Fixing revolutions.

Noriad2 said:

Incorrect. I've had a barb city turn into a regular city before anyone discovered writing (regular V36, minor civs option on). I've checked a save game to confirm. Also in the BUG options (revDCM tab) there is a minimum city size for barbarian civ, which is set to 3 but can be changed.

Click to expand...

I thought MW had mentioned Start as Minor in a Previous post in another thread. And I was pretty sure city size was involved. Also MW is now using the Crime Patch SVN version 9102. Still thanks for the "correction" and added info. You do seem to like correcting me.

JosEPh

There is a page in the BUG options which sets up what is required for a barbarian city to turn into a nation. You can vary all sorts of things - the big two, iirc, are the population of the city and how many nations have seen the city.

JosEPh_II said:

No we did not remove Rev. I would've been ecstatic If we had!

Click to expand...

I believe by ecstatic you mean a good feeling (sorry for my subpar english understanding), if so I must ask you to not think like that. The game already has the option to turn it on/off, so whoever wants to use it is free to use it, and whoever doesn't want to use it is also free to avoid it.

If you guys were to remove revolution altogether I would be really disappointed. I understand it's a messy feature by the way it was implemented, but to improve it would be far better then to remove it. Revolution is an essential part of a game like CIV IMO, I can't imagine playing a game without it as I have been playing with it for at least 6 years already.

Like what happened with me and RoM-AND. It was a good mod and all, but then Afforess came with the brilliant idea of removing Trade Routes altogether because they were costy to the game processing (which is true and should be adressed anyway) and also because they were hard to understand and completely independent from the player's choices (which was a really lame reason that shows the lack of knowledge on Trade Routes and how to manipulate them).

Spirictum said:

I believe by ecstatic you mean a good feeling (sorry for my subpar english understanding), if so I must ask you to not think like that. The game already has the option to turn it on/off, so whoever wants to use it is free to use it, and whoever doesn't want to use it is also free to avoid it.

If you guys were to remove revolution altogether I would be really disappointed. I understand it's a messy feature by the way it was implemented, but to improve it would be far better then to remove it. Revolution is an essential part of a game like CIV IMO, I can't imagine playing a game without it as I have been playing with it for at least 6 years already.

Like what happened with me and RoM-AND. It was a good mod and all, but then Afforess came with the brilliant idea of removing Trade Routes altogether because they were costy to the game processing (which is true and should be adressed anyway) and also because they were hard to understand and completely independent from the player's choices (which was a really lame reason that shows the lack of knowledge on Trade Routes and how to manipulate them).

Click to expand...

This version of Rev is not the same as LoR, as I understand it. This version was put in when Zappara was building RoM 2.91 and Afforess was still part of the Rev Team and also making AND as a Modmod for RoM. It has Bugs and problems. Some were looked at by koshling when he was still modding C2C. But he really did not want to deal with what he called "spaghetti coding".

I personally have not liked Rev since I was 1st introduced to it back in Rise of Mankind 1.0 when kalimachus(sp?) introduced it as a Modmod for RoM. I've been an opponent of it since before StrategyOnly bowed to the "crowd" and added it into C2C by using RoM 2.91 or 2.92 as C2C's base. His NWA never had Rev in it as he also "disliked" Rev as Much as I did back then.

And I would think Spirictum that after using it for 6 years you would realize how Bad Rev is for the AI. And that this is it's major Flaw. It will take a flourishing AI empire and destroy it. The AI's understanding of Rev is very limited If it exists at all in the version here in C2C.

I do understand phungus tried to make a better AI for Rev in LoR. How well he succeeded you will have to tell me. I never played LoR nor did I ever want to. About 8 years ago phungus and I exchanged "views" about Rev and it's real purpose for the AI and Player. Lets just say I was Not on his favorite poster list. Especially when I pointed out it's main Flaw in it's design. Did phungus ever overcome that flawed design? IDK and I never cared to find out.

But the Propponents of Rev almost made Rev mandatory here in C2C with their insistent clamor for it. Even koshling at 1st wanted it core. Their goal was You Had to play with Rev always On in C2C. I fought Tooth and Nail for the Option we have today. I made past modders of this Mod upset with me because of my arguments not to do so. And I was only a longtime tester back then and not part of the C2C team. Hydro even nicknamed me Mr. NoNo. But I persisted and I think wha I accomplished was for the betterment of C2C overall. Keeping Rev an Option. Personally, C2C doesn't "need" Rev to be a Great Mod. It's a Great Mod w/o Rev. And I intend to help keep it that way as long as I'm here.

Hopefully you now have a better understanding of my Stance over Rev. I won't take it from you, remove the Option. But if you (all those that play it) ever come to the realization it's Flawed and Broken, I'd be More than Happy to assist in it's removal. C2C wouldn't even lose a step with it gone, imhpo.

JosEPh

As I have said again and again without Revolutions you would not be able to have had as many modders working on C2C as we did. Without it you are limited to one unit infos file, one building infos file and so on. basically it would have been easier for each modder to make their own mod than work together on C2C.

Dancing Hoskuld said:

As I have said again and again without Revolutions you would not be able to have had as many modders working on C2C as we did. Without it you are limited to one unit infos file, one building infos file and so on. basically it would have been easier for each modder to make their own mod than work together on C2C.

Click to expand...

And I still don't understand that. If the coding for the Rev in C2C is subpar and no one wants to fix it then how or why would there only be "one unit infos file and so on"?

How does Rev allow more? And that process that Rev uses can not be duplicated without all the other Rev baggage?

And it wasn't the "not wanting to work on Rev" that caused modders like koshling and AIAndy, EldrinFall, ls612, and others to leave. And surely it wasn't Rev that attracted those same modders to C2C. If it had would they not have treied to improve it? Instead they shunned working on it.

JosEPh

JosEPh_II said:

And I still don't understand that. If the coding for the Rev in C2C is subpar and no one wants to fix it then how or why would there only be "one unit infos file and so on"?

How does Rev allow more? And that process that Rev uses can not be duplicated without all the other Rev baggage?

And it wasn't the "not wanting to work on Rev" that caused modders like koshling and AIAndy, EldrinFall, ls612, and others to leave. And surely it wasn't Rev that attracted those same modders to C2C. If it had would they not have treied to improve it? Instead they shunned working on it.

JosEPh

Click to expand...

We did not have a dll programmer. Therefore we needed a dll from somewhere which allowed such modding. At the time the best mods for modders were WoC and BUG. Basically, WoC allows multiple XML files and BUG allows multiple Python files.

Revolutions had both in it. Therefore using the Revolutions dll would give us WoC and BUG also. It also gave us BULL and many, many other things.

Unfortunately having so many things in it made it big. The actual Revolutions part of the code is mostly separate form the main lines of code so it can be ignored when working on other things. It is not actually "spaghetti" code but it is not as well structured as more modern coding traditions suggest it should be. (I am old enough to have worked on maintaining "spaghetti".)

JosEPh_II said:

And I would think Spirictum that after using it for 6 years you would realize how Bad Rev is for the AI. And that this is it's major Flaw. It will take a flourishing AI empire and destroy it. The AI's understanding of Rev is very limited If it exists at all in the version here in C2C.

I do understand phungus tried to make a better AI for Rev in LoR. How well he succeeded you will have to tell me. I never played LoR nor did I ever want to. About 8 years ago phungus and I exchanged "views" about Rev and it's real purpose for the AI and Player. Lets just say I was Not on his favorite poster list. Especially when I pointed out it's main Flaw in it's design. Did phungus ever overcome that flawed design? IDK and I never cared to find out.

Click to expand...

Quite the contrary, Rev was always a nuisance to me and a bucket of XP for the AI in LoR as far as I remember. My gameplay improved a lot only after I started playing LoR. When I started playing it I believe I was still playing on Noble. Maybe on Noble the Revolution was more a benefit then a nuisance to me, but I don't remember. But when I stepped up to Emperor, Immortal and then Deity, revolution was always a problem to me and an advantage to the AI, I can't remember differently. 100+ cities AI on Huge completely stable, or with some minor locations rebelling that were always loaded with AI's troops to quell it ASAP. OTOH, I had a game with a friend where both of us had 100+ cities, and the only reason we could keep our stability was because of continuous conquest of cities throughout the game.

If Phungus had made it better I'm not sure, but it was certainly scaled to difficulty. With all the bonuses the AI has on deity, revolution was a benefit to it in LoR. Also, I don't see a flourishing empire crumbling to revolutions something wrong, especially if this flourishing means overstretching or overpopulating. What I think is wrong is every flourishing empire crumbling to revolutions. That certainly did not happen on LoR (at least on the higher difficulties).

And I'll tell you what was the real Major Flaw of the AI back on LoR (as I hope this was already adressed since then): It's inability to deal with the "Shield Glitch". Back then playing on Deity could mean several restarts because of bad starting locations, and by that I mean places easy to be attacked, as Deity AI on LoR was ridiculously strong, especially barbs turned into minors. But then we discovered the Shield Glitch. If you could make a chain of units around all your cities by putting at least a single unit in each tile of the chain, the AI would stop targetting you with its main attack force. After some tests and gameplays we realized that when you did that the AI started to behave as if you weren't reachable by land (as closing the chain in a turn and passing it had shown us a great naval mobilization of the AI in the following turn, a behavior the AI only takes if it thinks it can't reach you by land, or at least it was like that in the past, I remember RFCEurope had this same issue too).

You can't imagine the good laughs we had when we fooled the AI so badly it was simply stuck going forward and backward because in one turn we opened the chain in its southern part, so the AI's main army would walk south, and in the next turn we closed on the southern part and opened in the northern part, then the AI would move north. It was hilarious. We left it doing this for ages until we developed nukes to kill its bizarre army.

JosEPh_II said:

Hopefully you now have a better understanding of my Stance over Rev. I won't take it from you, remove the Option. But if you (all those that play it) ever come to the realization it's Flawed and Broken, I'd be More than Happy to assist in it's removal. C2C wouldn't even lose a step with it gone, imhpo.

JosEPh

Click to expand...

If you're ok with not removing it, then its ok to me, I just don't get it why you'd prefer to remove it then to fix it. I'm not taking into consideration the effort to fix it, and that may be reason enough to you, but if it was doable and someone could do it for C2C, wouldn't you think it's better then removing it? I mean revolution is a great concept that takes a really important aspect of reality and puts into the game.

As far as I know we did include many of the fixes Phungus made. When we chose to use it it was in active development but RL hit all the modders at the same time.

If you're ok with not removing it, then its ok to me, I just don't get it why you'd prefer to remove it then to fix it. I'm not taking into consideration the effort to fix it, and that may be reason enough to you, but if it was doable and someone could do it for C2C, wouldn't you think it's better then removing it? I mean revolution is a great concept that takes a really important aspect of reality and puts into the game.

Click to expand...

Purely subjective playstyle preference Spirictum. Ones that I personally don't agree with.

As long as proponents don't push to make it mandatory to main C2C play (as was done in the past) I don't see the team removing it. Too much work.

As for "fixing/updating it" that would take a Motivated C++ programmer. And the only one we really have now is overwhelmed as it is. And that is why things have slowed down considerably. And will stay that way until another 1-3, Dedicated to the Mod, C++ programmers join the team to help Clear the backlog and Help fix the started projects.

This may be the slowest C2C has moved/advanced since it's inception.

JosEPh

I may try to help... I am learning programmer
Anyway I believe it is huge project such a dll.
I hope I will not make code worse to understand (sometimes I made code hard to read even for myself such as overusing ?: operator), but I am sure I can fix Something.

Just a problem for me is: even if I have time "free" I have problem to have time for one project or another, but hopefully I can help.

The biggest hindrance is that the Revolution code is part in the dll and part in Python, and as far as we can tell you need to understand both parts.

Spirictum said:

Like what happened with me and RoM-AND. It was a good mod and all, but then Afforess came with the brilliant idea of removing Trade Routes altogether because they were costy to the game processing (which is true and should be adressed anyway) and also because they were hard to understand and completely independent from the player's choices (which was a really lame reason that shows the lack of knowledge on Trade Routes and how to manipulate them).

Click to expand...

I thought that was one of his best ideas, personally, getting rid of both a resource hog and something over which you have little to no control, in favour of something much easier to identify.

Revolutions always seems to harm the AI more than me. But I always have to be aware of it, especially by sending more cops to my most populous cities. Crime is THE killer when it comes to revolutions.

But that's great! It opens up the whole thing about sending criminals into enemy cities and getting them to shatter under extremely high revolutionary stress. And recently (say, since the crime property was introduced, and now), the AI is better at filling its cities with cops, even in response to ME filling them with crims. A coherent and stable empire, say some nation on its own modest continent, will adjust far better to me sending mob cars into it than a bloated warmonger always having its newly conquered cities rebel without my intervention.

So I've found that it works well. Unless it's what is causing my current game to lag out, in which case it's bad. (Except I bet it's my rival having a 500+ stack of medieval siege engines with nowhere to go that's causing trouble. The AI, it REALLY sucks at killing Hitler, even when he has just 1 city left.)

And recently (say, since the crime property was introduced, and now), the AI is better at filling its cities with cops, even in response to ME filling them with crims. A coherent and stable empire, say some nation on its own modest continent, will adjust far better to me sending mob cars into it than a bloated warmonger always having its newly conquered cities rebel without my intervention.

Click to expand...

Awesome... sounds like my adjustments are working as intended finally!

Will be nice once I've got the AI programmed to use criminals against other nations properly as well. But it's very good to see appropriate law enforcement protocols are working. Hooray for small victories!

Arakhor said:

I thought that was one of his best ideas, personally, getting rid of both a resource hog and something over which you have little to no control, in favour of something much easier to identify.

Click to expand...

Fortunately people agree with him, beause if not he would indeed be doing something wrong. But IMO Trade Routes are a great part of the game, one that you can actually control a lot about although it's really costy to the game processing. Exchanging that to a connectivity commerce bonus seemed to me a downgrade to that feature which removed a great concept from the game, and one I was a lot interested on because it was really hard to manipulate and my discoveries about it are recent compared to many other aspects of the game.

To me he exchanged a hard to manipulate feature (which for me is a good thing, and it is possible to manipulate indeed), which could make big differences to different nations depending on the way it is worked out, with something that hardly replaces it in concept and is far easier to manipulate to one's own nation's favor.

I've re-lifed this thread to hopefully continue the discussion on revolutions that began in the Rethinking Crime thread.

In particular, I would like to ask for input from those who use rev:
- What revolutions have you observed in recent games? What seemed to cause them?

- To what extent should instabillity accumulate from different factors? Should extended war be worse than having a very large empire? Is it good gameplay that large separatist uprisings can occur in very small nations?

- To what extent should crime and revolutions interact (particularly considering that / already increases instabillity)?

I have my own opinions on all these factors, of course. For example, I absolutely hate the instabillity increase that occurs when a nation loses a city (this is the domino effect responsible for large AI empires suddenly imploding, particularly when combined with stagnancy mechanics and war weariness). But I'm interested in seeing what other players think.

Noriad2 said:

Losing a single city is not necessarily a bad thing in the long run if it is your worst city. There are penalties connected to the size of your empire. Upkeep (based on number of cities and distance from capital) is the most obvious, the need to defend it which needs units that also require upkeep, even happiness if you have the penalties from civics due to number of cities. Losing your worst city makes economic room for a new city that is potentially more valuable. Economic room of course slowly increases as the game progresses, as a civ gets new techs and develops its core cities.

Again, a city doesn't revolt away because everything is fine with it. It is probably completely overrun with criminals and probably costs the owning civ more than it benefits. A well-governed civ won't lose cities due to REV.

Click to expand...

I think the issue here is really that the AI doesn't manage crime very well (which hopefully is now fixed) and possibly that maintenance doesn't increase at a rapid enough pace commensurate to gold generation in C2C (although that can vary a lot depending on game speed). Perhaps more important is Revolutionary Sentiment and the snowballing effect of losing cities - this is what used to eat empires after they lost a few border cities.

I haven't observed many empire-wide collapses recently, which I think is due to an enormous increase in the stability granted by military units at some point in the last year-and-a-bit. I am not sure who did this, but the result was essentially to turn the AI's relentless over-accumulation of military units to some kind of advantage.

cammcken said:

Not all stability effects come from explicit sources like buildings, civics, and garissons. If I understood the hints correctly, "growth" and "stagnancy" also affect it. Whether that means technological growth, city growth, or imperical growth, I don't know. It seems like the creators were trying to give it a "natural" feel, instead of just fulfilling checklists. If so, Rev needs to be rebalanced from its BtS expectations of reasonable growth to an assessment of growth more realistic to C2C.
But I never mastered Rev either, so I don't know for sure. It must have been wildcards like these that kept throwing me off.

Click to expand...

If I recall correctly (and I may not, because Rev uses a complicated mix of Python and C++ that is damn hard to keep track of), stagnancy comes in two forms - an economic version (involving gold production rate) and an overall civ score version (which I think is the research-related one?). Modifying this would require some substantial searching.

One of the biggest problems with revolutions is that a lot of the underlying processes are never made explicit, so you end up having to dig through the code to figure out what is going on.

Squidimp said:

and possibly that maintenance doesn't increase at a rapid enough pace commensurate to gold generation in C2C (although that can vary a lot depending on game speed)

Click to expand...

This was being addressed and then some Eternity Players and Mod scenario makers pitched a fit over not enough gold (imagine that!). Plus it was then decided to revamp How some of the modifier chains worked in the Game speeds. These combined with adjustments to inflation, overall maint. costs reduced, and tilted to accomodate longer GS, pushed the balance currently back to reducing Maint costs. The effect of Game Speed maint costs can be seen in the financial screen under maint. There is a line that will show up that tells how much maint is increased by Game Speed. This area is being re-evaluated again and over the coming weeks adjustments will be made (again).

Squidimp said:

One of the biggest problems with revolutions is that a lot of the underlying processes are never made explicit, so you end up having to dig through the code to figure out what is going on.

Click to expand...

This is very true. And every Modder here has had their "own" area of interest and no one selected Rev. Koshling was the last one to truly make any changes until T-btrd's recent Rev index adjustments.

Hydromancerx said:

I like the idea of giving just the AI a big revolution stability bonus while the players don't get it. That way players can experience the fun parts of revolutions, and the AI will not fall apart because of revolutions

Click to expand...

I would play it. But so far that means only another thing to consider while managing cities. Part of the fun of revolutions is also, um, the revolutions. I want to see new civilizations break away. There should be at least some mechanism for introducing new civilizations. For those proponents of Barbarian Civs, revolutions is like BarbCivs after all the barbarians are gone. If Rev ultimately dies, I would like to see some other way for cities/land to go back to the barbarians.

I had a cool, off-topic idea on that note:


What should be avoided is an AI that tries to manage Stability but doesn't manage it effectively. Worst case is when they drag their economy down while trying. (Such as when they mass-produced expensive LE units before TB's crime tweaks.) However, that doesn't need to mean AI that are safe from revolutions. It could be that the AI recognize the risk-reward ratio of fast expansion.* Why is it a bad thing that empires fall apart? (That is, after all, what they tend to do irl...) There are two ways I can look at this:

1. In a multiplayer game, I'm okay with the AI being weaker. I usually do MP with my family: three humans and the rest AI. In this setup, weak AI are like puppet states tugged between the influences of the big superpowers. There's no loss to game play and there's the benefit of Rev's immersion. Isn't this what happens to Barbarian Civs anyway? Late to the game, they'll always be somewhat followers among the big players. If you're okay with having some small civs, why not a few more?

2. It's okay if empires fall apart, provided that whatever new Civ which rises up to fill that power vacuum is strong and competitive enough to take its place. When I was playing a "vanilla" Rev game, I once witnessed a Civ going on a warmongering spree, carving out a huge piece in the center of the Pangaea (It was Shaka, to no surprise). A rebellion arose, and the new Civ (Egypt, iirc) managed to defeat and conquer all of what was once the Zulus, staying strong. My point is, crumbling empires doesn't need to cause weak AI. So long as there is a balance: a) Some sprinters/rapid expansionists, who measure the potential reward against the potential risk and do not reach for low-reward goals. There is a positive snowball effect from conquering, irrc, which they could benefit from. b) Some cautious growers, prioritizing stability. c) And some opportunists, to fill the middle of this spectrum, who can step in when the big boys fall. With the new game mechanics favoring the bottom of the scoreboard (already being considered) like TD and WFL, and maybe some extra boosts when a faction gets out of "rebel" status into "independent nation" status, might it be possible?

The discussion seems to be assuming that all AI should be the conservative and secure type. What's wrong with letting some crash and burn, a la Rhye's and Fall?

*Please realize, I say this all without any knowledge of the process and difficulty behind making it. I'm just suggesting final-product visions, take it all in consideration.