Its very disappointing from my view. CIV IV is way better.
The mod BTS upto 50% faster by Sephi. Can that be included in a next release?
RevolutionDCM for BTS
- Thread starter glider1
- Start date
Its very disappointing from my view. CIV IV is way better.
The mod BTS upto 50% faster by Sephi. Can that be included in a next release?
AbsintheRed
Deity
Hm, I don't get you comment.
Glider said he will stick to Civ IV too
EDIT: Ohh I misread your first few words...
Glider said he will stick to Civ IV too
EDIT: Ohh I misread your first few words...
glider1
Deity
@AbsintheRed
Yep I'm committed to this mod for so long that Jdog is unable to begin contributing to Civ5. Within RevolutionDCM is the very best work that Jdog ever did that saved Civ4 from death, and I must respect that and if nothing else, the Revolutions code alone must be kept useful and working in harmony with the other main component Better BTS AI. I sincerely believe that without Jdog's vision, Civ4 would not be any better than Civ5 is today and so Civ5 without him, is at least a neutral step with positive hope for the future. It was he that saved Civ4 from becoming boring. To this day I struggle to understand how anyone could actually play Civ4 for any length of time without Jdog's many ideas that make the game more interesting and diverse. From that respect Civ5 is a definite positive first step. Jdog is currently completing a Phd and so his talents will benefit a a bigger scale than is needed here at CivFanatics. When he has time, he will return.
@BobeBrown
I have not even attended to the basic issue that there may be a fault with the attackers AI bombardment of city defenses with siege, or the defenders AI bombardment of units with siege from within the city tile. If you could please take a look at some time and report your findings that would be good. I did some testing a week a go and I struggled to see the AI building any siege but catapults.
As for Civ5 we must be very understanding of the developers. From the moment that the decision was made to fundamentally take Civ into a new direction, there was an unspoken mountain of work that needed to be done. I cannot reiterate just how complex it is to code a good AI for us under the Civ5 model with today's technology. With the very best talent and the totally amazing achievement to get any AI of merit at all under Civ5, the task to build a very high standard AI in the future will take years of work between both the fans and the developers.
I suspect it will be an insurmountable problem for anything but an AI with basic skill level's until the technology changes and we get a learning AI. I say this because of simple commonsense. There is not just one AI in Civ5, but many AI's. We now have unit movement that is equivalent to chess like games, but in those games we have just one AI dedicated to one task, and the player is prepared to wait minutes for the AI to decide. In Civ5 we also want quick turn times and multiple concurrent AI's undertaking multiple different kinds of simulated intelligent behavior.
Cheers
Yep I'm committed to this mod for so long that Jdog is unable to begin contributing to Civ5. Within RevolutionDCM is the very best work that Jdog ever did that saved Civ4 from death, and I must respect that and if nothing else, the Revolutions code alone must be kept useful and working in harmony with the other main component Better BTS AI. I sincerely believe that without Jdog's vision, Civ4 would not be any better than Civ5 is today and so Civ5 without him, is at least a neutral step with positive hope for the future. It was he that saved Civ4 from becoming boring. To this day I struggle to understand how anyone could actually play Civ4 for any length of time without Jdog's many ideas that make the game more interesting and diverse. From that respect Civ5 is a definite positive first step. Jdog is currently completing a Phd and so his talents will benefit a a bigger scale than is needed here at CivFanatics. When he has time, he will return.
@BobeBrown
I have not even attended to the basic issue that there may be a fault with the attackers AI bombardment of city defenses with siege, or the defenders AI bombardment of units with siege from within the city tile. If you could please take a look at some time and report your findings that would be good. I did some testing a week a go and I struggled to see the AI building any siege but catapults.
As for Civ5 we must be very understanding of the developers. From the moment that the decision was made to fundamentally take Civ into a new direction, there was an unspoken mountain of work that needed to be done. I cannot reiterate just how complex it is to code a good AI for us under the Civ5 model with today's technology. With the very best talent and the totally amazing achievement to get any AI of merit at all under Civ5, the task to build a very high standard AI in the future will take years of work between both the fans and the developers.
I suspect it will be an insurmountable problem for anything but an AI with basic skill level's until the technology changes and we get a learning AI. I say this because of simple commonsense. There is not just one AI in Civ5, but many AI's. We now have unit movement that is equivalent to chess like games, but in those games we have just one AI dedicated to one task, and the player is prepared to wait minutes for the AI to decide. In Civ5 we also want quick turn times and multiple concurrent AI's undertaking multiple different kinds of simulated intelligent behavior.
Cheers
AbsintheRed
Deity
It's great to hear this!
I won't go that far that it saved Civ4 from death, but RevDCM is truly awesome, and really important for many mods
Hephaistion
Warlord
I will go that far! RevDCM saved Civ4 from death FOR ME, anyway ... it's a monumental improvement on the original game and introduces all kinds of exciting dynamics, not to mention all the mod improvements that have been incorporated as the mod has grown.
+infinite rep
+infinite rep
Something I noticed, One time I started a game as rome and started very close to the greeks. I killed his scout and took his undefended city. For some reason the game didn't give me an option to capture the city, it just automatically razed it.
Why does this happen?
Why does this happen?
I'd assume because the city is too small. All cities lose population when being captured, some are destroyed by that loss (due to being so small).
I don't think this is RevDCM specific, I believe I had that a few times in vanilla Civ too.
KMRblue1027
Deity
That shouldn't happen. Any city that's producing culture is immune to autoraze.
glider1
Deity
Thanks Ben.
While you play, keep an eye out for:
1)whether you see the AI bombarding city defenses with siege (might not happen if the AI thinks it can capture the city quick).
2)Look also to see that the AI is building trebuchets in the appropriate eras.
3)Look too that the AI is hitting you with collateral damage from any siege in the city as a defense when you are attacking one of it's cities (that might not always happen).
It would be good to have play testers out there for these very subtle possible issues with the AI to track down if there are any. Keep the testing to the above cases otherwise it get's to complicated. If you are playing straight RevDCM, turn your autosaves to every turn just for the sake of record keeping while we play test. Are you playing the beta release from a couple of pages back or the official release?
Thanks!
Cheers
While you play, keep an eye out for:
1)whether you see the AI bombarding city defenses with siege (might not happen if the AI thinks it can capture the city quick).
2)Look also to see that the AI is building trebuchets in the appropriate eras.
3)Look too that the AI is hitting you with collateral damage from any siege in the city as a defense when you are attacking one of it's cities (that might not always happen).
It would be good to have play testers out there for these very subtle possible issues with the AI to track down if there are any. Keep the testing to the above cases otherwise it get's to complicated. If you are playing straight RevDCM, turn your autosaves to every turn just for the sake of record keeping while we play test. Are you playing the beta release from a couple of pages back or the official release?
Thanks!
Cheers
phungus420
Deity
- Joined
- Mar 1, 2003
- Messages
- 6,296
Why would he do that? We know the AI wol't bombard cities, that is the bug in the current version of RevDCM that needs fixing, it's what is holding back release of RevDCM 2.8. There simply are no other issues in the current SVN, just the fact that the AI wol't bombard city defenses (though it would also be a good idea to update the Better BUG AI code as well from Fuyu). Unfortunately it appears no one is able and/or willing to work on this, so RevDCM development has been stalled out for the past month; but having other users confirm the fact that the AI wol't bombard wol't change this, we already know what's broken; I also know I can't fix it, and it doesn't appear anyone else can or will either at this time.
glider1
Deity
Eh Phungus
Why wouldn't Ben do that Phungus? How do we know that he is not willing to keep an eye out? When you say that "we" know it, you may know it but I do not. Are we certain that it's not happening in Better BTS AI and only in RevDCM? If you are certain, when did it stuff up at what version of RevDCM? These are questions that do not require any software skills to answer. You do not have to answer them! You have done enough for this mod. I will try to sort it out in the absence of anyone else.
Cheers
Why wouldn't Ben do that Phungus? How do we know that he is not willing to keep an eye out? When you say that "we" know it, you may know it but I do not. Are we certain that it's not happening in Better BTS AI and only in RevDCM? If you are certain, when did it stuff up at what version of RevDCM? These are questions that do not require any software skills to answer. You do not have to answer them! You have done enough for this mod. I will try to sort it out in the absence of anyone else.
Cheers
Quick questions for veterans of the mod:
Can anyone explain to me what formula the game uses for determining whether, and to what degree, your civilization is in "Financial Trouble"?
I just ask because at times I'm getting flak for it even when my research slider is only on 70% or so and I have 500+ gold in the treasury while researching faster than any AI on the planet with my smattering of financial cottages.
How can I get my citizens to be quiet about the alleged financial instability? Do I have to run merchants and wealth until absolutely none of my slider is going toward gold generation?
I really like the concepts of this mod, but the financial aspect of the revolutions seems a bit weird to me so far. Refreshingly, most of the rest of it doesn't seem as weird.
Also, do any buildings help directly against revolution at all? (Courthouse, monument?) I've read through the Sevopedia but can't seem to find any articles about how the Revolutions mod mechanics work.
Can anyone explain to me what formula the game uses for determining whether, and to what degree, your civilization is in "Financial Trouble"?
I just ask because at times I'm getting flak for it even when my research slider is only on 70% or so and I have 500+ gold in the treasury while researching faster than any AI on the planet with my smattering of financial cottages.
How can I get my citizens to be quiet about the alleged financial instability? Do I have to run merchants and wealth until absolutely none of my slider is going toward gold generation?
I really like the concepts of this mod, but the financial aspect of the revolutions seems a bit weird to me so far. Refreshingly, most of the rest of it doesn't seem as weird.
Also, do any buildings help directly against revolution at all? (Courthouse, monument?) I've read through the Sevopedia but can't seem to find any articles about how the Revolutions mod mechanics work.
I must also say that RevDCM has single-handedly kept Civ4 alive for me. I can't imagine going back to playing regular unmodded BTS. Civ4BTS without revolutions and BarbarianCiv now feels like it would be like vanilla Civ4 missing core game mechanics, like civics or culture.
Anyways, quick question:
I've been playing a personally-XML-modded version of WolfRev 1.61, and I'm wanting to finally port my XML changes over into RevDCM 2.722, which shouldn't be difficult, right? After all, WolfRev was just an XML mod of RevDCM in the first place, right? I understand that some new unit-AI tags have been added in UnitInfos.xml, so I'll obviously have to have a look at that, but was there anything else changed at the XML level? Assuming the rest of it was python and SDK changes, it should be a pretty simple matter of copy-pasting the XML files over.
I do have one final question, however: in WolfRev 1.61, there was a "Revolution.ini" file that had all of the adjustable Revolution and BarbarianCiv parameters. I can't seem to find that file in RevDCM 2.722. Where is it, or if it has been integrated into another file, which one would that be?
Thanks a million!
Edit: Okay, I got everything ported over and booting up just fine (it turns out there were some XML changes to make after all, especially wrt the new inquisitions code that cropped up here and there), except I still can't find where I can customize the revolution parameters.
Anyways, quick question:
I've been playing a personally-XML-modded version of WolfRev 1.61, and I'm wanting to finally port my XML changes over into RevDCM 2.722, which shouldn't be difficult, right? After all, WolfRev was just an XML mod of RevDCM in the first place, right? I understand that some new unit-AI tags have been added in UnitInfos.xml, so I'll obviously have to have a look at that, but was there anything else changed at the XML level? Assuming the rest of it was python and SDK changes, it should be a pretty simple matter of copy-pasting the XML files over.
I do have one final question, however: in WolfRev 1.61, there was a "Revolution.ini" file that had all of the adjustable Revolution and BarbarianCiv parameters. I can't seem to find that file in RevDCM 2.722. Where is it, or if it has been integrated into another file, which one would that be?
Thanks a million!
Edit: Okay, I got everything ported over and booting up just fine (it turns out there were some XML changes to make after all, especially wrt the new inquisitions code that cropped up here and there), except I still can't find where I can customize the revolution parameters.
Mattygerst
King
For what its worth - I'm not sure of the exact problems phungus and glider are referring to because I haven't played REVdcm in some time...but...
I do use BBAI 1.01, and in all of my games thus far I do see bombarding of city defenses by the AI when they're attacking me. I also see them hit stack of death with siege from behind their city walls when I'm attacking. Is this not happening at all in Revdcm 2.722?..
Also - not to be brash - but, it would be entirely unfortunate to not update Rev with BBAI 1.01 - I'm not exactly sure but is there not a person around who is capable of doing so? BBAI makes Civ actually playable. And Revdcm makes Civ actually fun to play. Its a pretty big bummer if we will not see a revdcm 2.8 with bbai 1.01...pretty much have been looking forward to this all summer and now into the fall.
More or less - as a fan of this mod who has been waiting for the "finalized" release for some time - I guess I'm simply asking what stage is this mod stuck and what solutions need to be implemented in order to move the mod forward?...
I do use BBAI 1.01, and in all of my games thus far I do see bombarding of city defenses by the AI when they're attacking me. I also see them hit stack of death with siege from behind their city walls when I'm attacking. Is this not happening at all in Revdcm 2.722?..
Also - not to be brash - but, it would be entirely unfortunate to not update Rev with BBAI 1.01 - I'm not exactly sure but is there not a person around who is capable of doing so? BBAI makes Civ actually playable. And Revdcm makes Civ actually fun to play. Its a pretty big bummer if we will not see a revdcm 2.8 with bbai 1.01...pretty much have been looking forward to this all summer and now into the fall.
More or less - as a fan of this mod who has been waiting for the "finalized" release for some time - I guess I'm simply asking what stage is this mod stuck and what solutions need to be implemented in order to move the mod forward?...
ChrisAdams3997
Prince
lol, I'm not affiliated with RevDCM in any way, but I do have programming experience, and having read this thread yesterday I downloaded the SVN's for both RevDCM and Fuyu's latest BBAI (who's SVN at least is actually at 1.02 
) and just spent several hours today merging all the AI parts (didn't add in any of the new BULL stuff, though I might later). This week as I have time I'll run a test game to see if I missed anything important, but I think I was pretty thorough. This will be the largest merge I've done at once in cIV so far.
As for bombarding, it works fine in 2.722 I know as I've been using that for some time, the problem (which I've only read about here recently) came up apparently in one of the SVN revisions since. Having looked over the code dealing with the AI decision to bombard before attacking and compared it to 2.722, the most likely culprit I've seen (this'll mean more to those who've worked with or looked at the code) is a change in one calculation that subtracts the best attackers survival odds from the base value (default 350) that's compared against the relative strength of the two stacks.
Before, the attacker's odds where multiplied by just 2 then subtracted, so:
@ 50%, 350 - 100 = 250 OR...
an attacker's stack needs an advantage of 2-1/2 to 1 to ignore bombarding
The new code multiplies this by 4, so that:
@ 50%, 350 - 200 = 150 OR...
the attacker just needs 1-1/2 to 1 strength advantage
This could lead to stacks skipping bombarding where they have at least one really good city attacker in particular. To test it out, I'm trying a multiplier of 2.5 there instead, but I won't know more without some testing.
Once I have gotten a chance to test the new DLL out, I can post it and the code here for Glider to look over perhaps.
Chris
p.s. For those who might know-'bombard' can be an ambiguous term between ranged/archer bombardment and vanilla city bombardment, has the problem been with both or no? I don't typically play with ranged bombardment on, but I might need to to see what happens.
) and just spent several hours today merging all the AI parts (didn't add in any of the new BULL stuff, though I might later). This week as I have time I'll run a test game to see if I missed anything important, but I think I was pretty thorough. This will be the largest merge I've done at once in cIV so far.As for bombarding, it works fine in 2.722 I know as I've been using that for some time, the problem (which I've only read about here recently) came up apparently in one of the SVN revisions since. Having looked over the code dealing with the AI decision to bombard before attacking and compared it to 2.722, the most likely culprit I've seen (this'll mean more to those who've worked with or looked at the code) is a change in one calculation that subtracts the best attackers survival odds from the base value (default 350) that's compared against the relative strength of the two stacks.
Before, the attacker's odds where multiplied by just 2 then subtracted, so:
@ 50%, 350 - 100 = 250 OR...
an attacker's stack needs an advantage of 2-1/2 to 1 to ignore bombarding
The new code multiplies this by 4, so that:
@ 50%, 350 - 200 = 150 OR...
the attacker just needs 1-1/2 to 1 strength advantage
This could lead to stacks skipping bombarding where they have at least one really good city attacker in particular. To test it out, I'm trying a multiplier of 2.5 there instead, but I won't know more without some testing.
Once I have gotten a chance to test the new DLL out, I can post it and the code here for Glider to look over perhaps.
Chris
p.s. For those who might know-'bombard' can be an ambiguous term between ranged/archer bombardment and vanilla city bombardment, has the problem been with both or no? I don't typically play with ranged bombardment on, but I might need to to see what happens.
Fimbulvetr
Emperor
ChrisAdams3997
Prince
Which could be true. As long as that is why it's skipping the bombardment (I'm hopeful, as I don't see any other potential culprits yet, not much was changed in that part of the code), then it's just a question of balancing the numbers used properly.
I did start a game yesterday and played through about 50 turns without any problems, so it's a good sign... Now once I get to a point where someone comes with a nice stack with catapults, I can save it there and try some different number combinations and see how it reacts.
Just as a side note, I've also checked through the couple of functions feeding numbers into the decision (compareStacks; attackOdds; sumStrength) to make sure no changes there might have had unintended consequences and haven't spotted anything that concerned me there (and something wrong in those would have more widespread consequences of course, but just making sure ).
I did start a game yesterday and played through about 50 turns without any problems, so it's a good sign... Now once I get to a point where someone comes with a nice stack with catapults, I can save it there and try some different number combinations and see how it reacts.
Just as a side note, I've also checked through the couple of functions feeding numbers into the decision (compareStacks; attackOdds; sumStrength) to make sure no changes there might have had unintended consequences and haven't spotted anything that concerned me there (and something wrong in those would have more widespread consequences of course, but just making sure
).
There is a mistake in the tech diffusion code. The problem is the m_iTECH_COST_FIRST_KNOWN_PREREQ_MODIFIER.
If you set in TechInfo.XML for a tech that requires A and B by
i) A as OR-tech and B as AND-tech
ii) A as AND-tech and B as AND-tech
it should make no difference for the research cost. But with a value != 0 for m_iTECH_COST_FIRST_KNOWN_PREREQ_MODIFIER this is not true. In case i) you get a bonus for A in case ii) you will get none.
Of course you could also remove the line in int CvPlayer::calculateResearchModifier(TechTypes eTech) const or you could give the same bonus to the first AND, but why?
If you set in TechInfo.XML for a tech that requires A and B by
i) A as OR-tech and B as AND-tech
ii) A as AND-tech and B as AND-tech
it should make no difference for the research cost. But with a value != 0 for m_iTECH_COST_FIRST_KNOWN_PREREQ_MODIFIER this is not true. In case i) you get a bonus for A in case ii) you will get none.
Of course you could also remove the line in int CvPlayer::calculateResearchModifier(TechTypes eTech) const or you could give the same bonus to the first AND, but why?
Similar threads
