RevolutionDCM: Core Issues

[phungus420]

Hey glider & jdog. I've been thinking alot on it, and figure I should share my perspective here. This is some constructive criticism based off my experiences with RevDCM. Now you should realize before hand that RevDCM is a great mod, obviously. Keep in mind I use it as the base mod for my mod because it is so good. However as I see it there are 3 main problems with RevoltutionDCM, and I really want to highlight them here as they are very obvious and easily isolated.

1st is that it's not multiplayer compatible. My understanding is that jdog is working on this, but that it's a complex issue and requires a re writing of alot of the Revolutions code, so this wol't be out for a while, and may actually be scrapped altogether. I hope not, it would be a shame to keep RevDCM relegated to SP only. Could we possibly get an update on this?

Second is the scoreboard. On large maps the scoreboard gets very messy, and can simply run off the screen. This looks pretty terrible aesthetically, but is also an issue with general UI functionality. Please consider incorporating a scrolling scoreboard to adress this, it may appear minor, but for those that prefer large maps, that can spawn alot of civs, this is actually game breaking.

Third is a lack of transparency with revolutions. I realize this is a design philosophy difference between jdog and myself, but I still see no reason to hide information from the player. It's simply frustrating to have revolution problems going on with no known cause for it (and thus no known way to fix it). All this does is cause irritation, and I see no gameplay benefits here. This would be akin to hiding the + and - happiness and health benifitis/effects of some resources & buildings, while showing others because the devs thought that it should be common sense the effects of said modifiers. Think about how such a gameplay change would effect the game for the user, this is exactly how the current implementation of stealth Rev effects hinder the game in RevDCM. Any reason why jdog has decided to hide information from the player regarding Rev effects? Any chance future versions of RevDCM will go more toward a transparent aproach to information being available to the player?

These are the main problems with RevDCM, and I really wanted to bring them out in the open. Mainly because I think these could all be adressed, and these issues I can easily put my finger on point out, they are very tangible. I was wondering if anything was being done or any plans were being made to adress these issues, because I think if you fix these any other change/improvement to RevDCM are just gravy. These are your core issues that irritate users when they play, or simply make the game unplayable for one reason or another and I wanted to focus on them for a minute and at least get some feedback from jdog and glider and any other users of RevDCM on them.

 

[glider1]

Yeah thanks Phungus. Points taken except for the last. I remember a moment when Jdog asked a member of the civ community in Germany if having exchanged to them the Revolutions code, they could exchange with us the beautiful scoreboard scrolling code they had written. It seemed not to have happened.

On the last point, you do realise that many Revolutions variables are available as numbers in the Revolution Watch Advisor under the "cheats" tab? If there are more that could go in, I have shown you the system for adding them, but if you list which variables you would like, I will add them to the cheat screen myself at some point.

Cheers.

 

[phungus420]

I haven't fully understood the code involved in exposing the variables Hopefully I'll figure it out, but keep in mind everything I learn code wise is self taught, or rather I learn it from asking questions here and getting directions from the likes of EF. My main point is that the rev Idxs should all be displayed (with their cause and values) in the city RevIdx bar. There really is no reason for this information to not be there, and doing so would add transparency to a complex system that currently is very frustrating for the user.

Do I understand correctly from your above post that you and jdog are working on tackling the Multiplayer and Scoreboard issues? If so that's awesome, if those are fixed this mod would be pretty much perfect.

 

[jdog5000]

The scoreboard in Revolution has gone through a couple incarnations to date for exactly the reason you laid out ... too many civs spawning. Now that we're using the BUG system, we should definitely address this again. That scroll thing was cool, though it was a while ago ... I thought it was part of BUG, but apparently not.

As for multiplayer, there are some things I am working on which might make it work there eventually ... but I simply don't have time to pursue and debug multiplayer. I've been intending to switch the popup mechanism to the random event style system which would be multiplayer compatible, but that's a lot of work so it hasn't happened yet.

I agree with you on the transparency part, and that one is really easy to fix ... there aren't really any good reasons to hide that info.

 

[phungus420]

Thanks for the responses. I brought up these issues because my mod, Bronze, Steam & Tears is designed to simulate a "commercially viable" expansion pack. It's coming along well, but these issues (the scoreboard, lack of transparency and multiplayer compatibility) are the main flaws in the mod. Once these are fixed, the mod will comply with my vision for it.

I understand that multiplayer compatibility will be a serious PITA to code. So the fact that'll take a while is just how it goes. I'm very happy though that this is being addressed and a fix is in the works.

The scoreboard, and transparency in Rev Indexes being adressed would be really nice. Is there any chance you guys can make these a priority in the next update(s)? The ulitmate goal should be to have the RevIdx good/bad #s totally add up to the +/- rev index displayed, so that players can see and know the cause and their degree at a glance from the city display.

Also side question: I noticed that RevDefs and RevUtils were updated with my changes, but that Revolution.py wasn't. As far as I know my changes to RevDefs and RevUtils will have no effect without the subsequent code in Revolution.py. Does this mean a city distance and tax calculation similar to what I've uploaded is in the works? I'll attach my current Revolution.py file here with all the trait effects commented out. I've been using these numbers for the past month or so and am getting good behavior from them. This is the Revolution.py file used in WolfRevolution and BST (and outside of being updated to Rev 1.02, this is the same Rev.py file I've been using in BST and WolfRev for the past month, over 1000 downloads of these two mods in that time). The City Distance and Tax Code changes are commented with //phungus start and end tags.

 

[jdog5000]

The scoreboard, and transparency in Rev Indexes being adressed would be really nice. Is there any chance you guys can make these a priority in the next update(s)? The ulitmate goal should be to have the RevIdx good/bad #s totally add up to the +/- rev index displayed, so that players can see and know the cause and their degree at a glance from the city display.

Yes, they should be easy and I'll try to make them happen soon. The numbers won't add up though ... need to switch from integers to floats to be able to have that work and scale with game speed, but I was thinking of doing that anyway.

Also side question: I noticed that RevDefs and RevUtils were updated with my changes, but that Revolution.py wasn't. As far as I know my changes to RevDefs and RevUtils will have no effect without the subsequent code in Revolution.py. Does this mean a city distance and tax calculation similar to what I've uploaded is in the works? I'll attach my current Revolution.py file here with all the trait effects commented out. I've been using these numbers for the past month or so and am getting good behavior from them. This is the Revolution.py file used in WolfRevolution and BST (and outside of being updated to Rev 1.02, this is the same Rev.py file I've been using in BST and WolfRev for the past month, over 1000 downloads of these two mods in that time). The City Distance and Tax Code changes are commented with //phungus start and end tags.

Something is in the works, but life's been crazy for me for the last month ... moving and a bunch of other things all piling on at once

 

[glider1]

Something is in the works, but life's been crazy for me for the last month ... moving and a bunch of other things all piling on at once

I do not understand why in real life, chaotic events must pile up all at once, while in CIV they can be progressed one turn at a time

I will try to contribute to the interface enhancements to release some burdens.
Cheers.

 

[stagnate]

Great mod; I've been accessing it via Phungus' work, but think it's a perfect balance between the dynamic world of Rhye's and static nations of BTS.

I'm providing a second prod to prioritize the revidx information; I spent a lot of time trying to work through the various pages to figure out what I needed to do, and couldn't for the life of me couldn't get anything to balance out.

I did get some information from the boards which helped me understand what was happening, but it's damn near being usable out of the box which is just phenominal. Here's to the last 90% (90% effort in the first part of the project, and 90% in the last 10% .

 

[jmas]

This is just a request to say Pretty please add in more information about the RevIndex.

My first game of Legends of Revolution took a long time because I was trying to figure out how to deal with the rapidly rising RevIndex in many of my cities (starting with the first city I settled besides the capital, only five squares (some diagonally) from the capital (on a Tiny map). My economy was in great shape for most if not all of the game, so that wasn't an issue.

I found myself searching old threads for scraps of information, then using Worldbuilder to give myself techs (e.g. horseback riding, astronomy) and buildings to see if they actually helped or not (e.g. harbor helped, lighthouse didn't, although one old post by phungus implied that it would)--then I'd have to go back into Worldbuilder and take them out. I looked at the file "effect calculator.ods", however, it was hard to find and it was hard to get much insight from it without any commentary or guide. It seemed to not display quite correctly in Google Documents (I don't have OpenOffice 2.0 or higher right now). Google Documents didn't understand the culture rate formulae for example, and couldn't display anything for most of those cells; however the "happiness/unhappiness" part was also very hard to try to make sense of, and I wasn't sure whether this was because of Google Documents or simply the lack of explanation.

I also wasn't sure how much influence "Civ-wide" factors had, which may have been partly because the only time I saw them was when I got a warning that a city's Revindex was over 900. The + or - numbers for "Civ-wide" factors always seemed small, however it seemed to me (subjectively, as I played) that switching to representation (via the pyramids) had a significant effect. But I don't know if that's true or not.

Also, from a gameplay perspective, I was disappointed to find that the Forbidden Palace didn't seem to benefit the RevIndex of any city other than the one it was built in. Is Versailles the same way? And if I may ask, did you intend for the position of the capital to be so important, with the Forbidden Palace lacking the ability to deter revolution in nearby cities? In my game, I kept my original capital, with its abundance of food and early-built cottages (augmented with bureaucracy). However, from the perspective of RevIndices, it would have been better placed (to deter revolution) right in the middle of my empire. Did you intend for this to be a dilemma, or at least an issue, for the player?

These are just some examples, not exhaustive, of things that I was thinking about and that I was wishing for more information about. I can probably think of more suggestions but I think this is enough for this first post.

Anyway, I like the concept and I appreciate your work on it , and I think explaining more of the dynamics/ mechanics would be a very very helpful thing for me and some other people. And, if I understand the mechanics better, then I will be more likely to consider not nerfing the RevIndex accumulation for the next game I play with it (otherwise I will definitely do so, or else turn the Revolution mod component off).

 

[jdog5000]

I think part of the problem with visibility and understandability of RevIdx effects came from the switch from an instantaneous improving/worsening to a rolling average. There are a number of effects which occur at a moment in time (changing civics, accepting/rejecting requests, etc) that have lasting impacts on this rolling average.

(The "trend" average at turn say 101 is 1/3 change in rev index from turn 101 plus 2/3 times the trend average from turn 100. This way, significant recent events have a lasting impact on whether situation is considered to be improving/worsening but the impact reduces exponentially with time.)

I won't be able to fix the visibility of this in time for RevolutionDCM 2.5 as it will take some work, but will make it a priority for the next version.

Also, I highly recommend investigating the Revolution Watch Advisor which glider1 put together. I think you'll find it very helpful.

 

[stagnate]

Good news of the 1.19 work btw. Quick turnaround! Also glad to hear you'll prioritize working on improving the information after this release - I know that the code is pretty complex which makes it hard to display (completely agree with the rolling average, but understand that's hard to convey to the user).

The domestic advisor was great at looking at the environment without having to click through every city, but I didn't find it very helpful to determine the cause of instability (or reference what might improve it). I'll go back into it though and look it over pretty closely to give you some directed input.

I also think that a lot can be done just in term of describing what will improve or worsen the index for a given city, even if you can't tell what the primary cause is of instability, you know the types of actions that you can take to improve the situation.

Second the note from jmas regarding the Forbidden Palace. I busted my chops to get it created and placed, only to find that it didn't make a difference. From a 'realism' standpoint you might not want to make it behave like a palace (and/or from a code complexity standpoint), but for gameplay I think that both it and Versailles should both behave just like a capital for distance purposes, and lower (to some extent) the large empire penalty. Players expect the FP to behave this way for existing distance based features, and I think it should remain that way unless it really creates a gameplay issue.

If you want assistance testing or creating index display information just send me a message.

Thanks again though, this has really renewed my interest in Civ4/BTS!

 

[phungus420]

I think part of the problem with visibility and understandability of RevIdx effects came from the switch from an instantaneous improving/worsening to a rolling average. There are a number of effects which occur at a moment in time (changing civics, accepting/rejecting requests, etc) that have lasting impacts on this rolling average.

This isn't the issue. The issue is that the only thing the RevIdx bar ever shows is City Distance. Even if there are a whole host of things effecting the RevIdx of a city such as being in slavery and despotism, having lost a city, being in financial trouble, and having just built a wonder, the only thing that the user ever sees is City Distance. The user needs to be able to see the causes of instability, that's the issue...

 

[jmas]

Like stagnate, I also want to say thank you for the updating work done so far.

I agree with phungus inasmuch as most effects fall under city distance (a few others are visible, such as nationality, garrison, and culture rate, but in my game they were much smaller than the "city distance" which sounds like it is really a sum of a lot of other factors).

I highly recommend investigating the Revolution Watch Advisor which glider1 put together. I think you'll find it very helpful.

The one accessed with the button with "Che" Guevara's face? I used it, and it was helpful, however I didn't see "civ-wide" effects anywhere in it--that's why I mentioned the "civ-wide" effects. Mostly the advisor seemed to summarize information available on the city screens--which is okay in general, that's what the other city info screens do too. But I only saw "civ-wide" effects when a city's RevIndex went over 900 and I got a warning, and even then the information was hard to interpret. For example, it seemed like my "civ-wide" number changed very significantly with 2 turns of anarchy en route to new civics. That's OK in principle I guess, although the effect seemed stronger than I thought it needed to be, but on top of that, I don't think the number ever recovered very much over, say, a few dozen turns, which left me wondering whether any further civics changes (and resulting anarchy) would be a recipe for disaster. And whether it is a recipe for disaster or not, I'd like to have a sense of that from the information available to me.

I don't want to add too many more issues but here is one that's important to me: having better information about when (generally) is a better vs. worse time to bribe. I probably paid out a couple of thousand gold or more in bribes in the game. I did appreciate the message that I had bribed "x" number of turns ago so costs would be higher than usual. However I think more information would be helpful. Here's what I did:

When possible, I tried to use the largest possible bribes, made as close as possible to a RevIndex of 1000, with long periods in between (plus managing to get some cities stable enough in the meantime to avoid repeat bribing) which I thought was more efficient just from looking at the RevIndex. But looking at the file "effect calculator.ods", it looks like there may be a "snowballing" effect where the higher the RevIndex already is, the faster it grows, which may negate the effectiveness I thought I was seeing of waiting till the last moment before 1000 and then bribing. This is another reason why I think the subject of the RevIndex very much merits an analysis for players. I appreciate the analytical essays that are out there on many points of Civ4. Some of them get into some serious micromanagement issues, but I don't think people would make that accusation as much when their cities are at stake--I wouldn't!

 

[Draknith]

Top Notch guys! Honestly one of the greatest MODs that I've ever used. It incorporates so many ideas that I had wished were addressed in the Firaxis releases. Anywho, I wanted to let you guys know about the multiplayer functionality. I did not know that there was a problem because the last multiplayer game a friend and I played together we were utilizing the MOD as our core game mode. We ran into one sync error, but that was quickly remedied by reloading the game. Otherwise it was an intense game that created one of the best multiplayer games I've ever played.

Thanks again.

 

[Carwyn]

Top Notch guys! Honestly one of the greatest MODs that I've ever used. It incorporates so many ideas that I had wished were addressed in the Firaxis releases. Anywho, I wanted to let you guys know about the multiplayer functionality. I did not know that there was a problem because the last multiplayer game a friend and I played together we were utilizing the MOD as our core game mode. We ran into one sync error, but that was quickly remedied by reloading the game. Otherwise it was an intense game that created one of the best multiplayer games I've ever played.

Thanks again.

Which version of the mod were you using?

Thanks!

 

[glider1]

@Draknith and Carwyn
Yeah RevDCM is becoming really nice. Yes exactly the question, which version were you using Draknith and do you remember what settings were in play when multiplayer was working?
Cheers.

 

[Zeiter]

Hey everyone! Check out this fun and quasi-historical AI autoplay that showcases a lot of the dynamics of RevolutionDCM:

The History of the Decline and Fall of the Roman Empire
http://forums.civfanatics.com/showthread.php?p=8211171#post8211171

I love how stuff like this is possible with RevDCM.

 

[glider1]

Thanks Zeiter and great that you are enjoying it. Your feedback is most useful from a practical standpoint to get a feel for how this mod is going in the field, but I would need to know what version of RevolutionDCM you were using and what the modifications you made to it were. Things like the size of the map, the type of the map and the numbers of starting players is all useful information too.

Cheers.

 

[Zeiter]

Sure, no prob. I was playing a slightly-modified version of WolfRevolution 1.3.2 (just a few new units added and some unit values tweaked for what I perceived as balance purposes. I could send you the files I modified if you want the specifics). Standard map, 7 starting civs, tectonics, 60% water, raging barbs, aggressive AI, start as minor civs on (LOTS of early warfare!), randomized personalities on.

I find that the barbarian civ spawning rate is to my liking at marathon speed on larger maps, but when playing normal/standard, barb cities have a hard time hanging around for long because civs are so close by to pounce on them before they can blossom into barbarian civs. So on normal/standard, I find that allowing barb civs to come from barb cities at size 1, along with a doubled spawning rate, is necessary to get a decent amount of barbarian civ activity.

 

[glider1]

Ok thanks, that version of Wolf was an old RevDCM build not sure what version. Did you add the units as woc modules?
Cheers.