What questions do you still have?

[BornInTheLoo]

I'm looking forward to see what the late-game tourism will look like and if there are any Civilizations left that get bonuses towards this. So far it's America, France and England. Will there be others?

Roman tourism bonuses for monuments/theatres/arenas/aqueducts/baths?
Greek tourism bonuses for coast and cliff tiles?

 

[stealth_nsk]

I am no expert in coding, but from these two statements I conclude that the idea behind Civ V was to hire cheap low-quality coders who were unable to get things done properly in C++. Instead, they were able to quickly rush tons of LUA code, which eventually resulted in poor modding capabilities.
The question is (a) is that true, and (b) is that the case for Civ VI as well.

No, the problem of Civ5 was inabiity of Jon to properly manage project in situation of very limited resources. I can't say whether he would be able to manage project if resources would be more abundant.

In short - the system architecture, including the whole AI code was written by Jon himself and as many single-developer projects, architecture if far from being perfect.

This shouldn't be the case for Civ6. However, the case of limited modding capabilities will still be there. Modding interface is written on top of the game as an afterthought and for quickly changing conditions (like pre-release development) you just can't make mod interface rich enough.

 

[Brutus2]

Religious Victory question...

If more than 50% of every civilization's cities convert to one religion does the victory go to the civ that founded that religion or the one who owns the holy city?

For example, if someone was close to getting the victory could you capture the Holy City and snag the victory for yourself after they did all the work to spread it or would you have to completely eliminate the civilization that founded the religion to stop them from winning?

 

[LDiCesare]

Lua is 50 times slower than native code compiled from C++. Of course there's LuaJIT, but it's still way slower than native code and has its own issues. For processor-heavy tasks like AI you could ignore Lua and code in C++ right away. For professional developer there's not much difference in coding anyway. I'd say due to quite limited Lua capabilities and smaller amount of reusable code, coding in Lua could be even less productive.

As a professional developper who uses both languages for a living, I have to disagree.
Coding in any managed language (lua, go, java, C#, whatever) is so much faster than coding in C++ that the performance factor of the language itself becomes irrelevant most of the time. The productivity in lua is an order of magnitude higher than that of C++. You don't need to reuse any library. When you need something, you just plug the C function you wnat and call it from lua, but most of the time you'll just be coding game logic (or AI behavior actually). The time saved allows for choosing alternate algorithms or pinpointing the parts that need to be optimised in C++. A game should probably have half its code in lua and only the rest in native code (rendering engine, pathfinding, greedy algorithms).

No, the problem of Civ5 was inabiity of Jon to properly manage project in situation of very limited resources. I can't say whether he would be able to manage project if resources would be more abundant.

This I totally agree with. He made architecture choices that made modding difficult not because of language but for lack of software architecture skills. The changes to map scripting I quoted just don't make any sense from an architecture point of view. He could have achieved the same result with a mix of C++ and lua, but went all lua because it likely felt easier to implement.

 

[stealth_nsk]

As a professional developper who uses both languages for a living, I have to disagree.
Coding in any managed language (lua, go, java, C#, whatever) is so much faster than coding in C++ that the performance factor of the language itself becomes irrelevant most of the time. The productivity in lua is an order of magnitude higher than that of C++. You don't need to reuse any library. When you need something, you just plug the C function you wnat and call it from lua, but most of the time you'll just be coding game logic (or AI behavior actually). The time saved allows for choosing alternate algorithms or pinpointing the parts that need to be optimised in C++. A game should probably have half its code in lua and only the rest in native code (rendering engine, pathfinding, greedy algorithms).

1. I wouldn't put Lua in the same line with Java and C#. Lua is minimalistic language, it lacks too many instruments.

2. Modern C++11 (don't even need 14) is quite nice. Smart pointers (including reference counting strong/weak pointers) lets you nearly forget about memory management. And if you need further optimizations, C++ has a lot of tools for it.

 

[LDiCesare]

1. I wouldn't put Lua in the same line with Java and C#. Lua is minimalistic language, it lacks too many instruments.

2. Modern C++11 (don't even need 14) is quite nice. Smart pointers (including reference counting strong/weak pointers) lets you nearly forget about memory management. And if you need further optimizations, C++ has a lot of tools for it.

I'm aware that lua has its issues (nil and false, global vars by default...) even in the limited set it handles. But lua is not for the same tasks as java, C# or C++. It is for those parts of your code that will change often, such as game logic or AI or behavior, for which you want subject matter experts instead of programming experts to be able to code. They should not be able to shoot themselves in the foot, be able to make changes often, and lua offers that whereas compiled languages don't.
C++ has had 'smart' pointers for a long time (through boost for instance). It remains a difficult language. Reference counting is really a poor man garbage collector. The syntax of templates and lambdas is intimidating for instance. It lacks refactoring tools of C# but it's indeed fast. Yet, when you want to go really fast for a game, you're probably going to end up writing shaders anyway. And it's got very poor support for parallelism, contrary to lua's coroutines.
C++ can definitely get you great performance, but it would take much much longer than getting decent performance with a mix of languages. If you mix, you can profile, locate where time is spent, and recode that in C++. You'll end up a few % slower than pure C++ code, but you'll have spent far less time and money, and your code will likely be most robust because you could test it more.

 

[gunnergoz]

You don't invest this kind of development money without a master strategic plan. Civ V was clearly released with incremental expansions and improvements in mind. Civ VI as revealed so far seems to have many aspects of those Civ V improvements and expansions built in from the start. So what is there left to expand the game with, beyond DLC like maps, nations, leaders? What is the strategic plan for Civ VI beyond the obvious DLC?

 

[Bechhold]

Random events? Accurate Earth map?

 

[SJSerio]

Random events? Accurate Earth map?

Also colonies... And maybe civil war...

Sent from my LG-H345 using Tapatalk

 

[stealth_nsk]

But lua is not for the same tasks as java, C# or C++.

Yes, and I'd say more - Lua is not for big parts of code. That's why I don't like the idea of large part of the game being written in it. It's handmade OOP is quite weak for complex architectures.

Reference counting is not good if you dive into it. It's used by Apple in both Objective-C and Swift. And all those complexities of C++, they aren't required if you don't optimize hard. You could write Java-like code with classes, objects and interfaces and a bit of extra syntax for smart pointers. Of course all this will not work if you have to use 3rd-party C++ code, especially legacy code.

You don't invest this kind of development money without a master strategic plan. Civ V was clearly released with incremental expansions and improvements in mind. Civ VI as revealed so far seems to have many aspects of those Civ V improvements and expansions built in from the start. So what is there left to expand the game with, beyond DLC like maps, nations, leaders? What is the strategic plan for Civ VI beyond the obvious DLC?

Well, no. Blizzard invests huge money in their projects without knowing release date or whether the game will be released or not. Firaxis don't have Blizzard resources, but I'm pretty sure developers don't know what they going to implement in expansion - it will depend the first months of real world game experience.

 

[JimmyBanks]

Wow this thread morphed into code speak? I really wish I followed along but you might as well be speaking icelandic <- do they even have their own language? not sure.

 

[stealth_nsk]

Wow this thread morphed into code speak? I really wish I followed along but you might as well be speaking icelandic <- do they even have their own language? not sure.

Well, it's long and poetic discussion about how moddable Civ6 will be

 

[JimmyBanks]

I've never used a mod in any Civ. I'm thinking Civ VI will be different.

 

[SJSerio]

I've never used a mod in any Civ. I'm thinking Civ VI will be different.

There were some great mods in Civ V that really enhanced the gameplay.

Sent from my LG-H345 using Tapatalk

 

[stealth_nsk]

I've never used a mod in any Civ. I'm thinking Civ VI will be different.

Well, I've tried to use Vox Populi, but it has too huge graphical glitches on my hardware. That's quite weird, since all "official" games, including unmodded Civ5, work like a charm. But well, the lack of hardware testing is quite expected downside of mods.

Other than that I always feel like I'm cheating when playing mods, especially in strategic game, so trying to avoid this too.

 

[gunnergoz]

Well, no. Blizzard invests huge money in their projects without knowing release date or whether the game will be released or not. Firaxis don't have Blizzard resources, but I'm pretty sure developers don't know what they going to implement in expansion - it will depend the first months of real world game experience.

Blizzard is a different outfit I have little knowledge of. I'm speaking specifically of Firaxis here and what Sid Meier has in mind for a long term plan for his new baby. Sure, there will be flexibility and opportunism as components of any good plan, but by and large, I'm betting they have a road map fairly well figured out, including what DLC's (i.e. country and map packs) and when. Expansions are probably a bit more fluid, as you suggest, based in part on market experience and feedback.

 

[stealth_nsk]

Blizzard is a different outfit I have little knowledge of. I'm speaking specifically of Firaxis here and what Sid Meier has in mind for a long term plan for his new baby. Sure, there will be flexibility and opportunism as components of any good plan, but by and large, I'm betting they have a road map fairly well figured out, including what DLC's (i.e. country and map packs) and when. Expansions are probably a bit more fluid, as you suggest, based in part on market experience and feedback.

We've seen how plans for the first XCOM were shifted based on player reaction. It was quite large. The third planned DLC was cancelled and content developed for it was included in expansion not planned originally.

I expect Firaxis to have cleat plan for those 4 DLC mentioned in Digital Deluxe. Further DLC and expansion(s) are likely to have rough timeline, but no content proposed for them at this point.

 

[jekke]

No, the problem of Civ5 was inabiity of Jon to properly manage project in situation of very limited resources. I can't say whether he would be able to manage project if resources would be more abundant.

In short - the system architecture, including the whole AI code was written by Jon himself and as many single-developer projects, architecture if far from being perfect.

This shouldn't be the case for Civ6. However, the case of limited modding capabilities will still be there. Modding interface is written on top of the game as an afterthought and for quickly changing conditions (like pre-release development) you just can't make mod interface rich enough.

Could you please elaborate a bit on that? What do you exactly mean by "limited resources", not enough money and as a result, inability to hire top coders? I am struggling to find a link between "limited resources" and "limited modding capabilities".

Or is it just that introducing modding capabilities into the game is an extra work which is to be done on top of actually making a game? And since there was no extra money, this part has been skipped?
But then, the decision whether to stick with simple-and-fast but limited LUA or use more powerful and difficult C++ had to be take in advance... I am just confused how it all worked.

 

[stealth_nsk]

Could you please elaborate a bit on that? What do you exactly mean by "limited resources", not enough money and as a result, inability to hire top coders? I am struggling to find a link between "limited resources" and "limited modding capabilities".

Limited resources here means limited time and money. You need to spend resources on modability and it's not top-priority task especially on game start so if the development is rushed, this hurts modding capabilities.

 

[local_hero]

I watched the livestream where mr Beach played as Barbarossa (the same one where he accidentaly revealed all leaders) and he says that unique districts are built in half the time and doesn't require the population.

Does anybody know more about this? Are they considered "free" or would for example England be able to build their Royal Navy Dockyard while at pop 1-2 and have to wait until pop 6 for their second? Unique districts seem strong either way.