What should be the team's focus for the next release

[Koshling]

I'm guessing that we could (mis)use GeoRealism to generate maps for the other Multi-Maps (except for Galactic, but that is another issue entirely). But my biggest disincentive to working on Multi-Maps is that it will be incredibly hard to get to or test them in a game setting, combined with the fact that the TH era is a bit of a mess and has very little content or balancing, so it seems that we are at another catch 22 situation there.

We just add multiple prehistoric maps artificially as a test vehicle. Can make it a game option 'start on two worlds' with an alternate timeline 'inter-dimensional portals' in which either a 'magic' tech is discovered that enables map-crossing, or a natural feature (a portal) is simply on the map that does it. IMO this has merit in its own right as an alternate timeline (with naturally occurring map portals), but it is a good test vehicle too.

 

[ls612]

We just add multiple prehistoric maps artificially as a test vehicle. Can make it a game option 'start on two worlds' with an alternate timeline 'inter-dimensional portals' in which either a 'magic' tech is discovered that enables map-crossing, or a natural feature (a portal) is simply on the map that does it. IMO this has merit in its own right as an alternate timeline (with naturally occurring map portals), but it is a good test vehicle too.

I wasn't really refering to testing issues, we can always fix that with creative debug code and maps. I was saying that until normal players are getting to that point in normal games, either from faster turn times or from starting at a later era (or both) then there isn't a whole lot of point in adding actual new maps. That was why I was asking Thunderbrd about the later start stuff.

 

[Thunderbrd]

We just add multiple prehistoric maps artificially as a test vehicle. Can make it a game option 'start on two worlds' with an alternate timeline 'inter-dimensional portals' in which either a 'magic' tech is discovered that enables map-crossing, or a natural feature (a portal) is simply on the map that does it. IMO this has merit in its own right as an alternate timeline (with naturally occurring map portals), but it is a good test vehicle too.

Yeah... talk about taking a 'New World' setting to a whole new 'dimension'! I like this idea actually. Gets us working on and hammering out elements of multi-maps without having to conquer our biggest bugbear - the generation of and interaction with new map scripts.

@ls612: What do you feel we need for the later start development plan? Can we start a new thread or do we already have one for that? I know DH has spent a lot of time considering that and has done a number of things to move in that direction already so at this point we'd need to figure out what we need exactly (and how does this really differ from an Advanced Start gameoption? I know there are intended differences but it should be discussed.) What I think we need to do is identify EXACTLY what needs to be done in code, xml, python etc...

 

[RidetheSpiral23]

I agree with this except that I'd add an appeal for us to make the Nomadic Start optional, thus demanding that it BLEND in with the early game balance and progression expectations as we've already established it - things could be adapted a bit to make more room if it needs more 'time' to resolve this stage of the game during a Nomadic Start but all in all, it should not demand big changes to the tech tree or a huge number of new units.

Would it be acceptable to make the option of not playing nomad start to just start at sedentary lifestyle? I feel like it is going to be hard to mesh the two options together and balance them.

I agree with not needing to change much in regards to techs and units.

 

[Thunderbrd]

If we find its nearly impossible to do so I'd take a concession on that but I don't think it'll be as tough as you may think. However, what is it you see that would make that as hard as you think it will be? Maybe I'm overlooking a consideration or two.

 

[RidetheSpiral23]

If we find its nearly impossible to do so I'd take a concession on that but I don't think it'll be as tough as you may think. However, what is it you see that would make that as hard as you think it will be? Maybe I'm overlooking a consideration or two.

Since running through all the XML for the leaders over and over to make sure the are matching up well enough with the normal method and making sure they are diverse enough I have been taking breaks here and there with a write up of nomad start concepts.

It is going to take a lot of discussion and thought before anything is done. I guess I was more on the train of thought of replacing the current structure than balancing the two options. I will putting up a full layout of my ideas here in the near future. I was going for more of a promotions on units opposed to buildings and I feel like that would be hard to balance. I also was liking the idea of switching up gameplay mechanics to add more diversity to gameplay as C2C is such a long and epic play through. I like the idea of a discovery approach to tech research. You wouldn't select techs they would get researched based on the way you are playing. It sounds weird without explaining in depth. I guess if I keep in mind balancing it with normal play while considering nomad options then it could work.

 

[ls612]

@ls612: What do you feel we need for the later start development plan? Can we start a new thread or do we already have one for that? I know DH has spent a lot of time considering that and has done a number of things to move in that direction already so at this point we'd need to figure out what we need exactly (and how does this really differ from an Advanced Start gameoption? I know there are intended differences but it should be discussed.) What I think we need to do is identify EXACTLY what needs to be done in code, xml, python etc...

Procedural generation of Civ stuff (cities with buildings, units, improvements). Here is my original idea.

The second thing would be an actual game change, and a big one. Essentially, here is how it would work:
1. You can start in the Classical or Renaissance Eras, and that would work normally for buildings that care about those things.
2. If you chose to start in one of those two eras, some things would happen.
3. Each civ would get all of the techs up to the start of the chosen era (same as it works now)
4. After map generation, each player would be iterated though a number of times equal to the suggest size option in the WorldSizeInfos (Classical), or until there isn’t any room left in your continent (Renaissance). In each iteration, a city would be placed at the best city spot near the start position, so starting in Classical would give 8-14 cities at game start, and Renaissance would give 20-30 cities at game start.
5. Each city would then be given population equal to ½ the total food yield in its fat cross (or equal to that in Renaissance).
6. Each city would be connected to the other cities with the best available road at the time (think like Via Appia).
7. All resources in city vicinities would be improved and connected to the road networks.
8. Each city would get one worker (two in renaissance) to improve the other tiles.
9. For Each City certain buildings would be built (probably based off of the start conditions in the XML, although other algorithms could be considered). I am unsure of what specifically this would be, that would be open to input.
10. Each City would get 2 Archer-types and 3 Town Watch types to defend with (or double that in Renaissance).
11. World Wonders would be randomly built in cities based off of their size.
12. National Wonders would do the same, excepting that they would be done for each player as opposed to globally.

This would give a valid and very good advanced start option and a good way to jump into later eras right away, which would considerably alleviate the issue of people only ever seeing the early eras.
This is a major proposal, but I think it would be technically doable, and since it would be in the DLL it would be compiled and therefore not take too much time (and whatever time it takes would be a one-time occurrence).

 

[Thunderbrd]

Since running through all the XML for the leaders over and over to make sure the are matching up well enough with the normal method and making sure they are diverse enough I have been taking breaks here and there with a write up of nomad start concepts.

It is going to take a lot of discussion and thought before anything is done. I guess I was more on the train of thought of replacing the current structure than balancing the two options. I will putting up a full layout of my ideas here in the near future. I was going for more of a promotions on units opposed to buildings and I feel like that would be hard to balance. I also was liking the idea of switching up gameplay mechanics to add more diversity to gameplay as C2C is such a long and epic play through. I like the idea of a discovery approach to tech research. You wouldn't select techs they would get researched based on the way you are playing. It sounds weird without explaining in depth. I guess if I keep in mind balancing it with normal play while considering nomad options then it could work.

So tech discovery wouldn't be about purchasing the tech so much as it would be about qualifying for it and being awarded it via the event system? Could work but might be much harder to ensure a fair progression rate between civs. Or you'd have progress scales for each tech and various differing activities would earn more on those scales... but this method would make each tech require a bit of specialized programming.

I'd ask for more explanation on these ideas, in more depth, in the nomadic start thread so we don't bog down with the train of thought here.


@ls612: Alright, taking a look at this in greater detail than I have before:
1-3 are as it is now.

4. After map generation, each player would be iterated though a number of times equal to the suggest size option in the WorldSizeInfos (Classical), or until there isn’t any room left in your continent (Renaissance). In each iteration, a city would be placed at the best city spot near the start position, so starting in Classical would give 8-14 cities at game start, and Renaissance would give 20-30 cities at game start.

How does this differ from the Advanced Start mechanism that allows you to have a number of points to build an initial civilization with the default point setting established at the beginning of the game? Is this just a way to make things get setup in a more streamlined fashion? I suppose I don't like Advanced Start for just that reason, a lot of BS setup time. But then again, having my cities placed randomly wouldn't sit well either as I already tend to demolish enemies just so I can place cities in the newly invaded territories where I feel they make more sense to go. Maybe if we had something LIKE the Advanced Start but say give the player X number of cities, a starting spot, then click to plant and a limit to how far away from another city you can plant another... but then again that might not work either because it could conflict with the other players around you, possibly leading to setup situations where the last player in a region doesn't have enough room to place all his cities... I dunno. I suppose the more I think about it the more I can see the rationale behind the Advanced start mechanism as it is. (At least where cities are concerned.)

5. Each city would then be given population equal to ½ the total food yield in its fat cross (or equal to that in Renaissance).

I'd think you'd want to establish a core city 'birth date' then space out more 'birth dates' in the rest of the cities then base the size on a presumed amount of food collection into a calculation formed around the presumed age of the city.

6. Each city would be connected to the other cities with the best available road at the time (think like Via Appia).
7. All resources in city vicinities would be improved and connected to the road networks.
8. Each city would get one worker (two in renaissance) to improve the other tiles.
9. For Each City certain buildings would be built (probably based off of the start conditions in the XML, although other algorithms could be considered). I am unsure of what specifically this would be, that would be open to input.
10. Each City would get 2 Archer-types and 3 Town Watch types to defend with (or double that in Renaissance).

All just ways to streamline and enforce simpler rules to the Advanced Start really. These, I don't think are a bad idea but it'd take some careful thought how to enact this sort of start along with the city and territory capture system built into the Advanced Start option. Perhaps another option entirely that gives you a limited Advanced Start mechanism then streamlines the rest in the manner you've just expressed.

11. World Wonders would be randomly built in cities based off of their size.
12. National Wonders would do the same, excepting that they would be done for each player as opposed to globally.

This would be frustrating, as mentioned before, but what COULD be done is have all wonders equally divied out to players or put all wonders that would've been built up to this time into a pool and in turns the players pick one and place it into one of their cities until no further wonders remain. This would include both Team Projects and World Wonders.

Then national wonders could be placed one at a time (and potentially skipped if the player doesn't want it.)

This mechanism could be a follow up in a new Advanced Start style gameoption that bundles in with the concepts above.


Alternatively, just establish a lot more points into the defaults on the current Advanced Start method and establish some 'best practices' on how to set things up and post those best practices here on the site to help people get themselves set up faster. It's a bit like using the WB and can be a large and frustrating setup job if you aren't too familiar with it. But it does take all this into account, just not as simplified.

 

[ls612]

@Thunderbrd:

I don't know about others but I have absolutely no desire to spend several hours in Advanced Start micromanaging the making of a later civ on a later start era. That and the AI has no idea how to properly use Advanced start. With this the AI would get the same stuff to start more or less as the human, and it would go far faster.

 

[Thunderbrd]

True... thus why I'm suggesting we design a streamlined Advanced start, primarily because randomly placed cities and wonders are simply intolerable. Getting a full understanding of how the current advanced start works in code would be the key to working that out I think. Probably a complex web of selection lists, much of which could pretty much be copied over or called as is for the new method. It'd be more about what to take out (carefully) of its options and then the rest takes place afterwards, autobuildings, routes, improvements etc...

Either way there'd have to be some supportive AI.

I suppose if you aren't as anal as I am about city placement and wonder placement, the routine you suggest could be pulled off easier but it'd be hard to want to use it for those reasons. Additionally, how would you address the automatic assignment of city placements? Wouldn't be much different working that out as it would be to working out the AI for the other method anyhow.

 

[ls612]

True... thus why I'm suggesting we design a streamlined Advanced start, primarily because randomly placed cities and wonders are simply intolerable. Getting a full understanding of how the current advanced start works in code would be the key to working that out I think. Probably a complex web of selection lists, much of which could pretty much be copied over or called as is for the new method. It'd be more about what to take out (carefully) of its options and then the rest takes place afterwards, autobuildings, routes, improvements etc...

Either way there'd have to be some supportive AI.

I suppose if you aren't as anal as I am about city placement and wonder placement, the routine you suggest could be pulled off easier but it'd be hard to want to use it for those reasons. Additionally, how would you address the automatic assignment of city placements? Wouldn't be much different working that out as it would be to working out the AI for the other method anyhow.

Same way the AI addresses the optimal placement of cities, more or less. It does that pretty well nowadays, so I personally wouldn't mind that for my civ at the start, if I was choosing to start later in the first place. Remember that you adn the AI are starting off on equal footing in such a situation.

 

[Dancing Hoskuld]

For Buildings we have a thread. I reckon that barbarians settling as a nation should be treated in the same way as if they had started in the average era of all other nations or maybe one behind to give a better game.

Reminder of definitions:-

Star in Era = all cities get buildings in the list appropriate to the build era

Free at Tech = only new cities get the building free after the tech has been discovered. It does not work for buildings that require the Palace and maybe some other buildings at the moment.

Auto build = any city that meets the requirements gets this building

Note 1. Auto build checks for all prerequisites. Free at Tech checks for none.

Note 2. There has been a request that no buildings with any negative values be included in either Start in Era or Free at Tech.​

For cities, units and improvements we should also look at the barbarian settling down code and perhaps the rebellion code for the units also. In fact for units I would have thought all three would use the same code modules.

 

[Thunderbrd]

Same way the AI addresses the optimal placement of cities, more or less. It does that pretty well nowadays, so I personally wouldn't mind that for my civ at the start, if I was choosing to start later in the first place. Remember that you adn the AI are starting off on equal footing in such a situation.

While I don't know much about that code, I would greatly disagree that they place their cities well. (I don't know though its possible that Koshling has improved things there and now they aren't loving putting cities one tile in from the coast, one tile away from a river, and spaced out as if the fat square had a two tile radius...)

As I said, I usually prefer to destroy the enemy and rebuild the land from scratch rather than accept the overall impeded ultimate potential thanks to the horrible ways they choose to layout their nation. Its just not worth the developed city usually.

 

[Dancing Hoskuld]

Having two cities being able to work the same resource plot is often an advantage in C2C.

Also, if you weren't around for the earlier ears you just have to live with the "bad" choices your predecessors made.

 

[strategyonly]

Having two cities being able to work the same resource plot is often an advantage in C2C.

Also, if you weren't around for the earlier ears you just have to live with the "bad" choices your predecessors made.

If it should NOT be that way, how to fix?

 

[ls612]

While I don't know much about that code, I would greatly disagree that they place their cities well. (I don't know though its possible that Koshling has improved things there and now they aren't loving putting cities one tile in from the coast, one tile away from a river, and spaced out as if the fat square had a two tile radius...)

As I said, I usually prefer to destroy the enemy and rebuild the land from scratch rather than accept the overall impeded ultimate potential thanks to the horrible ways they choose to layout their nation. Its just not worth the developed city usually.

Well if you see the AI in the process of making those 'bad' choices please post something for Koshling to look at. But in my game, which was started with more or less V27 AI it has been quite good at placing cities in good spots if good spots exist. If good spots don't exist or if they are in contested area then problems can occur, but that won't be an issue here as they don't need to worry about starting cities being seized by enemies during generation.