I want to create a folder to keep them in for safety purposes.

[GShock]

Beginning Tests on 1.97

[GShock]

I was surprised to see the new bow "ronin" (wako) and I like it. Meanwhile, congrats on another excellent choice, the teppo ashigaru recruitable in Satsuma. You're right of course, the Shimazu bought the first teppo then duplicated them later on but they were surely the first ones to employ the fire arms (and not the clan Otomo at Bungo). Excellent.

I was surprised also to see the melee ronin was missing. I didn't understand why and then I found it was recruitable from Osumi (but not Satsuma). What changed? A localized recruiting capability where the wako can come from Satsuma while the melee ronin comes from Osumi? I think ronins should be recruitable at any tier level in any province. Anyway their current upkeep cost seems rather low and, yet again, I think ashi should have a higher upkeep, the second highest (ranking should be: ronin most expensive, ashi right second, samurai third). That would ensure a sort of "population cap", especially if the numbers are corrected. IIRC statistically samurai were really few when compared to ashigaru numbers. Perhaps 3 to 1, hence a bow ashi of 90 men should correspond to a bow sam of 30 men only. You can do this because you have an outstanding barrage of differently armed ashigaru to keep things lively (and appropriate), if you want.

So what is the logic behind the wako and ronin? Why 1 in satsuma and the other in osumi? Just curious.

The yashi recruitable after bushido is learned costs more than the standard yashi... it requires an art, it costs more but has worse combat stats... odd. I had already noticed it, is it intentional? I don't understand why.

The bow as a weapon is one of the hardest to master and a weapon for the elite not for the masses. I think the bow ashi are too numerous, too deadly and think their accuracy should be just slightly lowered (still much lower than the bow sam's) as well as their morale (they seem to hold too much at lev 0xp).
Besides I didn't count how much ammo do they have as BASE (without the art that doubles it)? Shouldn't they have the same ammo as the bow sams? Quivers don't come in different sizes but I have no historic data at hand. I have the impression it's too much ammo already and when doubled it's even more.

There's a prob with sieges (another one lol). Have you checked how many men are lost by the defender if the attacker keeps the siege? After one season has passed, the losses are just THREE men. Can you do something about that (checked in hyuga, castle lev1)? Same should be in winter for attrition losses.
Also, the losses implied in the autoresolve seem a bit too off when compared to the manual battle. This is especially true in sieges where the attacker should be more heavily penalyzed. Wht I am saying is that it seems to me it's easier to win by autoresolve (and not just in sieges) because losses are equally shared among all troop types while if you manuallly do the siege you will take many many more losses.

Please see if you can make the mod reduce the number of provinces needed to win and force more vassals. Vassals are useful. They fight pirates, they fight battles, give you half their koku and you can trade with them for extra money whereas it's very hard to trade with other major clans. Factions that are about to be killed should actually ask to become vassals but it's the AI that should do that when convenient (sometimes it is but right now if you need 25 provinces at least it's a suicide to take a vassal instead of its last province).

[TylerX5]

I started a Chosokabe campaign on h/h and it is really difficult to earn money without main land land trade which makes everything else hard as well. Is there anyway you could mod the game to make it so the Chosokabe could land trade if they capture a province on the main land?

[GShock]

build a light bune asap and go East to make first contact with other clans. Chosokabe needs a very careful go at the beginning but once you've got the trading port you're good to go because you have a major resource: wood.

[DaVinci]

while the melee ronin comes from Osumi?

That's a wako sword warrior, all wako's are recruitable by default via "pirate's lair" building (name? or something).

The yashi recruitable after bushido is learned costs more than the standard yashi... it requires an art, it costs more but has worse combat stats... odd. I had already noticed it, is it intentional? I don't understand why.

Did you notice the name-part Inaka? And house name Shimazu? The Shimazu is a house soldier, merely standing army, the Inaka is peasant, raised from the region. Therefor needs more time to getting them being "soldiers" and more money as they are not available for harvesting the land.

The bow as a weapon is one of the hardest to master and a weapon for the elite not for the masses.

Therefor the recruitable number is limited. Bow Ashis are a sort of semi-elite Ashigaru soldier in my mod. Btw., there is a thread "Unit Guides".
Bow Samurai are standard Samurai (besides the Yari Samurai), fire a few deadly arrows and go into melee. Samurai's were trained and equipped with bow, yari and sword, this is not possible to reflect - the compromise in this mod is the bow Samurai who is an expert archer and uses this weapon short. For arrow raining, the Bow Ashi is the one. As for lethality, some guys found out that the number of casualties per battle by arrowswas significant in the period. I formerly had weaker arrow lethality what made Ashi archers not cost-effective, if this now might disbalance battles, i can reduce it again ... but after all they are expensive units, and also have good melee value.

There's a prob with sieges (another one lol). Have you checked how many men are lost by the defender if the attacker keeps the siege? After one season has passed, the losses are just THREE men. Can you do something about that (checked in hyuga, castle lev1)? Same should be in winter for attrition losses.
Also, the losses implied in the autoresolve seem a bit too off when compared to the manual battle. This is especially true in sieges where the attacker should be more heavily penalyzed. Wht I am saying is that it seems to me it's easier to win by autoresolve (and not just in sieges) because losses are equally shared among all troop types while if you manuallly do the siege you will take many many more losses.

There is a file that can influence autocalc battles, perhaps i'll take a look.

Please see if you can make the mod reduce the number of provinces needed to win and force more vassals. Vassals are useful. They fight pirates, they fight battles, give you half their koku and you can trade with them for extra money whereas it's very hard to trade with other major clans. Factions that are about to be killed should actually ask to become vassals but it's the AI that should do that when convenient (sometimes it is but right now if you need 25 provinces at least it's a suicide to take a vassal instead of its last province).

The problem is the win conditions are not simple db files, it needs startpos.esf modding. I wanted to avoid that.

[GShock]

Here's a little analisys of the unitstats in regards with some of the basic units recruitable by Shimazu at the first turns (just conquered Osumi). There's something to discuss about.

I grouped in color-code stats that can be assimilated to multiple unit types (i.e. green is used by everyone while grey only by missile troops) for ease of reference.

Since these numbers make no sense when taken by themselves but do make sense in comparison with the numbers of other unit-types, I'll begin with the upkeep with my personal opinion on each relevant stats and stating since now it's guessed figures to be fully tested (which is relatively easy indeed, much less problematic than it seems) before actual implementations.

I'll take the basic value of 10 for morale (as I said many times the troops of SR2+ but also of other mods stand ground too much) so all troopers should have a basic value of 10 with the current modifiers. The CLASS is the main modifier, secondly, missile troops have a negative modifier for morale because they are skirmishers (as you can see, modifiers are ultra light. The concept starts around the MAN and how he is employed not the numbers which must be similar to one another and much less relevant).

A-Teppo -1 (is missile unit) = 9
A-Shin Tachi 0 =10
S-Ronin Daikyu (Samurai units should have +3, [ronin +2] but missile = -1) = 11
A-Naginata 0 = 10
A-Yari Shimazu 0 =10
A-Yari Inaka 0 = 10

I'm taking the same base for combat stats of 10 (green). Here it's the CLASS and the WEAPON that make the difference plus the unit type.

Melee Attack
A-Teppo sword +1 minus missile troop -1 =0
A-Shin Tachi sword +1 (is melee unit) = 11
S-Ronin Daikyu sword +1 (is missile unit -1)= 10
A-Naginata 0 (not as effective in the fray because it's cumbersome) = 10
A-Yari Shimazu 0 (more cumbersome than nag -1 and it's def weapon -1) = 8
A-Yari Inaka 0 (same as above) = 8

Obviously, after the charge bonuses are discarded the sword is the strongest weapon but do mind that samurai here have higher morale and sword equipped will beat other weapons both in morale and in melee attack (they kill more). Missile troops also are losing in melee. This is obviously intended to have balance in realistic unit numbers where ashi are 3 times more numerous but at disadvantage with morale and skill.

Melee Defense
A-Teppo sword +1 minus missile troop -1 =0
A-Shin Tachi sword +1 (is melee unit so has no penalty) = 11
S-Ronin Daikyu sword +1 (is missile unit -1)= 10
A-Naginata 0 (not as effective in the fray because it's cumbersome) = 10
A-Yari Shimazu 0 (more cumbersome -1 but +1 is def weapon) = 10
A-Yari Inaka 0 (same as above) = 10

Charge Bonus (10 = 0 = no charge bonus)
A-Teppo sword 0 has no charge bonus minus missile troop =10
A-Shin Tachi sword 0 has no charge bonus but is melee unit +1 = 11
S-Ronin Daikyu sword 0 has no charge bonus - missile troop -1)= 10
A-Naginata +1 (long slashing weapon) = 11
A-Yari Shimazu +2 (longer stabbing weapon) = 12
A-Yari Inaka +2 (same as above) = 12

Bonus vs Cavalry (10 = 0 = no bonus)
A-Teppo sword 0 has no bonus minus missile troop -1 = 9
A-Shin Tachi sword 0 has no bonus = 10
S-Ronin Daikyu sword 0 has no bonus - missile troop -1)= 9
A-Naginata +1 (long slashing weapon) = 11
A-Yari Shimazu +2 (longer stabbing weapon) = 12
A-Yari Inaka +2 (same as above) = 12

Armor
The japanese never made a mystery of their armor being just a bit more than fashional... but the samurai DO have an armor, ashigaru basically had a lot less than an armor indeed. Some were even armorless.

10= armor = 10 (selected guessed basic value)
A-Teppo ashigaru -1, missile troop -1 = 8
A-Shin Tachi ashigaru -1, melee +1 = 10
S-Ronin Daikyu Samurai +2 - missile troop -1)= 11
A-Naginata ashigaru -1, melee +1 = 10
A-Yari Shimazu ashigaru -1, melee +1 = 10
A-Yari Inaka ashigaru -1, melee +1 = 10

Range (here it's about weapon type and training skill) 10=basic value
Teppo had a maximum range of 150yds (though I doubt they could really hit anything beyond 100yds) while the shortbow (bow ashi) was effective at 200yds and the longbow (bow Sam) almost got to 300 yds. Range is one main factor but accuracy is the most important factor. These 2 are linked and must be compromised together.

A-Teppo ashigaru = 150
S-Ronin Daikyu Samurai = 300
A-Daikyu Ashigaru / Wako = 200
S-Daikyu Samurai = 300

Reload skill (I think it's reload time + reload skill in S2TW) taking the expressed value as SECONDS between reloads.

A-Teppo ashigaru (very difficult to reload in combat, each shot should take about a minute time and I am taking as reference the U.S. civil war muskets where a GOOD soldier could shoot 3 rounds per minute) = 45

S-Ronin Daikyu Samurai = 20
A-Daikyu Ashigaru / Wako = 30
S-Daikyu Samurai = 20

Accuracy (taking 10 as base value, most likely this factor is heavily modified by weather/distance. This stat can be used as master to affect the effectiveness of all missile troops in combat thus giving more relevance to melee vs missile which was traditionally more inclined to the Sengoku Jidai warfare)

A-Teppo ashigaru (very accurate +5) = 15
S-Ronin Daikyu Samurai (trained) +2 = 12
A-Daikyu Ashigaru / Wako (poor) = 10
S-Daikyu Samurai (trained) +2 = 12

Ammo
Very easy here it's about number of shots but having no reference I am guessing the size of quivers and balls+flint+powder for teppo.

A-Teppo ashigaru = 24 (half the standard value of US civil war equip)
S-Ronin Daikyu Samurai = 12 (good arrows)
A-Daikyu Ashigaru / Wako = 16 (poor arrows)
S-Daikyu Samurai (trained) = 12 (good arrows)

Upkeep
Here the concept is simple, the base value of one soldier is 1 rice bowl a day per 1 year divided on 4 seasons. This means this value has to keep in mind the Samurai/ashigaru ratio in the armies which was basically 3 to 1 (actually much much much more in favor of ashigaru). For ease of reference I am making the count on a Samurai unit being 30 men (all samurai units being the same size) while Ashigaru units are 90 men size.

Also, Ronin cost more than anyone else in upkeep, should be recruitable everywhere (according to the structures, I like the idea of the wako daikyu and ronin shin tachi difference) and have high replenishment rates.
Ashigaru should have intermediate replenish rate and high upkeep (arms stripped off the agriculture!) and clan samurai should have low replenishment rate, very low upkeep and very low recruiting costs. I'm making a table for ease of reference.

Type Number Recruit Upkeep Replenish Max units
Ashigaru 90 Medium High Medium Unlimited
Samurai 30 Low Very low Low Many
Ronin 60 Very high Very high Very fast More than Samurai

(Extra modifiers: Siege unit recruiting and upkeep + 50%, Cavalry unit recruit and Upkeep + 25%, missile unit recruiting and upkeep +10% ) The estimated upkeep cost of a 90 man ashigaru unit (the weapon used for melee makes no difference but of course 10% to be added in case there's swordsmith in the province) is 90k per month meaning 270k per turn (just for the food!). Such system would require a thorough estimation of how many men each single region may sustain so that the upkeep expenses + the limits on samurai and ronin numbers may effectively add a population cap which is missing in S2TW.

Now this is more or less what I had in mind. Of course it's just a few figures and basic concepts, I more expressed it so you understood where I start from (1 man) and how then everything else adds up to make the army. Back to playing S2R+, if you survive reading this post, let me know what you think.

I hope I didn't mess up too much with maths ... lol

AHAHAHAHAHA

[DaVinci]

Good God ... lol ... no, not a problem, i know far more complicated unit models.

I like your cost-approach there somehow, but it would change everything, the whole balance would change, my friend. This would need a lot testing-phase in how far the impact works, perhaps one can leave the other cashflow model of the mod as is. However, i would like to test it somehow, also some of your combat-change ideas, but i have to look into it deeper, only crossread as of yet. But really, as already expressed, i wanted to finish the mod soon ... but your contributions are too valuable to miss them in this mod, so the development might go on after v2.0 (which was planned as final version). Your model could be a new design-approach, which could be stuff for a S2R+ relaunch after v2.0.
Btw., you should enter the dev forum as S2R+ team member, so we can discuss etc. all such complete new approach-things internal. For this you look at the "Settings" tab, and there you go for group-membership application (edit: here calls "permission-groups" .. i'll check this when log-in the next time to see if you applied for that).


Btw., i found and will apply (implement) the Oda-Tokugawa vassal mod, big thanks for this tipp :)

[GShock]

The point of my suggestion on starting as scratch is that this is the only way to develop things slowly but consistently towards the final goal which is perfectly reachable, unlike in other games. Once the basic concept of history is commonly shared in dev schemes, then it's rather easy to say: hey, this is a game designed to sell to the masses, we can make it somehow 100 times better by taking out the crap of it and making people understand it's not a shopping center, it's war and troops must be used accordingly.

A strong melee requires a strong armor to stop units from dying too soon, however, this messes up with missile damage. It's just an example of how ***the more you get away from the MAN factor (1 man = 1 man not 1 superman) the more things start to go bad***. (I think it's possible to slow down the combat animations, wouldn't that allow longer battles with more time to react? What does the other .pack file do?)

I am particularly concerned about the terrain/weather factors. You can for example alter the bonus vs cavalry but you CAN'T alter the Guard ability factors, how much the weather will affect missile hit ratio, how BAD the cavalry will behave when fighting in the woods and so on. The hard-coded factors must match with the unitstats file and do not forget the more XP, Arts, equipment (etc etc etc etc) get into the game the more things will go awry. Cavalry for example never seems to get really tired. It keeps running up and down the field (but in a more general sense, all troops in SR2+ seem to never get tired enough according to me because their combat stats are already highly affected by too many positive factors and fatigue has been blessed and said goodbye).

The AI of SR2+ behaves very well I have some more data after this morning of tests. First of all, despite the so many units you introduced which have no corresponding factors in vanilla I just met and defeated a sallying out Shoni army at Bungo in a 1200 vs 1200 battle (he shouldn't have sallied out but that's really something HARD to find I guess) and its troops were VERY finely balanced between missile, different ashi and a few ronin and samurai. Outstanding.

The typical result is that I won and in the following siege I had something like 800 men vs 40 men of the AI. Such is the catastrophe when the units fight for too long. They are so exhausted they can't escape the chase and it's the opposite of what should happen. We must absolutely keep the AI armies alive even when defeated... on the winning side I did lose 1/3rd of my men and yet no units were disbanded (that's another thing we must absolutely find. In vanilla it does work though the threshold is still too low imho)... still, because of your excellent idea of making maximum number of (some) units capped, CTL M is mandatory and so some units are lost in the reshuffle (though sadly, the experience mix should make the new units drop and it doesn't, damn CA, but better than nothing). Perhaps lowering the general's morale is the key so that when his unit is weak he will rout taking his army with him, and making it fight another day. PERHAPS I said... impossible to say without starting from "scratch". What's sure is that I seldom win without killing a general in battle and that's very bad. He seems to be pretty reckless.

Meanwhile:
1) Todofuken ability is supposed to give you the knowledge of building the roads. However this doesn't happen, the option is still greyed out after having the art (and the money). Probably you associated the roads to the Equal fields (just straight below Todofuken) which I would associate to trade posts & roads. Check that out.

2) My monk has spreaded rebellion all around Shoni's lands, however the spawned armies are well too weak to trouble the defenders. Shoni is Christian, my Monk is Zen so it should be a religious revolt with greater numbers and possibly a few ronins. If it's possible, there's room for improvement here.

3) I recruited and produced both teppo units. The ability to fire in ranks is MISSING so they only do the volley. However, I noticed something that I had been looking for: the LOS concept. CA this time did it nice, Muskets are not able to fire above your own troops so they don't shoot if they don't have a line of sight, while archers do. I was happily surprised. :)

4) Chosokabe in the hands of the AI also seems rather weak. Sagara didn't do anything while I conquered Ito and declared war on Shoni. It's still camping in Higo with castle full of troops... too static. Chosokabe ... well I can't see from Kyushu but definitely didn't conquer a single province so far... either static or constantly being defeated (without elimination though).
Currently it's year 1548.

5) We could use more generals (more requests to enlist) They seem to be rather too few.

6) It's somehow odd to see that at the first round of siege (stronghold) the shoni armi lost THREE men in 3 months (1 season) of siege... damn that's bad.

7) Winter losses seem too little. 2 men per unit in a winter season.

[DaVinci]

GShock, please head over to the private dev forum (internal 1.97.5 is waiting for a download by you).

[GShock]

I got it, seen changelog, ready. :)

[TylerX5]

Sorry if you've already discussed this topic, but over at twcenter.net there was a topic (http://www.twcenter.net/forums/showthread.php?t=434941) about the AI spawning units out of the player's view, which isn't really a fair challenge but a poor substitute for one. Yarkis de Bodemloze came up with a beta mod (http://www.twcenter.net/forums/attac...2&d=1300633011) to fix it. I think this is something to look into that could make this mod better.

[DaVinci]

Sorry if you've already discussed this topic, but over at twcenter.net there was a topic (http://www.twcenter.net/forums/showthread.php?t=434941) about the AI spawning units out of the player's view, which isn't really a fair challenge but a poor substitute for one. Yarkis de Bodemloze came up with a beta mod (http://www.twcenter.net/forums/attac...2&d=1300633011) to fix it. I think this is something to look into that could make this mod better.

Thanks, It's known. If you scroll through the thread, you'll see that i commented there as well (mere the last pages).
In S2R+ i have solved that issue in the meantime (since a few versions), at least i never encountered such sudden unit uprisings since ages*. If you experience that though, please describe the situations.

* what i mean here rather is that AI didn't apply unit caps, else, in general i don't believe in AI unit spawns out of thin air.

[DaVinci]

Info for the meantime: Me and GShock working deeply on the S2R+ development, many things will be changed/improved with the upcoming public version.