Obsolete buildings provide nothing anymore, they are cleaned out of the city so we can save memory per city at next save break.
Quick Questions , Quick Answers
- Thread starter TaylorItaly
- Start date
Somebody613
Somebody613
- Joined
- Nov 18, 2019
- Messages
- 2,813
It would've been much smarter to convert them into something like a "special Great Person - Wonder Man" (or "Archeologist", lol) on obsoleting.
Or better yet, remove "Wonder buildings" in the first place, replacing them with various amounts (and/or bonuses) to that "special GP".
It would definitely consume waaay less memory, yet would also never obsolete - because I find the entire IDEA of "obsoleting Wonders" to be kinda stupid to begin with.
Pyramids are still a popular tourism site, despite totally having lost their ACTUAL historical "Wonder meaning".
And the same applies to any and all surviving real life "Wonders" - they change context, but they never "obsolete" to the state of providing NO "culture/tourism/science/whatever".
With the exception of those that had been totally and utterly destroyed, of course - but those are very clearly a minority of real life "Wonders" (especially those "weird in-game" ones).
I implemented xml tags and code so that a building can become a different building upon obsoleting, so wonders that shouldn't get lost to time can obsolete into a new building to change its stats towards tourism or other stuff.
Very few obsolete buildings actually served a purpose, so I felt it made sense to clear them out of memory per city (not actually fully done yet, changing what obsoletion does was just step one, step two will liely break saves so it is saved for later) and rather have the building transform into something new on obsoletion if we wanted a building to change its effect over time.
Before this change, only culture from a couple buildings, some city defense modifiers, and great person contributions continued to work after building obsoleted. With the new system a building that obsolete can have whatever building stat we want after obsoleting by replacing it with a new building.
I only bothered to make a few world wonders obsolete into a new building when I implemented the tag, and changed how obsoletion works, because I am not an XML modder, I leave xml design up to modders like raxo.
Very few obsolete buildings actually served a purpose, so I felt it made sense to clear them out of memory per city (not actually fully done yet, changing what obsoletion does was just step one, step two will liely break saves so it is saved for later) and rather have the building transform into something new on obsoletion if we wanted a building to change its effect over time.
Before this change, only culture from a couple buildings, some city defense modifiers, and great person contributions continued to work after building obsoleted. With the new system a building that obsolete can have whatever building stat we want after obsoleting by replacing it with a new building.
I only bothered to make a few world wonders obsolete into a new building when I implemented the tag, and changed how obsoletion works, because I am not an XML modder, I leave xml design up to modders like raxo.
Spoiler List of wonders I made special obsolete variants for :
Last edited:
My only concern is that a neural net AI would be impossible to manually tweak/adjust to get specific change in AI.
I.e. expand upon it with traditional human AI programming. Might also be harder to debug certain bugs? Performance? May be non-issues, just thinking out load that it might need to be implemented as an option.
As long as the questions are far shorter than your first post here, then yeah, that's what this thread is for. ;p
We've long delayed working seriously on AI coding, C2C AI is not actually far developed from vanilla relative to other C2C changes away from vanilla.
We've recently started talking about working more seriously on AI though, so maybe the timing of your arrival is ideal.
Last edited:
Somebody613
Somebody613
- Joined
- Nov 18, 2019
- Messages
- 2,813
[SUGGESTION]
Team Project Super World Wonder: Text Wall.
+9001
for all other players.
Team Project Super World Wonder: Text Wall.
+9001
for all other players.@JaguarDemon Welcome to civfanatics btw.
We have a discord channel here https://discord.gg/cZ5N3Jdq
Feel free to start a group PM either here or on our discord server, main people to include would be me, Thunderbrd, MattCA, and Flabbert..... perhaps also Blaze (not very active in a while).
We are the current active coders on C2C.
We have a discord channel here https://discord.gg/cZ5N3Jdq
Feel free to start a group PM either here or on our discord server, main people to include would be me, Thunderbrd, MattCA, and Flabbert..... perhaps also Blaze (not very active in a while).
We are the current active coders on C2C.
Thunderbrd
C2C War Dog
@JaguarDemon We don't make a practice of turning down offers of help where the help is needed. This is a team that has and always will give of themselves just to see the overall whole be better for it. There have been only a few cases where the personalities and culture clashed to a point we had to make any 'decisions' to turn down an offer or two of assistance. We're all volunteers and do our best to respect that about each other not only from the perspectives of what we're willing to do but also what we expect of each other, which really comes down to no expectations at all, just thankfulness usually, for what contributions we can afford in our lives to give. So what I'm trying to say, in a nutshell, is that we'd be thrilled to have your help, particularly if it is as techno skilled as you claim, which I believe - I've met some of the best thinkers on this project.
I've recently been looking at a large scale of architectural plans to greatly improve our AIs, and we've been getting a lot done in that direction from a few different project fronts. Therefore, before you dive in, I'll need to somehow find the time, best perhaps on Discord, to dive into the architectural redesign concepts that I've spent more than a year or so giving plans and considerations without quite yet putting any of those plans into action.
However, its understood that a neural experience driven AI could potentially achieve better, sure. To do so, we'd really have to talk about some of the technical details and limitations we may be working with here and how we would actually go about this in a manner that could not only be designed, perfected, and applied, but also helping to define where the edges are between directed AI and honest artificial data driven AI. The data storage and management on this sort of project as I understand such AI concepts could be... well to say demanding would be laughably understating it if I understand how these things work. What I think you might find in its implementation is just how many broken or poorly established AI input points there may actually be. If you are as good as you say, I might expect if you joined us, the first thing you'll find is how insane things actually are under the hood - it's decades of different design technique and eras of 'best practices' being on display here, like any very very old home might display in its very architectural core, and then there are issues interfacing with the deepest part of the core that MAY end up being deeply frustrating as well. I won't sugar coat that we often have deeply talented contributors push us to simply start over from scratch thanks to some great limitation discovery or another discovery of deep flaws, not the least of which is our overall memory limitations thanks to this ancient exe, and that is always on the consideration heap. A neural AI would probably need to at least be an external process that somehow plugs in without creating much more memory weight on the underlying process thread.
I'm positive towards getting your help. I'm curious if the plans you have can be implemented. Even moreso to see how much room it gives us to manipulate the way the end result works.
At the same time, be patient with all of us, myself as much or more - I have not had much time to devote to C2C of late which is frustratingly sad to me and I feel like life keeps forcing me to deal with other projects harder and harder the more my will to get back into a groove of working on things here continues to grow, almost as if the hand of the divine is personally offended that I would mod further. But I do at least try to find the time to discuss the thoughts I always have about the project. So yeah definitely join us on Discord for more chat and we can explore what you have in mind on a deeper level.
I've recently been looking at a large scale of architectural plans to greatly improve our AIs, and we've been getting a lot done in that direction from a few different project fronts. Therefore, before you dive in, I'll need to somehow find the time, best perhaps on Discord, to dive into the architectural redesign concepts that I've spent more than a year or so giving plans and considerations without quite yet putting any of those plans into action.
However, its understood that a neural experience driven AI could potentially achieve better, sure. To do so, we'd really have to talk about some of the technical details and limitations we may be working with here and how we would actually go about this in a manner that could not only be designed, perfected, and applied, but also helping to define where the edges are between directed AI and honest artificial data driven AI. The data storage and management on this sort of project as I understand such AI concepts could be... well to say demanding would be laughably understating it if I understand how these things work. What I think you might find in its implementation is just how many broken or poorly established AI input points there may actually be. If you are as good as you say, I might expect if you joined us, the first thing you'll find is how insane things actually are under the hood - it's decades of different design technique and eras of 'best practices' being on display here, like any very very old home might display in its very architectural core, and then there are issues interfacing with the deepest part of the core that MAY end up being deeply frustrating as well. I won't sugar coat that we often have deeply talented contributors push us to simply start over from scratch thanks to some great limitation discovery or another discovery of deep flaws, not the least of which is our overall memory limitations thanks to this ancient exe, and that is always on the consideration heap. A neural AI would probably need to at least be an external process that somehow plugs in without creating much more memory weight on the underlying process thread.
I'm positive towards getting your help. I'm curious if the plans you have can be implemented. Even moreso to see how much room it gives us to manipulate the way the end result works.
At the same time, be patient with all of us, myself as much or more - I have not had much time to devote to C2C of late which is frustratingly sad to me and I feel like life keeps forcing me to deal with other projects harder and harder the more my will to get back into a groove of working on things here continues to grow, almost as if the hand of the divine is personally offended that I would mod further. But I do at least try to find the time to discuss the thoughts I always have about the project. So yeah definitely join us on Discord for more chat and we can explore what you have in mind on a deeper level.
Thunderbrd
C2C War Dog
Well... posting your list of questions, perhaps in a new thread specifically for discussing your more in depth exploration than the 'quick' questions and answers thread for the more casual player, may be a good idea, since discussions on discord can wander a lot. Maybe you can maintain your garden of questions on the first post and where we are with answering them in general, but it might help you to have that project list of things to discuss located in its own home. Discord is for direct real time discussion when we can find each other but the forum is great for more patient and much longer and deeper involved conversations that need to make sure all bases are hit.
SOME of the more technical minds tend more towards the chats on discord and some of the team prefers the discussion to be here so ... there's that too to consider. IN part it's a who you want to address preference as well.
Thunderbrd
C2C War Dog
Fortitude would be pretty much the same thing on all units and shouldn't 'just' show up there though I'm not sure where it does at the moment. The Combat Mod threads should break that down and you can see the 5 intro threads to the Combat Mod linked to on the 1st page of the Modder's thread. Interesting if its been put to use somewhere a bit early at this point. I don't think it actually has a lot of application yet if some of the other things its there for also haven't been put to use yet.
Sakura Matou
Warlord
Sorry to be a pest and if this questiion is stupid, but in the BUG menu in the REV tab is the " Barb Tech Percent Setting" Reversed? Recently there was a fix to the SVN that when set to 0 they woould get most techs in the game. Now it seems when I set it to Zero when A barb becomes a full fleged Civ they get All techs I have then some. Every time now they take all Religious Techs and grab Sed. Lifestyle 
So was wondering if Zero actually means 100 percent of Techs I own.
So was wondering if Zero actually means 100 percent of Techs I own.
Zero is now actually almost 0, that is less than 5.Sorry to be a pest and if this questiion is stupid, but in the BUG menu in the REV tab is the " Barb Tech Percent Setting" Reversed? Recently there was a fix to the SVN that when set to 0 they woould get most techs in the game. Now it seems when I set it to Zero when A barb becomes a full fleged Civ they get All techs I have then some. Every time now they take all Religious Techs and grab Sed. Lifestyle
So was wondering if Zero actually means 100 percent of Techs I own.
Before fix it was more of Not A Number - first minor civ got all possible techs that exist in game including Transcendent ones.
Sakura Matou
Warlord
Yeah they are no longer getting all Techs any more but they seem to be getting all the ones in the current era I am in ((Prehistroric Era)). When they Form a Civ the get all the religious techs and hit Sed Lifestyle. Oh well.
You should read the tooltip for it.
0 mean that at least 0 percent of the players need to know the tech for the barb to get it for free. (bug fix here was that at least 1 player must know the tech rather than 0 so that barb civs don't get all the techs in the game)
50 Means that if at least 50% of the players know the tech, barb civ will get it for free.
100 means that 100% of all players must know the tech for barb civs to get it for free.
Last edited:
Sakura Matou
Warlord
Aha thanks! Did read the tool tip but I mis understood it! I thought it meant as in they would get Zero percent of the techs I knew. Would explain why they were getting all the techs I had haha! I'll set it to 100 now.
((sorry english isn't my first language))
((sorry english isn't my first language))
Somebody613
Somebody613
- Joined
- Nov 18, 2019
- Messages
- 2,813
There's a very annoying downside to this popup in the first place:
You only get to see the options of WHAT to build - but you don't exactly see WHERE it sits.
Not all resources are THAT much obvious visually (or are obscured by units altogether), nor should I be forced to literally deduce the required relevant improvement by heart, dammit.
You may call this "being a noob", but I rather call it "bad UI choice", lol.
Maybe verify your settings, because I have a giant light beam, in the middle of the screen, pointing to this resource.
So effects disabled must be unselected
Somebody613
Somebody613
- Joined
- Nov 18, 2019
- Messages
- 2,813
Clarification of what I meant:
I can see the POSITION well enough.
I can't exactly always see the RESOURCE under it, though.
I usually tend to rely on mouse-over for resource data more often than not - because I seriously don't feel like MEMORIZING ALL THE MAP RESOURSE IMAGES, ya know.
And also some earlier improvements are still too obscure to GUESS correctly without actually reading their effect.
To sum up:
This event causes me to 50%-80% guess what resource MIGHT be in question, and then guess what improvement is NEEDED there to begin with.
Not all improvements are obvious in regards to their resources, either.
Why is this so hard to understand?
Like, do YOU all guys remember ALL resource-improvement correlations (including those early ones, that are most ambiguous) by heart, really now?
Similar threads
