RevolutionDCM for BTS

@Phungus
When you think of ranged bombard, think of combat in the field between stacks, with the exact same collateral damage calculations from standard BTS, except that the seige does not have to attack to achieve it. Without attacking but just bombarding, the head unit does not receive any damage as per standard collateral. Only attacking damages the head unit as well as effecting collateral as per standard BTS. When range bombarding a city, the calculation is the same as standard BTS bombarding, except that there is a small chance of damaging the defenders in the city and a greater chance of completely missing altogether. Straight bombarding cities is far more effective when trying to reduce a citie's defenses, but range bombarding does have it's place if you want to introduce a chance of weakening the city defenders in order to speed up the capture of the city. Range bombarding from behind city walls at the attackers in the field, the odds of success are significantly reduced. If you have enough seige, you can improve those odds and begin to wear down the attacker, but it's difficult to pull in sufficient seige to make range bombarding city attackers effective. It's desperation stuff really but if all else fails, range bombarding from behind city walls into the field can make the difference. Best of all is to build fortifications outside of the city and then range bombarding from there. Forts are great for this purpose. However that carries associated risks and possibly some loss of productivity if you are building forts to protect a city on prime agriculture. I cannot remember the exact odds for the various scenarios, but can supply if you want them. The previous DCM bombard was badly unbalanced because the distributed effect of range bombard on the victim was significant across the stack (including damage to seige), effectively killing off any chances the offensive stack had of success.

@Cripp
I Think you are referring to Jculture. I have that on the to do list for sure.

@Avain
Good question on Inquisitions. The Inquisitions in RevDCM is totally tailored to the Revolutions mod at the moment. It's a great way to moderate and tame Revolutions, as is the limited religions option. However, the RevDCM Inquisitions AI could be made to work without Revolutions, but haven't bothered because that area is covered by other mods like WoC inquisitions and OrionVet. If you think it would be good, will add it to the list.

The main work that will be going on behind the scenes as far as I'm concerned (Jdog is continuing high level work on the AI and Revolutions), is to attempt to add WoC compatibility to RevDCM without breaking backward compatibility with the range of mods that are already incorporating RevDCM in non-WoC format.

@Xaxxa38
Try range bombarding using straight RevDCM and compare with range bombarding from your Amra mod. Perhaps that will help clarify where the problem is.

Cheers.
 
A quick change you can make in the spirit of influence driven war: set the free culture value in the XML to 0. In the base game, a city generates 20 free culture on plots for every point of difference between that plot and the maximum radius. Removing this makes culture much more important and prevents scenarios where taking those last couple of tiles next to a city impossible because they might be getting a +80 bonus for free.
 
glider, in terms of ranged bombard, I suppose my question is how is a hit% determined. I've set it so Artillery and Mobile arty have a DCM hit% of 100, but they still miss quite frequently...

Edit: Also something of a Bug report from a user:
http://forums.civfanatics.com/showpost.php?p=7989493&postcount=647

Not sure about it though, I never really make it to the late game where I'm bombing resources, and you can't tell if the AI is bombing resources from watching the AI... Has anyone else noticed this?
 
@Phungus
Attack city range bombarding:
Here are the global alt defines variables for ranged bombarding A CITY assuming you are bombarding from the adjacent tile (taken straight out of CvUnit::bombardRanged).

If a city still has defenses:
DCM_RB_CITYBOMBARD_CHANCE: default 5 (20%). Lowering the number increases chance of ranged bombard happening over straight city bombard, otherwise raising the number decreases chance. The lowest possible number is 1, which would mean range bombard would happen 50% of time relative to standard bombard. If you set it to 100, there would be a 99% chance of standard bombard happening and a 1% of range bombarding.
DCM_RB_CITY_INACCURACY: Odds defined by the unit XML (for trebs this means their accuracy is 50/100 or 50%) If range bombard is to occur according to the above probability, this means a 50% chance that collateral damage will be issued on the city defenders otherwise a complete miss.

If a cities defenses are down:
The chances of ranged bombard happening over standard bombard are 100%
DCM_RB_CITY_INACCURACY: default 350 (for trebs this means their accuracy is 50/350 or 22%) This means a 22% chance that collateral damage will be issued on the city defenders, otherwise a complete miss.

In other words, if you are hitting a city with defenses still in place with trebs, but decide to range bombard attempt, your odds are 20% that it will happen at all (otherwise just straight bombard damage will happen), and then 50% of effecting collateral damage or complete miss on the defenders if it has happened. If the defenses are down and you decide to range bombard attempt, it will happen 100% of the time, but only with a 22% success rate otherwise complete miss. This simulates the difficulty of collaterally hitting units that are still shielded by the rubble of the city with range bombard, when what you should be doing is moving into the city and all out attack moving into the city with seige and your infantry.

Defending city with range bombard
DCM_RB_CITY_INACCURACY: default 350 (for trebs this means their accuracy is 50/350 or 22%) This means a 22% chance that collateral damage will be issued on the city attackers, otherwise a complete miss. However if you have lots of seige, each one get's 22% odds which means that if you have five trebs, one could very well be successful. This can have a big impact in defeating the attack stack.

In other words, if you are trying to range bombard from within a city with trebs, you don't have good chances because trebs struggle to be effective behind all the impediments of what are city structures.

Field Bombardment
This is the case of tile to tile combat including forts and assuming adjacent tiles:
Odds are defined by the unit XML (for trebs this means their accuracy is 50/100 or 50%). If range bombard is to occur according to the above probability, this means a 50% chance that collateral damage will be issued on field stack otherwise a complete miss. For mobile artillery it means 80% (from memory). These odds are not changed from standard DCM.

This discussion has not included the probabilities of plot bombardment succeeding. If there are any errors in what I have written here, it is because it is quite hard work translating C++ code into readable english language and that translation may contain errors.

Enjoy.
 
@Avain
Good question on Inquisitions. The Inquisitions in RevDCM is totally tailored to the Revolutions mod at the moment. It's a great way to moderate and tame Revolutions, as is the limited religions option. However, the RevDCM Inquisitions AI could be made to work without Revolutions, but haven't bothered because that area is covered by other mods like WoC inquisitions and OrionVet. If you think it would be good, will add it to the list.

The main work that will be going on behind the scenes as far as I'm concerned (Jdog is continuing high level work on the AI and Revolutions), is to attempt to add WoC compatibility to RevDCM without breaking backward compatibility with the range of mods that are already incorporating RevDCM in non-WoC format.

I'd like to have the possibility to play without the Revolution component (and of course anyone who misses it could just turn it on).
As RevDCM incorporates almost all the mods I like (just I'm thinking about turning Revolution OFF for the moment), I'd like to ask you to add this to your list if possible.

Also, do you use Unofficial Patch version 0.21? (in the readmes included I could only find v0.19)
 
Also, do you use Unofficial Patch version 0.21? (in the readmes included I could only find v0.19)
All the unaltered gameplay mods (BUG, Better AI, and the Unoficial Patch) are included in RevDCM, and are currently up to date.
 
Is there a fix for the TXT_KEY_REV_AND?

I see where it occurs in the Python... but I'm not sure why it's not or what would fix it.
 
Duuk, at the moment I am confused about this problem too. I look at the code and there appears to be nothing wrong. Hmmmm.

@Avain
Ok will make the inquisitors AI independent of revolutions, good idea.

Cheers.
 
I originally posted this in the WolfRevolution thread, but at Phungus' suggestion I tested with RevDCM v1.01 and I'm having the same problem...

I've been noticing that aircraft bombing tiles have seemingly been nerfed along the way... I have noticed this for a few months and originally chalked it up to bad dice... now I think it is a bug...

The conditions to recreate are pretty simple... start with any aircraft and bomb a mine or oil somewhere... it just never seems to succeed. I'd wager I've tried at least 40 or so missions in no less than 10 different games without a single success... Contrast this with vanilla bts, where you can get a significantly better success rate than 0 out of 40. BTW, I don't count missions where my bomber intercepts something or is intercepted...

I'm including a save game below where I worldbuilder'd in a bunch of bombers and went to war with Joao for the purposes of demonstrating the problem... if you try to bomb his sugar tile (or any other AFAIK), it never succeeds.

Thoughts?
\gg

The save game is linked here:
http://forums.civfanatics.com/attachment.php?attachmentid=211141&d=1240007257
 
@Gatling
Yep, got you. Thanks for the save game. Makes it so much easier. I have not checked the save game but will use it to check that the bug has been squashed. Think I know why it could be happening. If it is a confirmed bug and I do fix it, will release an interim DLL and report.
Cheers.
 
glider, quick request:

When you upload the fixed source code for the above bug, could you also include jdog's fix for the reinforcements and maxglobalinstance units as well? He said he updated the fix for the SVN (only applies to mods with wonder units, but Bronze Steam and Tears is one, and I want to get that in the next release), but I can't figure out the SVN :dunno:

Also if you're going to update the .dll, You might as well include the Exmine City on Conquest modcomp. It's totally stable, and the source is very simple and well commented. Took me all of a minute to merge it into the RevDCM core.
 
@Phungus
Not sure about any new mods, but at least a RevDCM 1.02 should come out reasonably soon. I will look at Gatling's bug and see how it goes with Jdog and test the water before diving in.
Cheers.
 
OK, I hear that. It just sounded like you were going to release a quick fix for GG's tile bombing bug immediatly. In terms of the Examine city on conquest feature I figured you could just add that in, because it really is very simple. Two quick snippets that are commented and it's done, I'm not exagerating when I said it only took me a minute to merge it into the SDK so it's not time consuming or difficult in the slightest, unlike the more common with SDK modcomp merges which can be time consuming.
 
Well done GatlingGun! Nice find and thanks again for the save game which cut debug time down by a third and made it heaps easier for me. The bug is traced back right to the origin of the DCM mod itself. It was a very deceptive bug indeed. Without DCM airbombing turned on, bombing plots with aircraft appeared to work perfectly except for a 0% success rate. With DCM airbombing turned on, behaviour was always normal.

Here is a zip of the fixed DLL and code. GatlingGun's save was so good, it also allowed me to double check the other DCM airbombing behaviours except that the DCM AI airbombing mission decision code has not been checked. Keep in mind that this DLL is the latest build off sourceforge and not an official release.

@GatlingGun
Might be worth to double check the fix yourself in case I missed something. :goodjob:

Cheers.
 
You know what Revolution needs? It needs a tag so you can make Civilizations foundable only by Revolution or Barbarian promotion. In other words, just as some civs can be only AI playable, there would be some civs that are simply not in the running to be generated at the start of a game, they can only show up randomly during it.

Or maybe even better, a tag to determine what eras a civ can start in. So, if you start a game in the Ancient Era, only a set of Ancient Era civs can be generated on the map or selected to be played or produced by revolutions or from Minor Civs settling down. Only those with
<EraStartables>
<EraStartable>ERA_ANCIENT</EraStartable>
</EraStartables>

The above might be for Sumeria, whereas Rome might use:

<EraStartables>
<EraStartable>ERA_ANCIENT</EraStartable>
<EraStartable>ERA_CLASSICAL</EraStartable>
</EraStartables>

meaning that it would be among the possible civs generated on a the map for an Ancient or Classical era start and one of the civs that might be generated during those eras by Revolution or Barbarian promotion.

and America might be

<EraStartables>
<EraStartable>ERA_RENNAISSANCE</EraStartable>
<EraStartable>ERA_INDUSTRIAL</EraStartable>
<EraStartable>ERA_MODERN</EraStartable>
</EraStartables>

meaning America would not appear on any map before the Rennaissance, but might appear on any after that.
 
I agree completely with Tholish. I would also propose adding a tag for civs which can only be created by colony (so I could finally be satisfied with the inclusion of America/Canada/Australia).
 
I would also propose adding a tag for civs which can only be created by colony (so I could finally be satisfied with the inclusion of America/Canada/Australia).
America has been the strongest, and most influential nation on the planet in the post WW2 world (and arguably post napoleonic). It also pretty much invented the modern notion of liberal government, over 100 nations constitutions are modeled on the US. Hell the very notion of a constitution sealing the social contract was created by the United States, before the American Revolution any so called constitution was between a Sovereign and some other entity claiming protectorship of The People. To deny America's influence in history, it's significance, and it's Greatness as a World Class Empire, is to deny reality.
 
America HAS been very influential in history, but it need not necessarily have been that way. In an alternate history America may never have emerged, or its analog may have emerged somewhat earlier, later, or elsewhere. When you start a random game, you are asking the computer to plunk you down in an alternate history. In fact, the addition of effective tags like this would make it easier to calibrate for your own vision. Since America would be one of the few later era Civs, its emergence would be highly probable.

Another thing that would be nice would be if the same sort of thing could be applied to Leaders, so that if America emerged in the Rennaissance it could only have Adams or Washington, if it emerged in the Industrial it could also have Lincoln or Jackson, if in the Modern era it could have Roosevelt or Reagan. The idea is that the only anachronism possible should be the longevity of leaders. Its understandable that if the civ has been around since Ancient Times it is still represented by the same leader, unless it has been destroyed and reconstituted like Egypt or Greece. If the Ptolemaic dynasty had survived to modern times, it would be understandable to have a Cleopatra, but if it is wiped out by Rome and remains under one empire or another until modern times, when it breaks free, it should restart with a Nasser or a Sadat leader. But for a leader and civilization from the future to appear prematurely is bothersome. No Stalin vs Mao in the middle ages.
 
Back
Top Bottom