- This topic has 333 replies, 11 voices, and was last updated 12 years, 2 months ago by
.
-
Posts
-
Go with more than three layers. Yes, many of them are unused, but when we have tiles that bleed out, it'll be easier to give each terrain that bleeds out its own layer.
The flare format requires 3 layers, but it will export all of the layers into it… so.. no limit on layer count afaik.
busy night
* optimized several things by thousands of percents
** mostly graphical and a few other small things
** makes it far less painful on lower end systems (like this netbook heh)
** not necessarily going to be thrown away later, so not a waste of time
* initial move to a proper application / game loop
** no longer sitting in a buried loop somewhere
** things are much more 'background processing' friendly
** beginnings of input abstraction
What must happen next:
Coming up with a cleaner way of handling overall game state, variables, etc
Right now, several things require a lot of string searching / matching across several arrays of structs and all that jazz. While that was fine for proof of concept, it really needs to go away and cleaner state variables implemented etc.
… rolling along / progressing just fine….
It's progressing slowly.
What's kinda fun is that the code base cleans up a bit while I progress in this part of the endeavor :).
The project is also becoming much more usable (with proper loop and state system), which is always nice. The optimizations done last week have also made it very "playable" (horrible word, there's no game yet lol) on my netbook, so platform horsepower shouldn't be a concern.
I still need to clean up how I'm gonna handle game variables. I keep switching back and forth from just working from a master list vs building sub lists for each action. Always the thought of each province has it's own global sub list that doesn't get rebuilt constantly except when someone joins / leaves a province is probably a valid compromise (decided: [X] hehe).
So next few days:
* provincial sublist (easy)
* move everything to working off of the provincial sublist (tedious but easy)
* "button" code for making gui stuff easier / better
* implementing the provincial screen gui
* cleaning up the general-select-for-actions screen gui
After that:
* provincial screen actions go through the RPN stuff and causes things to happen
* load the flare map format
* convert existing battlescape map stuffs to flare format if someone hasn't already made the maps
** easy enough since i've already got the battlescape maps in a different format
From there… I dunno. It'd prolly be time to let a few people look at it, since it should become interactive enough for anyone who wants to contribute.
Down the line in possibly this order:
* battlescape gui interaction
* beginnings of actual game flow (ie: only being able to control one ruler's territories at a time, 'end of turn', etc).
* 'events' system thing (yuan shu / sun ce independence, random promotion, spoils of war, seasonal things, etc)
Ugh.
There are days when I want to work on this, yet I *know* I won't be able to make any meaningful progress; this is one of them, I'm afraid. So all I can suggest is the idea of officer blocks (sub-forces). From my experience, there were four main officer blocks historically:
* Liu Bei (and Guan Yu, Zhang Fei, etc.)
* Sun Ce (and Zhou Yu, Huang Gai, etc.)
* Lu Bu (and Zhang Liao, Gao Shun, etc.)
* Emperor Xian (and the entire Han court)
Since we seem to be transitioning from "RoTK2" to "other Three Kingdoms game", it may make sense to make it harder for to split them up or remove their soldiers. Similarly, if you recruit Liu Bei, it makes sense for Guan Yu, Zhang Fei, etc. to follow him — after all, even though Liu Bei currently serves Tao Qian, they're loyal not to Tao Qian but to Liu Bei.
So how do you peel officers off of an officer block? I see three ways:
(1) You can always recruit them, but it's tricky. In 194, for example, we're not tracking Zhou Yu's loyalty to Yuan Shu, but his loyalty to Sun Ce. So this is rarely an option, but it worked for Xu Shu.
(2) If you defeat them in combat, you can snag them that way. Cao Cao had no problem temporarily getting Guan Yu!
(3) A ruler may be able to convince officers to leave an officer block serving them. Of course, if the block leader finds out, their loyalty is likely to drop…
So what other effects would an officer block have? I can see the following happen:
* The block leader is more likely to declare independence (Liu Bei, Sun Ce, and Lu Bu all did that!)
* Officers in the block are usually less likely to be recruited than if they were your direct subordinates.
* Block leaders may recruit enemy officers, even if they're not in force leader's province.
So think of them as a double-edged sword, and remember that the vanilla version had a bunch of forces that really should be officer blocks:
Scenario 1:
Sun Jian (Yuan Shu)
Scenario 2:
Sun Ce (Yuan Shu)
Scenario 3:
Liu Bei (Liu Biao)
Scenario 4:
Liu Bei (Liu Qi)
JIN XUAN, CONQUEROR OF WORLDS (Liu Qi)
Han Xuan (Liu Qi)
Liu Du (Liu Qi)
Zhao Fan (Liu Qi)
And yeah, we'd need slightly different rules for His Emperorness…
Yeah, I'm trying to get it to the point where there can be meaningful contributions by other people…. otherwise it's more or less 'only my project' which is not the intention.
Cliques is an interesting addition to the loyalty / ruler hierarchy.
Oh… clique might be a mostly American term. A loose definition, lets pretend you're in high school. You have the jocks, the cheer leaders, the geeks, etc. In general, each of those could be considered a 'major clique'. Within, say, the cheer leaders… you'll have several smaller groups. They usually don't like each other or something. Each of that smaller group of 'friends' would be your typical clique.
Anyway… the interaction of individuals, cliques, rulers, etc could be a nasty nightmare to try to figure out. Formulas would be kinda neat.
I do like the idea.
Lets say Liu Bei and his entourage are serving under Yuan Shao. Cao Cao manages to recruit Liu Bei away from Yuan Shao.
A) do close supporters of Liu Bei (Yu/Fei) automagically follow Liu Bei ? Or maybe if their loyalty is under XX to Yuan Shao ? (if loyalty-to-clique-leader * partiality-to-clique-leader > loyality-to-ruler, convert as well?)
B) if the members of the Liu Bei clique don't follow him over to Cao Cao, does their loyalty to Yuan Shao drop ?
C) lotsa questions like that
Instead of Cao Cao managing to recruit Liu Bei, lets say he manages to recruit Guan Ping. How does this work out ? Does Liu Bei's clique dislike Yuan Shao now and suffer a drop in loyalty ?
Another scenario: while serving Yuan Shao, Lu Bu kills that traitorous bastard Liu Bei. Now what ?
A) Does the clique get a new 'leader' or is it left void ?
B) Does the clique loyalty to their ruler decline ?
So, lotsa questions here… and it drastically changes traditional game play (i can see two modes of play coming up… 'free' vs 'enhanced' hehe).
Some clique-implementation questions…
In an effort to keep it somewhat dynamic or less prone to falling apart as a game progresses…
A) does each general keep a familiarity list with every other general ?
B) serving within the same province makes it go up ?
C) doing the same task together makes it go up ?
D) combat on same side makes it go up ?
E) what causes it to go down ?
What about people that, say, couldn't stand each other ? Zhou Yu and Zhuge Liang ? I doubt serving together would make things better ? If there are things that causes familiarity to go down, could you eventually get Guan Yu to hate Liu Bei ?
Could general stats help this (charisma stat becoming even more godly ?) ??
I see *alot* of questions about this stuff… but it's quite intriguing!
Oh and "other people working on it" …
Depending on how 'mod-able' it is, it might require actually working within the code base as well. I'm trying to keep as much as possible as /NOT/ being hard coded though.
Compiling the code would probably be impossible unless you picked up a license for the compiler (i do make use of some 'professional only' functionality).
Without an events system, the primary "working on it" currently would be…
Things that are useful no matter what:
* filling out more information per the ~450 generals (date of birth, personality, expected lifetime, etc…. boring tedious work heh)
* filling out more start-of-scenario data (i believe I only have the data for the first scenario put in)
* flushing out more generals used within the first scenario (i've got most of the rotk2 generals, but there are more that were active during that time…)
* more generals such as the bandits and stuff ?
* filling out RPN functions for things (even if it's just regular math and i make it rpn for ya)
Ideas to toy with:
* possibly figuring out how to construct this loyalty / clique / blocking system :P
Stuff that is meh:
* graphical skinning possibilities
* fiddling with battlescape maps maybe, i dunno
After the events system gets put into place … there will be a little bit more mod-able stuff … but… huge drastic stuffs… well… there's a lot that would basically be better off as something like "the game is written entirely in LUA, the graphical front end is generic" heh. (oh, and note: i'm not gonna make it lua-friendly).
Oh, and AI flow charts / pseudo code is handy — but given that we've got the core game functionality in so much flux… that's a complete bitch to try to even guess at implementing now heh.
RULE OF THUMB: The block leader is essentially a mini-ruler that serves the actual ruler. Things that would annoy another force would usually annoy the block leader, but then again, you can always reward the block leader.
Also, I somehow left off the most important officer block of all… the Sima Yi/etc. block.
-=-=-
Presumably, the block (or clique, if you prefer) is loyal to Liu Bei, not to Yuan Shao. So if Liu Bei is recruited, everyone in the block (or almost everyone) would follow him.
Any officers who don't follow Liu Bei would be reassigned loyalty to Yuan Shao, which would presumably be higher than loyalty to Liu Bei.
If Cao Cao recruits Guan Ping from the block, Liu Bei and Yuan Shao shouldn't be having a falling-out… unless Yuan Shao ALSO has a Cao Cao block. ;)
If Liu Bei gets killed and the block has a strong heir, such as Guan Yu (speaking of which, we need to revamp the inheritance system), the block should be completely or almost completely intact… although it may cause tension between the block leader and the ruler. But if there's no clear heir (such as the Lu Bu block), the block could easily "lose" (to Lu Bu's boss) many members or even stop existing.
Each officer has a compatibility rating; familiarity lists probably aren't needed.
Serving in the same province is probably the norm; scattering the block across your various provinces too much may cause tension between the block leader and the ruler.
Combat shouldn't affect block loyalty. Of course, if Lu Bu is bribed in combat, a large portion of the block may also switch sides!
For a block to weaken or fall apart, the only things I can see affecting it are outside pressure and the death/capture of the block leader.
For a block to form, presumably you need someone with a desire to rule (the block leader, if the RNG wills it) and at least two officers who have (1) low loyalty and (2) better compatibility with the future block leader than the ruler.
If an officer has low loyalty and better compatibility with the force leader (or the rival block leader, if they're in a different block), the block leader may try and recruit them into their block.
Finally, as a block gets too strong, I can see then doing these things five things (in increasing block strength):
(1) Humbly requesting that the ruler not do (anti-block action)
(2) Strongly suggestion that the ruler not do (anti-block action)
(3) Outright vetoing (anti-block action) — which will completely prevent (anti-block action) and/or cause trouble if you [try to] do it anyway.
(4) Claiming independence based on their strength (unless the block leader has high loyalty)
(5a) Outright deposing the ruler based on their strength (unless the block leader has high loyalty)
(5b) Becoming the heir, even though someone else should rightfully be the heir.
No sensible ruler will let the block get strong enough for 5, 4, or even 3, but obviously some rulers DID let them happen.
Also, I somehow left off the most important officer block of all… the Sima Yi/etc. block.
As for stats, they probably won't DIRECTLY affect things, but if you have only 15 charisma, expect to have officers with low enough loyalty that the block leader can start handing out invitations…
(and psst: Zhuge Liang and Zhou Yu didn't actually have an issue; that was novel-only!)
EDIT: To clarify, here are the block level descriptions and examples:
(1) The block is so weak you'd barely notice if they revolted (You have 1000 soldiers, 50 of them loyal to Liu Bei)
(2) It would annoy you if the block revolted, but the outcome is never in doubt (You have 1000 soldiers, 150 of them loyal to Liu Bei)
(3) You can crush the block if they revolt, but it requires a major operation (You have 1000 soldiers, 300 of them loyal to Liu Bei)
(4) If the block revolts, they're nearly as strong as you are! Tread carefully… (You have 1000 soldiers, 500 of them loyal to Liu Bei)
(5) Sorry, Liu Pan; the block's now stronger than you! (You have 1000 soldiers, 700 of them loyal to Liu Bei)
The numbers aren't set in stone, but…
some of this sounds down right dangerous to fully implement. it'd be moving away from the player being an absolute dictator … and attempting to loosely manage a simulation (akin to those games where, say, you're a football coach or owner but never actually "play" the soccer game itself…. a simulation engine determines what happens).
i can see varying degrees / incarnations of this mechanic taking a lot of play testing to see if it's liked / hated or if it can be broken, etc.
stuff like, is it possible for Cao Cao to convert Guan Yu to be within his clique rather than in Liu Bei's ? showering with gifts, etc.
Quote:As for stats, they probably won't DIRECTLY affect things, but if you have only 15 charisma, expect to have officers with low enough loyalty that the block leader can start handing out invitations…Poor Jin Xuan…. cough.
Anyway… I love a lot of the concepts here… definitely gonna have to debate this a lot more once we get into a playable game (no, i'm not trying to close the subject, it just gets really complicated and very theoretical without being able to test it).
As far another thing to possibly do:
* a list of 'surnames' and 'first' names for use with random general promotion after combat or similar
Lists cooperating.
Basic button stuff put in.
Did some digging – apparently language supports function pointers. That's handy.
The officer block's only dangerous if the player lets it get too strong. Yeah, some officers will likely start up some blocks, but only a careless player will let them get too big!
Also Jin Xuan has 37 charisma; it's Lu Bu who only has 15 charisma!
And yes, the player loses a little control, but keep in mind that having an officer block can arguably be beneficial; those officers Liu Bei recruits form outside the force are additions to your army, and as long as you keep Liu Bei loyal he'll keep his block loyal!
-=-=-
Surname suggestions should also include entries for:
(1) The family name of the force leader
(2) The family name the province governor
(3) The family name of a randomly chosen officer in the province (reject if no stats >80)
That said, the family surnames as of the Liu Song dynasty (420-479) are at http://en.wikipedia.org/wiki/Hundred_Family_Surnames and should be fairly similar in the han dynasty.
They should also be usable for given names; double-character names are much less common for given names, but they do exist (for example, Xi Zhicai and both of Huangfu Song's sons, Huangfu Jianshuo & Huangfu Shuxian). Technically they were banned, but seeing as the banner was the infamous Wang Mang (best known for interrupting the han dynasty 9-23 AD), the elected officials would look the other way.
For the list of all given names in RoTK2, see http://pastebin.com/66YW7jTf — typos have been corrected, but a few non-RoTK2 names (such as Zhou) are missing. Non-chinese names are also missing.
-=-=-
Now, all we need to do is to peel some people off of John Boehner's block… :Þ
You know, if we go ahead with officer blocks, the ruler should be able to give sub-block rewards; presumably that would involve giving Mi Zhu a horse and a little note thanking him for assisting Liu Bei. Doing so would make him more loyal to Liu Bei AND make Liu Bei should also be a bit more loyal to you. Remember, officer blocks aren't necessarily bad!
So…. If I'm … say… Dong Zhou. I have my own block prolly?
If I take the game to 220ad and beyond, there won't be much of my block by comparison.
I'm going to have mostly a conglomerate of opposing blocks.
How much control do I really have?
Is this then the loyalty structure is I'm to keep the block leaders loyal to me so my influence is felt?
We'd have to definitely show block affiliations for each general. As well as possibly block 'leader'?
'Spy' options grow to some degree as well.
Basically I'm fearful this will turn the player into the useless emporer in some games that makes requests of subordinates.
- You must be logged in to reply to this topic.