Tip(list) of the day! :)

RSS
Email

0 Kudos

Author	Comment
Bozo.shoguntotalwar	Tip(list) of the day! :)	Lead	[-]
Posts: 24 (11/16/09 10:48:29) Reply Quote More My Recent Posts	Tags : tiplist, day Since I can't stop playing this game I thought about taking notes on suggestions I am sure will be helpful for other games of the branch for future releases or... "retro engineering" I am by now playing the hard level and everything is different now. All clans are more aggressive and alliances are volatile while the supply production is so low my 2 Takeda armies are almost always out of supply calling for a more defensive play. I noticed the structures by each castle are different every new game you start. That calls for a different strategy every time, it's just perfect. Too soon to assess the capabilities of the AI and the overall level of gameplay now. I think an option to semi-randomize character statistics would REALLY benefit this game. I have NEVER played a civil war game (got all of them) with fully revealed statistics before, part of the good commanding strategy is to put the best man in his most suitable position and semirandomizing + hiding statistics makes all games more challenging because you come to know of their stats only after they get involved (in combat or by some other mean). This is another thing that would keep longevity very high. Anyway it was just an observation, trial and error makes the game much much harder for any player but you know that governor might NOT revolt, that general might NOT be so good.... every game = totally new game, and new quest to discover who's the best man for the best job. 1) I have noticed that now the mons appear on the character banners and in the field they are correctly represented with an exception... the red ones. There's many red banners on the field and very little difference between them. I think we need more banners, with different colors to help making things different and divisions more clearly identifiable. What I am saying is that all mons correspond to each div correctly but the red ones don't because they are too similar, that's all. Better use some other color instead of drawing new mons. That saves time and is a resource effective solution. 2) CTL A (select all divs) is still not working after the patch. On the other hand we could use a CTL+key to select all troops of the same kind. CTL Y (yari ashigaru) CTL C (cavalry) CTL M (Archers) CTL K (Kachi samurai). The 1 2 3 etc keys work perfectly and the CTL+banner correctly selects multiple divs. It would be great at this point to have a CTL1 to build a group you can call with SHIFT1. It's basically the same as TW engine but still retaining standard 1 2 3 etc keys to call on a single div, CTL+number builds the group and SHIFT+number selects that group. TW engine also has CTL+H for hand to hand troops, I would skip that but in our case it would do CTLC+CTLY+CTLK... essentially everything but archers would be selected. 3) Some DIVs regardless of their initial stance do not want to move after the army mode order is given. They are generally the ones in the back of the deployment line. It's proper that the general guarding the flags does not move but all other divs should move by that order... well some of them, just don't. 4) I've noticed we are not told when the General routs. He just disappears and of course all of the troops rout immediately thereafter. I think the player should be warned that his general is routing. It's not a minor event to pass unreported. 5) I think we really need a key button to stop and another to deselect a division. In the middle of the fray it's hard to move the mouse everywhere, select this and that, try not to make mistakes and key shortcuts always help. A button I would find particularly helpful would be the "go to army mode". 6) Driver problem? Sometimes I've noticed after rebooting ... that the screen starts in 1024x768. Never happened before to me. I'm not saying when I quit the game, it's ok it goes back to 1440x900, just next day as i start the laptop sometimes it starts in 1024x768. I would love to try the game in my standard resolution... unfortunately in last patch it didn't work... since I saw nothing about it in this latest patch I haven't tried yet... will let you know, most likely it would solve this very minor problem. 7) Thought about a CTLT to increase speed in the battle mode? Thought about selecting a division, then right clicking on the minimap could be handy to give move orders instead of rotating cameras everywhere (it's useful for long distance marches). Similarly, we could use a SHIFT+right click to give a waypoint move order. 9) It's nearly impossible to select multiple divs or even a single div if the HOLD LEFT MOUSE + DRAG is not really precise... precise to the last man. 10) Considering the hard level is really tough on supplies, maybe the supplies at normal level aren't being used at all? This is truly worth checking. 11) Engagements. Armies didn't always fight on first sight. I know the timeline is seasonal and not daily... but I wish to move an army to block some other army's path without engaging it. Unfortunately as soon as the 2 armies get in touch I am always asked if I want to fight or simulate but not if I want to wait and see what happens. If I manage to block a mountain pass or a road, it's the enemy who's got to engage me and I am on defensive instead of offensive which is a big difference. Is it possible to have a button that says WAIT instead of just YES play the battle or NO simulate it? That being said, let me go and play now bc this game is really too addictive! Just thought about reporting my impressions now that I have 80% of the game functions in my grasp. Ming I hope you take this list as it is: Totally constructive critic to make the game even better than what it is: The best in his segment!
Bozo.shoguntotalwar		#1	[-]
Posts: 25 (11/17/09 03:02:44) Reply Quote More My Recent Posts	Updates on the tiplist... oh the more I play it the more i love it! 12) Are we sure the army defending in a siege should really rout? I think they should only abandon the field when both flags are cut down, that means the attacker has to enter the keep and take arrows till the last moment. At that point, the battle should end because we know the defenders of a siege CANT escape the castle and get out of it. The result should be 0 survivors and 90% of the generals dead (seppuku), 10% captured. 13) A strange episode happened just right now... I was attacking an enemy castle to which i had laid siege. There were 2 armies inside, and i could see the camp icon INSIDE the castle. If within the castle walls there's no supply consumption why did these 2 armies camp there? They ideally should have gone outside the castle to organize a skirmishing battle, cause as much damage as possible and then retreat inside. That would delay me one turn at least. On the other hand, and ignoring this losing strategy the AI adopted, I proceeded with the assault and beat the first army during the siege. At the beginning of next turn I could see the castle was mine (I could even post a Governor there) but there still was the second camped army inside. I had to re-engage and this time the battle was in open field. I won again and finally the castle was empty of enemy armmies so i could disband my army inside the new castle. Something ain't right here, just thought about reporting it. 14) I had already said it looks really bad to see 1 man surrounded by 50 and still escape alive. This one man, 90% of times is a general on horse. I understand generals must be protected from normal combat rolls but if a general is routing, to avoid this gamey show to take place, I think he should rout much much faster than this. He will still escape, just like now, but at least it's more realistic... he routs and routs for good with much less chance for foot soldiers to catch him if he's much faster. 15) Actually right clicking on the minimap works as move order so I take back 8 ) I still would like to see the selected div(s) on the 3d world highlighted in the minimap. That would help identification and moves taken exclusively via minimap (it's generally long range marches). 16) Since Taking enemy flags is an important thing (and by the way I like it in hard level there's much less promotions and medals to generals!) I suggest to decrease the time it takes to really finish the battle. The idea here is to take the flag before the battle is over and not exploit the humongous time available for the player once the battle is over, he clicks continue and then goes for the easy kill on the flag because the enemy is all routing and can't stop him. If you click on continue battle, you should have no more than 1 minute to trample as many routing troopers as you can and that includes the flag. If you didn't manouver with the purpose of cutting the flags per time and focused all your troops on the main combat then you should understand the fleeing general and his staff will take the flags with them, you can't ask for 5 minutes to take that flag for free and then take promotions and medals for it. You didn't take it in combat, you are exploiting a gamey feature. I think 1 minute is enough after you click continue battle... this also helps with the problem of fleeing generals who are surrounded by your whole army and they still don't get cut down. Just a few more cents... expect more.
Kuroitatsunoshi		#2	[-]
Posts: 74 (11/17/09 04:19:22) Reply Quote More My Recent Posts	I'm glad to see other people love this game as much as I do. I'm enjoying reading your notes. Some notes on your notes: For the randomization on Generals, it might make for a better game, but would make for a worse simulation. The game was designed to be more of a simulation according to an interview we at the Samurai Archives did with Ming. Maybe giving the player the option to randomize or not might work. 3) The two 'reserve' divisions (the ones in the spaces on either side of your Marshall's camp) are not affected by army orders. They stay there to guard the camp and wait to be committed at an opportune moment (unless you instruct them individually to do something else). That's probably what you're seeing. 12) Japanese castles were built with many, many gates and entrances/exits-there was almost always a back door to escape by. Even in the massive siege of Osaka, the majority of the defenders escaped the siege. In addition, Japanese strategy usually called for leaving an escape route open for a beaten enemy-a cornered enemy was considered to be far more dangerous to fight, and leaving an escape open for them was usually done by any attacking army (field battle or siege). The casualty rate in any pre-modern Japanese battle therefore rarely, if ever, exceeded 10 % (killed and wounded) in a battle. 16) Flags in a general's camp were often abandoned, and looting the enemy camp was a major post battle activity. Depriving your samurai of the chance to 'loot the flags' and chase down a beaten enemy (who was given an escape route in hopes of inciting a flight) wouldn't be accurate.
Bozo.shoguntotalwar		#3	[-]
Posts: 26 (11/17/09 06:32:55) Reply Quote More My Recent Posts	Hello Kuro, wait for me to complete a campaign first with a couple of clans, then perhaps we can play a Multiplayer. Yeah I just love this game. I know the game is as historical as it can be. Retainers, famous and less famous generals are all there. It must have taken a huge deal of study to build such a game but to a keen eye, just the battlefield of castles is enough to tell. The problem of semi-randomization is different from full randomization. The semi-randomizer still keeps the core of statistics intact with a slight + or - to the basic stat. This means the overall stats won't be too different from the historical recreation. This is how the most evoluted games based on characters work (example, GG war between the states). The full randomization might give... well it would surely give unpredictable results but it still would fit into the "what if" scenario this game struggles to recreate. I personally don't like much the full randomization but it does give much a higher challenge to the player than the semi-randomization does. However, neither system of randomization works when the statistics don't go hidden. The 2 things must always be together because it doesn't make sense to hide Shingen's stats if you already know by history (or by gameplay) that he's the best general the Takeda have. I think giving the choice to the players is always good. If it doesn't work for a player it will certainly work for another and, provided the game only allows multiplayer with identical setup (which is how it actually is) neither party would be dissatisfied. I personally would be greatly satisfied because I talk on personal experience, there's no game where semi-random + hidden stats has disappointed me and all players I know who have tried, have never gone back to the default system. 3) That's certainly what I am seeing, thanks for the clarification. It does make sense. All the army moves but the general and these 2 divisions unless manually instructed to move. 12) I know that but the problem is that the defending army actually doesn't flee anywhere once the battle is over. If the soldiers and generals could be scattered among the neighboring friendly provinces, I would be extremely happy to make it so. However, what happens is that basically at some point in the battle (and the reason is definitely not the killing of the marshal that's for sure) the AI simply routs. There's no point in making it rout so soon because the AI always leaves a garrison by the last gate, in proximity of the second flag (where the keep is). I can understand a rout and a defeat in a siege if the defenders are beaten but if 2 or more divisions have not even been engaged I can't see a good reason to make the defender rout. Actually the first line at the gates should go back to rally with them for the last stand before the final rout... which would end as you said outside of the castle but I still stress the importance of not losing those troopers. If they escape, they escape. If not, they should fight to the death and, in any case, they shouldn't have a mass rout so easily... they are inside their castle not on the open field. 16) I understand that but unfortunately I still think the available time AFTER the battle is concluded is too long. It's an exploited promotion basically though much attenuated at Hard level of difficulty. I'm not saying this should change ... just that the available time to slaughter the runners and take the flags is much more than necessary... especially if you have cavalry... which is.... always. A minute less would do... and it would also help the AI regroup for next (possibly last) defense. 17) *NEW* I failed a siege, my Marshal routed and my army routed through the gates of the castle I was sieging... perhaps the path to the rout should take the soldiers back to their flags and on those edges of the map and not inside the castle itself on a much longer route (and full of enemies who will slaughter them)
Kuroitatsunoshi		#4	[-]
Posts: 75 (11/18/09 04:24:51) Reply Quote More My Recent Posts	12) This is pretty standard for the times-especially for armies full of ashigaru. Once the walls of a castle were breached, these lower level troops tended to dissolve away and desert en masse. They'd simply go back to their farms or cities, dreading the day when a warlord would press them into service once again. Contrary to what most Western books have to say, Japanese warriors rarely would make a last stand and fight to the death-the division usually guarding the second flag in the castle would represent the daimyo's most loyal and engaged retainers who would have the most to lose in status and wealth if the Lord were to be killed-and even they rarely fought to the last unless it was the daimyo's home castle and there was nowhere else for them to run to. 17) In a way, this shows one of the main features of a Japanese castle, and something they were designed for-they were intentionally laid out to confuse and misdirect attackers and make it more difficult to advance on the keep, or retreat once inside. But I've had a few instances where I haven't been happy to see my troops stuck inside, too!
minglee7		#5	[-]
Posts: 1013 (11/18/09 08:02:52) Reply Quote More My Recent Posts	Thanks a lot for all your comments and supports. We are taking notes and will try our best to improve the game from it. Regards the routing issue, because we are using a chain-of-command system, every individual units' behaviour will bubble out to the whole army. The balancing is quite time consuming for us so it won't go extreme one way or the other. We did put some bonus for castle defender's morale adjustment, but when the general's leadership or loyalty is low, they still relate to the collapse of morale pretty quickly. I understand as a game play, it might look silly to see the whole army run away even before the enemy starts punching them, but for now, maybe just think it as a historical spec in some degree. Same applies to the trapped attacking soldiers. In the code, we ask all attacking units to retreating toward their army's original attacking direction, but sometimes the path they find make them look like they are running to a wrong direction. And, sometimes they really got trapped because of bugs. (sorry ^^)
Bozo.shoguntotalwar		#6	[-]
Posts: 27 (11/19/09 00:01:03) Reply Quote More My Recent Posts	Eh... I had already posted this but eventually I clicked the wrong forum button... luckily I could retrieve my notes from the garbage bin so I can repost it again, adding some more since I've been playing extensively. It's ok Ming there's no bug-free game in the world. I am happy you're taking notes, these impressions can really improve the game if implemented and, besides, if other games share the same engine and code, by implementing in one you can easily port into others thus bettering all games not just this one. 18) (multiple issues) I am at war with Nagao clan (the one who owns Echigo) and attacking Kasugayama castle. I have now arrived to the 8th consecutive siege with multiple armies, all of them committed and destroyed while one always keeps the siege in place to avoid the trapped defenders from getting reinforced. Kasugayama has started with almost 5000 troops and good generals and they keep improving thanks to my defeats. I cannot field more than 12 divisions and even if i destroy some of theirs, there's simply too many troopers inside this castle to start with and they are always numerically superior...at least so far. There seems to be no way to beat the defenders with less than 20 sieges and I repeat, with 4 or 5 armies at time in full strenght I can still only commit one at time, with 4 armies I won't commit the 5th to keep the siege in place so I further stress the importance of fixing the problem of besieged troops never dying to starvation, disease or desertion. During more than a year siege by now, I haven't witnessed a single Nagao army come and help rescue the besieged. The Nagao seemingly have no friends and no other clan came to help them either. This is pretty bad for the AI and needs to be worked on imo. On the other hand I am pretty sure the last army I always leave besieging being the weakest, if the Nagao had chosen to attack after winning 4 battles in a row, they would have surely defeated me (lacking supplies sometimes or numerically still inferior to their huge garrison). That's another thing to work on imo. I was also helped by the Uesugi clan who have a "request war on clan" in place but no good so far. I spotted a serious problem in pathing for units, on the right way up to the castle troops seem to get lost and go behind instead of climbing up to the gate on the right, I tried with move and with attack order btu they still often get lost. Since this is a very hilly castle, I suggest the camera to fly over hills and objects else we can see through the terrain which is no good. Cam view and move is already hard enough and this could help a lot even though I know there's limits to the height the camera can go to. 19) I managed to spy on the capital of the Nagao (Yoita) and found out they have 20 Ninjas. This is pretty bad for I suspect the fact they have so many (and used so little despite being heavily sieged in their only non-capital castle available) will prevent other clans from having any (me included, I've noticed I was contacted by ninjas only after a few were slain while trying to kill other daimyos, luckily not mine [Shingen]). At the same time I wonder how could the Nagao pay to hire 20 ninjas (I counted them it's not a fantasy number they really have 20 ninjas hired!). There's definitely a problem here. 20) There seems to be no way to go back to the siege once an assault is started. I have noticed even if clicking retreat immediately after the assault begins and thus taking no losses, one of my armies was utterly vanquished. Shouldn't it go back to the siege, especially considering there's 4 more armies besieging this castle? 21) I have issued multiple requests of attack to the clan Uesugi. I mean while this treaty is in place (and they have abided to our agreements) I asked them to attack the Hojo. They said they would but they didn't. I am wondering if this option to call for attacks on other clans shouldn't be disabled when one such treaty is already in place. See... If you are to attack Nagao I can't ask you to attack Hojo until the treaty expires, that's what I mean. 22) Given the fact that once a castle reaches the supply limit and it's got full structures (no more empty lots to build on) that castle is pretty done and can't help others, we could do something with supplies in excess (cant be used for an army because there's none and cant be used to build because there's no place to build further structures) convert in gold (100 supplies = 5 gold for example), sell/give to other clans to improve relations (Uesugi selling salt to the Takeda for the same price they were buying it from the Hojo is an historical example) or even transfer to other castle (transfer 100, get 50). Something like that? 23) Besieging Uehara (after the gov rebelled... hey it took him years to rebel this time because I took away all his generals) I won this siege with just 2 standing and badly hurt divisions, when all of my army was routing the AI suddendly retreated. Don't know... sounds gamey, could be exploited. If the goal of an assault is to take the keep, with only 2 DIVs I couldn't have but yet again... 24) Very good thing is when I am beaten in the assault, the AI has many many times swooped down to take my flags... but even better, i committed sometimes all the army on one gate and from another gate while i was winning badly they swooped down to take my flag and they won the siege. Fantastic. This is really smart of the AI and I loved it. 25) We should update the .pdf manual to include the fact that special characters like Ninjas, new generals, hired special ninja (corruptor, teacher, assassin, etc) will always and only arrive at the capital. Also, we should put a note on how to move the capital because there's no instructions on that (even if we know you need to move the ruler to govern a new castle to move the capital it's nowhere written that the ruler automatically governs the capital castle). 26) The killing of one of our generals, the routing of the marshal are events that should be highlighted with game messages during the battle. 27) At the beginning of a turn, I got displayed on the main map a portrait of a daimyo whom i couldnt' recognize. It's an event like box, like the one telling you of the first battle with teppo ashigaru but in this case there was no text. Wondering what that is... in any case, the text is missing so these events should be rechecked. Cheers that's all for today.
minglee7		#7	[-]
Posts: 1014 (11/19/09 06:59:54) Reply Quote More My Recent Posts	Thanks a lot for the posting. For a quick note, 27) is really a bug. We thought we fixed it already, but I guess we didn't. We will take a look and try to fix it in the next patch. 18) Based on the current code which I believe still have a lot of room to improve, the AI takes several measurement on whether to break out or to rescue. For example, not all governors or rulers think the same way, some will weight heavier on their own safety, some will risk more for betting on the luck. In general, they check to see if the defender can create an extra army to break out, it always keep about 500+ men for self-defense. If an extra army can be assembled, then it will check to see if nearby enemy is weaker then this army. It also measures how many "good generals" should be dispatch in this army. Sometimes, an large potato army is worse then a small elite army. As for the rescue, each governor acts based on a castle policy given by the ruler, and plus his own personality to come out a "real castle policy" that the governor will actually executes. If the policy is set to defensive, the governor will try to help a besieged friendly castle in range, if his/her castle has extra army and supply to dispatch and if the defending castle's self-defense odd is bad, and if there's no other friendly armies are on the way...etc. Similar calculation applies to foreign helps. As a clan agrees to attack on a clan, it suppose to lay down its own defense and put the clan into an all-out-attack mode for the treaty. However, it only "suppose", the ruler sometimes still have other ideas in their mind. (Or, somewhere alone the line, I believe there're still some bugs to cause them not to act decisively) Anyway, thanks again for the posting and we will keep working on it. As I'm programming for the AI of Takeda engine, I noticed the game speed slows down a lot compare to earlier versions so I have to be careful not to put too many layers of AI. However, with the new hardware's help, we could slowly put them back on and make the game more fun in the future. ^^
Bozo.shoguntotalwar		#8	[-]
Posts: 28 (11/19/09 09:10:00) Reply Quote More My Recent Posts	As a matter of fact, I was assaulted by the besieged party just a couple of hours ago. They lost and we're still there sieging. The problem of this all is that at hard level, sieging an army for an indefinite number of turns costs you a lot of supplies. It doesn't make much sense when you see the enemy not suffering the effects of the siege. Supply is there, men don't die, no chance to make them surrender ... it doesn't look too good but I understand that while the besieged AI calculations make sense, the overall strategic AI should be reformed to put a priority on saving one of its castles. Things being what they are, you can disband your besieging army into the newly conquered castle and proceed to attack the next one... with the AI not taking action to save the first castle, there won't be any action to take the second one and then the third... and you have basically won over it without much fun, with an exploit merely by sending time and time again the same army that got defeated in the siege while another army pins the besieged inside and stop them from reinforcing. I think, strategically speaking, that the various AI should be ready to counter an offensive from the player based on the fact that this time period was effectively a civil war whose end could only be the victory of one clan over all the others. This means a small AI should ally with a bigger one (i.e. the one North east of Uehara should ally with Nagao) so that big rivals can clash actively instead of a passive AI which makes you become big (Takeda only starts with 2 castles) without doing anything to prevent it, even at its own expense as mentioned earlier when it lets you come unchecked to their castles and unchecked you can besiege for an unlimited number of turns and finally take it... still unchecked. What I am saying is that if Takeda captures a castle, all the neighboring factions who are not totally Takeda-aligned should perceive this as a threat and this should be perceived as a MAJOR threat for smaller neighboring factions just because the civil war can only end with the supremacy of one clan on all others. If smaller factions ally with bigger ones to whom Takeda are not allied, then the moment you try to capture them, the bigger clan behind them will intervene. That's how it was and that's how it should be imo. As Takeda, I would really think three times before attacking Uesugi if Uesugi is allied to Hojo... you know.... Sometimes the daymios only needed an excuse to declare war on their enemies. Attacking one of their allies would really be the best excuse in the world to make them declare war on me. Unfortunately it doesn't happen in the game. If the AI is fixed for the sieges, the sieges themselves fixed properly and the strategic AI is reworked to act like this, there will be much more action. At present time I have only seen 2 Nagao armies and none of them intervened to stop my sieges...at the same time they didn't really attack anywhere else. Seemingly just wandering around without a clue. Believe me I know how hard it is to program an AI...and I have no idea how you're going (if you're going) to change this but if you can make it, perhaps you should apply for a job at NASA ! )))
Bozo.shoguntotalwar		#9	[-]
Posts: 29 (11/22/09 00:14:57) Reply Quote More My Recent Posts	28) Need a reverse deployment button in army deployment screen. This "mirror" button would allow us to reverse left/right deployments actually doubling the existing formations. 29) Jar counter icon appears even if you have less than 1000 supplies in the castle. If less than 1000 there shouldn't be such icon (which is useful to see quickly which of your castles can build a facility) 30) The number of generals displayed in a castle from the main map being 40 is actually incorrect. I have 12 in one army, 6 in another, 2 in place and 20 lieutenants. I think it would make much more sense instead of displaying "Number of generals: 40" something like: "Number of generals: 18, 2, 20". 31) The generals listed into an army (in the army button mode) all show a sword in their icon (because of course they are in an army but if they are being transferred to join an army they show the blue region icon) how about displaying their division type here instead of the sword? Keep the sword for kachi div type and build new icons for their div types. That would save a LOT on micromanagement and we know how important it is to have balanced armies. 32) 5 years without a single ninja asking to be hired by me. It means all ninjas are taken (as per problem pointed out in this thread). I propose to reform the ninjas so that if a ninja is not used for 5 turns he leaves the faction and asks to join somewhere else. This is a serious thing to check ASAP. This time I ctl-c this message because the browser went into timeout and i lost the one I had written... lol
Bozo.shoguntotalwar		#10	[-]
Posts: 30 (11/22/09 00:33:39) Reply Quote More My Recent Posts	33) The nasty CTD taking place when you pillage an enemy farm is still there. :-(
minglee7		#11	[-]
Posts: 1017 (11/22/09 09:11:05) Reply Quote More My Recent Posts	Thanks again for the comments. ^^ 28) Right-click on the icon in any of the divisions on the deployment screen. You should be able to reposition them and even reset their formations before the battles begin. 32) The ninja is set to help smaller clans so the more castle you have the less they will come co help you. The maximum ninjas a clan can have are suppose to be 10 or 20. And the maximum ninja th world can have is about 100 or 200. I will take a look to tune the number better. Oh, you might know it already but just in case, if you hover your cursor on top of any clan's icon in the place mode's map, you will see other clan's relationship to each other. If you invade a small clan, all the neighbors will feel threatened and become hostile to your clan, ... and so on. Also, if you grow power and start to unbalance the world power, there's a chance all the clans will start to unite together to against your domination. When a small clan is under siege, sometimes it will try to hold on its position and wait for new developments from diplomacy. They will try to ally with your enemies and hopefully your other enemies will create a second front to force you call back your armies. It's quite a complicate calculation but through a campaign sometimes you will see those things happen. To attract your enemy to ally with them, the under sieged clan must show some worth value for the ally to be. The piling up military strength is one of them. The AI will calculate both clan's military strength compare to your strength plus your allies' strength and use that as one of the factor for the decision. Well, that's why sometimes it's not always good to sacrifice those extra units in the castle. ^^ Last Edited By: minglee7 11/22/09 11:51:55. Edited 5 times.
Bozo.shoguntotalwar		#12	[-]
Posts: 31 (11/22/09 12:38:31) Reply Quote More My Recent Posts	Ever played Madden NFL ? you hit a button and all the formations or plays come reversed. You don't need to manually do anything, they just come reversed. So I could attack on the right and use the Form Ability without having to manually stay there 10 minutes to move, rotate and change formation or aggressiveness to all divisions. Point is to reduce micromanagement. :-) The number of ninjas is DEFINITELY 20. I counted them. However, it doesn't look like they really are using them that's why i suggested for a "rotation" system. Still... something is not right. 20x250 is a big sum of money I don't think the Nagao Clan could ever afford. I have seen what you mention here with other clans being affected by my hostile actions. The problem is that Imagawa has been at war with me for 10 years already and they didn't send a single army. That's why I am having a fair game with Nagao... I can send there 4 or 5 armies because no other hostile AI is coming to kick my ass and no Nagao-friendly AI is coming to help them either. I think the overall strategic AI thinking is good. What's really missing is a bit of extra aggressiveness from the AI at the tactical level. I wouldn't wanna be wrong here but supplies are a real issue at hard level. Some castles have few slots to build and can't build the temple or enough farms which means for AI morale and supplies could be an issue in the decisional phase to launch an attack. What I am saying is that certainly the diplomatic AI is very good, I agree with you but something could be worked better on the way the AI pursues a declaration of war. Essentially, there's no point for Imagawa to declare war on me (and there's no point for me to fear the Imagawa if I attack one of their allies) if, after they declare war on me, they don't do their best to at least capture ONE of my castles. I really would like for this to change because I am also at war with Oda and they didn't send anyone against me either. So I assimilated Uesugi with marriage, made extinct the 2 clans bordering Uehara (Fukushima and Katsurao) and already captured 2 of the Nagao fortresses while being at war with Imagawa and Oda and taking no hits in my south while all of my armies are north. That's what I am saying.
Anguille		#13	[-]
Posts: 156 (11/22/09 23:45:56) Reply Quote More My Recent Posts	Could it be that Imagawa and Oda are at war with each other? Grand Ages: Rome...the next generation of city-builders
Bozo.shoguntotalwar		#14	[-]
Posts: 32 (11/23/09 01:39:06) Reply Quote More My Recent Posts	They are both at war with me and I have not seen any army coming from them to my territories. The point is very simple. While the diplomatic AI is good as Ming said, with taking decisions on which party to side on, what comes next is not appropriate. A declaration of war should be followed by an attack. The stronger the clan the stronger and more numerous the attacks should be with an outcome of peace or submission that certainly can't ever come if the AI doesn't ever attack. I think the problem could be caused by the fact that it takes more than 1 turn to reach the enemy territory... perhaps the AI moves towards them and then goes back because another threat approaches. Another possibility lies in the shortage of supplies or lack of army morale. Again, and this is very important, the AI definitely is unable to manage the way sieges really work here so it sends just one army which will immediately storm the enemy castle. If this army is defeated during the assault, the AI remains with no standing army (generals are coming back while the army is disbanded) and the no longer besieged enemy castle has plenty of supplies and reinforcements again before this AI makes another move. It's a non-winning strategy when compared to mine: I am sieging with at least 2 armies one of which is just holding the siege to prevent the besieged from receiving reinforcements. If my assault is a failure, I will keep the siege intact and then send another army to the same castle. When my third army arrives, it holds the siege while the second one storms the castle again (which has not received reinforcements so it's weaker). It is in my case only a matter of time before i succeed since every assault weakens the defenders and I keep reminding you all, I have several enemies none of which has ever made a move to force me to withdraw from a siege. Maybe, just maybe, the AI should accept peace treaties even when a clan is very unfriendly and very powerful like my Takeda are now, so that it could better focus on closer and more dangerous opponents. I am talking about both Oda and Imagawa refusing my peace offers. They don't attack but they don't make peace with me. I think if they made peace they could at least focus on closer and more dangerous enemies than me. I have 5 armies and they are all committed against Nagao in the north which again is the perfect opportunity for them both to strike me hence my requests for peace. It all feels like empty declarations of war, I'd just like for the AI to pursue them with action but I'm perfectly aware that they should declare war on me since I'm conquering unchecked the whole Japan in the North.
Anguille		#15	[-]
Posts: 157 (11/23/09 02:11:54) Reply Quote More My Recent Posts	What i was saying is that maybe they are busy fighting each other (this is happening in my campaign too) and they don't have enough troops to come after you as well. While not systematic, i've seen the AI sieging my castles. Just three questions: which difficulty level are you playing, which faction and how many castles do you control? Cheers Grand Ages: Rome...the next generation of city-builders
Bozo.shoguntotalwar		#16	[-]
Posts: 33 (11/23/09 03:41:54) Reply Quote More My Recent Posts	Hard Level, Takeda, many castles. Point is that if Imagawa declares war on me, then it should seek peace with Oda before doing so. The same goes with Oda. You don't declare war and then forget about doing it like they both did. On the other hand, I have asked several times for peace so I could focus on Nagao. They both always refused... then I realized they are too busy or weak to interfere and kept moving all armies to the North. Obviously once I'm done with Nagao it's their turn. This must change imo. It's not a matter of rash manpower, it's AI logic. The diplomacy is very good but the way they pursue a war is too light both in offense and in defense.
minglee7		#17	[-]
Posts: 1018 (11/23/09 11:07:14) Reply Quote More My Recent Posts	Thanks for the comments. We'll try to put a check for opportunity attacks when nearby enemy castle's defense is low rather than concentrate on the target enemy. That should fix some degree of the problem. However, as a side note, we particularly code the Imagawa AI to be more conservative than normal AI to reflect the historical weakness of the clan except the march for moving toward Kyoto. The thing I really like to see in the Imagawa case is to bring Hojo into their side and let Hojo do the job for them, but I guess that doesn't happen in your case. ^^ Oh, the other thing is, even if you are at war with a clan, if you don't actually engage with them in battle, the relationship between two clans will resume to neutral after awhile.
Bozo.shoguntotalwar		#18	[-]
Posts: 34 (11/24/09 01:25:57) Reply Quote More My Recent Posts	It's not a problem of specific AI coding, all the AIs act like that. They send one army, they don't pin down the castle in siege and send multiple armies till the job is done. They do send help however when they are besieged but not as a priority (and it should) and they declare war without really pursuing the offensive intent. The concentration of the target enemy (as you said) should be exactly as you proposed with your fix. Demolish the enemy clan one castle after the other, starting with the closest one exactly as you said. I think this check you'll introduce will certainly make things better but, to be honest, the AI should be taught to pin in siege AND assault at the same time so that when its army is sent back after a failure, the castle is still sieged and may fall to the second or third assault when the first army is ready to go back and strike again or when the second army sees the defenders weakened. It would help to cut the supplies and make the defenders actually desert and die if no help is sent or the siege is not broken. It would certainly help the AI provided it knows that the castle will surrender one day or another... so if a rescue party is coming the AI must go with another army to intercept it. It's the castle the key to offense, not the entire clan but that castle, to take them all one by one, one after the other, that's what I mean. 34) I have seen you can manipulate an army on the field from its home castle relieving generals and sending new ones with reinforcements. There are 2 observations to be done here. One is that if you click on this army on the map, you will see the newly sent generals still greyed out even if after they have arrived to the army, and the second, is that iif you have to disband a division to replace it with a new one coming from the home castle, the men of the disbanded division are immediately sent back to the home castle (while their general has to travel back for turns). I think this is improper. It would take a major rework to fix this issue so I propose when disbanding a division on the field to still make the men become available immediately but to make 2/3rds of these men actually lost (deserting on their way back). 35) It is really hard to handle troops on the map when woods are present. I don't know if it's possible like in STW to make the trees become "ghostlike" almost invisible but it would help a lot to be able somehow to see your division and the enemy divisions in woods. 36) Yoyta castle has a problem in the map. The hill immediately ahead of your besieging army on the path to the right is somehow impassable (it's that small "vein" of the big hill that leads to the main central gate with the tower). Troops wanting to move right in front of your flags to the right, generally get stuck there and since they can't climb this small hill, they can't find a way to bypass it. I lost a siege because of this issue. They get "lost" on the top of this very small hilly vein of the main hill and can't get down on the other side. Neither can they go back down from the left side, back towards the flags and then around this small hill on the right to reach the path going up on the main hill from the right. I repeat the castle is Yoyta. 37) If I assault the same castle twice in the same turn, shouldn't I find the gates I destroyed still broken? That's another thing that would help offensive maneuvers and sieges being extra hard sometimes. After all, If i am not mistaken, when ninjas burn down gates this effect lasts much longer than just 1 turn (and i think it shouldn't. It should last exactly one turn with the ninjas or with the army destroying the gates). 38) Again the problem with invulnerable generals. Keeping a whole division to fight a single general is already bad enough. But when this enemy general passes through 2 of your divisions (one of which is your own marshal) and hits your flag that's really a catastrophe. It appears to me some generals are really impossible to kill. Maybe it's their title, maybe it's their skill but they really slaughter many many troops single handedly. As I said, they can also take down flags and I repeat after their div is entirely lost, and they are totally surrounded they don't die and they don't run away. I have suggested making them rout faster to keep them safe but the fact they are almost invulnerable is bad. Now I am also recommending that when their division is lost, actually much sooner than that, they should really rout. This incident with that single handed general fighting his way through 2 of my divisions and cutting down a flag cost me a siege. It's very serious.
Bozo.shoguntotalwar		#19	[-]
Posts: 35 (11/24/09 05:24:39) Reply Quote More My Recent Posts	36b) There's another map problem point at the hill right in the lower left corner of the map (where your leftmost division starts). Most of the troops are able to pass and get to the main hill in the middle but a dozen remains right behind this hill and cant seem to move past it (unless you call back the division and move it towards the middle first and then only then on top of the main hill in the center). I took a screen of both spots with my mobile phone in case you need I'll post them. :-)
minglee7		#20	[-]
Posts: 1019 (11/24/09 06:27:46) Reply Quote More My Recent Posts	36) I'll check the map. Maybe we missed some nodes in the corner. Thanks for pointing out, and all the other comments. ^^

<< Previous Topic Next Topic >>

Yuku free message boards

Tip(list) of the day! :)

Tip(list) of the day! :)