Will the Civ 6 modding community be the quickest ever to establish itself?

[Chinese American]

The game would run like pure crap.

LuaJIT is very fast and almost approaches C/C++ in speed. There's also Havok Script, which is basically Lua with (proprietary) optimizations and enhancements, and is used in many AAA games across all the major platforms and consoles. Just found out that Civ:BE already uses Havok Script. So the transition should be seamless to develop the new game in the series almost entirely using the same, fast, powerful scripting language.

There will be many benefits. This will simplify development for the developers. They don't have to go back and forth between C++ and Lua. Changes can be tested almost instantly without recompile and restart application. New concepts and content can be created and tested more rapidly. i.e. new DLCs, civs, traits, etc.

 

[Delnar_Ersike]

LuaJIT is very fast and almost approaches C/C++ in speed. There's also Havok Script, which is basically Lua with (proprietary) optimizations and enhancements, and is used in many AAA games across all the major platforms and consoles. Just found out that Civ:BE already uses Havok Script. So the transition should be seamless to develop the new game in the series almost entirely using the same, fast, powerful scripting language.

There will be many benefits. This will simplify development for the developers. They don't have to go back and forth between C++ and Lua. Changes can be tested almost instantly without recompile and restart application. New concepts and content can be created and tested more rapidly. i.e. new DLCs, civs, traits, etc.

I've fiddled with implementing LuaJIT into Civ5's DLL (compiled it from source using VS9.0, compiled the DLL with LuaJIT's .lib file and headers, included a replacement lua51_Win32.dll file with the mod that was just the compiled LuaJIT .dll renamed to overwrite the original Lua DLL), and in summary, I'd say it's basically Lua's C++: it makes Lua code run fast, but when something goes wrong, you have absolutely no idea why it went wrong and debugging is a nightmare (even more so than usual C++ debugging). Also, LuaJIT absolutely does not run at speeds comparable to C++ when it comes to Civ-specific code in my experience: when I added the ability to call Lua code that could modify a building's flavor value and used it to execute some simple conditionals that ran long enough to reliably time (e.g. add growth flavor to granaries for each wheat, bananas, and deer in an owned plot in the city's radius), that section of code ran about 100x slower in LuaJIT than an equivalent in C++ (I timed it with a profiler), even with hexspace-optimized for() looping logic (i.e. x and y coordinates are looped in a way that allows all looped coordinates to be valid, instead of having to check for tiles that are actually too far away from the origin in hexspace and discard them, thereby wasting cycles). When looking at online benchmarks comparing different languages, remember that some languages are built to be faster at certain tasks than others; a classic example is Fortran, which people find is a lot more sluggish at vector operations and OOP-related fanciness than C++, but is significantly faster when it comes to crunching large arrays of numbers (which is why lots of physics and computational chemistry simulations are written in Fortran). It's possible that LuaJIT has performance comparable to C++ for certain algorithms, but that doesn't matter when the majority of the DLL's code is compromised of algorithms that do not fall under this category.

Considering that despite the greatly reduced performance of Lua code, the game's main CPU bottleneck is still the pathfinder (again, we know because we've run profilers), which relies on no non-C++ code, I very much think that moving the CvGameCore code from C++ to LuaJIT would be terrible idea.

 

[zwei833]

No one talkig about the unit models modding? I think the art style of Civ6 units are pretty similar to the unit models in Civ3, so perhaps we may be benefited from the Poser model resource made by the Civ3 modders more than ten years ago.

I would like to import this Landsknecht unit model made by Kinboat into Civ6. If we can have proper importing tools (most likely it would not happen, I think).

Spoiler :

 

[SMcM]

Think I might stick around to ensure that at least several iterations of the Isle of Man are released.

Thinking about doing a British Isles civ pack though.

British isles pack sounds cool!

Would civs be something like:

-Ireland
-Scotland/ Picts
-Wales
-Mann
-Cornwall

You could also have interesting leaders for the Channel Islands, like Sibyl Hathaway for Sark and St Helier for Jersey.

 

[zwei833]

Wrong post. Deleted.

 

[Mudokon Pop]

Think I might stick around to ensure that at least several iterations of the Isle of Man are released.

Thinking about doing a British Isles civ pack though.

I was thinking of the same thing - I'll certainly be doing the Gaels anyway - at least from a modelling/textures/art perspective.

 

[Homusubi]

What i'm getting from all this chat is that modders hate change and are very stubborn people.

If you try modding Paradox games you'll understand why.

 

[hangman]

This has meant that modding never really gets established with the Vanilla version of the game - so far! I believe that with Civ 6 a lot of experienced modders will at least try their hand at modding Civ 6 and that we'll hit the ground running with a good selection of early mods available by Christmas.

Thoughts?

I definitely want to get going out of the gate this round. I think I've picked up enough art skills that I can probably start cranking out decent quality stuff early on, and I want to get better at game scripting. However, I'm definitely going to keep one eye on the expansions, since I typically don't work that fast and I see a lot of modders' work ultimately end up obsolete due to expansions. I'll definitely be highlighting the most obscure cultures first. I'm sick of the Mayincatec rotation, so if there's no Chimu/Purhepecha, they're both getting an entire unit/building series, modability permitting.

I've always wanted to see multi-tile, settleable atolls (as a legit terrain instead of a feature), but from what I learned screwing with Civ5's tile files, it would be insanely difficult if there's no support from the devs for hardcore terrain editing.

No one talkig about the unit models modding? I think the art style of Civ6 units are pretty similar to the unit models in Civ3, so perhaps we may be benefited from the Poser model resource made by the Civ3 modders more than ten years ago.

I would like to import this Landsknecht unit model made by Kinboat into Civ6. If we can have proper importing tools (most likely it would not happen, I think).

Spoiler :

Civ units have been fully 3D since Civ4, importing them directly would be impossible. However, previous work is always helpful for when we need some inspiration for new units, so you may find a unit maker willing to make one or two for you, especially if you have mods of your own.

 

[zwei833]

Civ units have been fully 3D since Civ4, importing them directly would be impossible. However, previous work is always helpful for when we need some inspiration for new units, so you may find a unit maker willing to make one or two for you, especially if you have mods of your own.

I mean import the Poser model file of the civ3 unit graphic, not the flc file.

 

[raen]

Since Civ 1 I try to mod it, always with the purpose putting Portuguese in it. It was Hex Editing fun.

Civ II and Civ III were more difficult to edit for me, and at the time I had still few skills, so I left that behind. Then Civ4 appeared, and alot of resources were built (including colonization wich uses same engine), so I had many graphics to choose from later on, I only had to build the metadata and ot modify some graphics. Insertion of javascript language (instead of only dll) also helped.

Civ 5 was a performance disappointing for me and also many few resources available, maybe my expectations got high with so many resources for civ4.

I will be back modding civ 6, and I think civ 6 modding can match civ 4, but it depends on performance, tools, programming language, and most important people liking the game, wich looks that way for what was shown so far.

 

[Thormodr]

Think I might stick around to ensure that at least several iterations of the Isle of Man are released.

Thinking about doing a British Isles civ pack though.

Being of Manx descent myself, you would be a lifetime forum friend if you ever made a Manx Civilization for Civ VI.

 

[Craig_Sutter]

Think I might stick around to ensure that at least several iterations of the Isle of Man are released.

Thinking about doing a British Isles civ pack though.

I'd be first in line stealing the graphics... for dark ages stuff.

 

[Craig_Sutter]

Just a thought... there was some question as to whether Civ V graphics would fit thematically with Civ VI... how about revolutions. Is there a possibility those units/Leaders/buildings etc. could be ported. I never played nor modded that game... but I recall seeing it and it is more cartoonish. Can the graphics files be accessed as we would a pc game? Just wondering...

 

[snafusmith]

I'm a firm believer that a mod community grows when the volume of art grows, primarily units and leaderheads. You can do all the coding you like, but ultimately people want to be able to SEE a difference.

I started making art way back in 2005/2006 when the devs released the tools to make units for civ4. I saw (and in some respects drove) a large growth in mods during and after that time. For several years there was good growth and development in civ4 mods because of the availability of units and leaderheads.

The same thing happened in Civ5, but slower. The devs really never released the tools needed to make units; modders ended up making them themselves. Only once the modding community figured out how to make stuff, did we start to see a lot more (quality) mods.

I believe the same will be true for Civ6. If we get access to tools so we can make units, then we'll see good growth in Civ6 mods. If not, things will go slow.

***Keep in mind I'm a unit modder, so I'm a bit BIAS ***

 

[Deliverator]

I agree that art support is very important.

Firaxis never finished mod support for 3D graphics for Civ V, what they did provide was incomplete, broken and unexplained. If we'd had Nexus Buddy 2 in it's current state back in 2010 there would probably be a lot more 3D art for Civ VI and a lot more 3D artists would have made the switch.

Civ VI can't be worse in terms of 3D support and may be better if XCOM 2 graphics mod support is any indication. Also some of the quotes sound promising:
"A lot of the tools will be similar to what we had in Civ V. [Even] if you're a highly skilled modder, it was extremely difficult to work on leaders, and our elite units in the game, so we're paying special focus to that." I presume this must refer to modding the graphical side as modding the gameplay of leaders and units is not that difficult! As I suggested elsewhere something like the character customisation from XCOM2 for making 3D leaders would be awesome!

What is really needed for units is a way to export as well as import graphics so that new models can be rigged to existing animations or we can adapt existing models. Unfortunately I think Firaxis will be highly unlikely to provide this, but we'll see. They have created a brand new engine which could be an issue. If there graphics format is something unique/exotic then there won't be any tools for importing or exporting it unless they provide them or the modding community hacks them. We'll find out more in a few weeks!

 

[notque]

I think so provided that Civ 6 is an upgrade over Civ 5, and not a new engine. I think especially true if Firaxis makes the positive decision to release the source early. I think we're going to be surprised how similar many things are, which allows us to move quickly.

I think the modding community for Civ 6 may be "boosted" via the use of tools for more "normal" people to be able to mod.

 

[Deliverator]

I think so provided that Civ 6 is an upgrade over Civ 5, and not a new engine.

They've specifically said it is a new engine: "Notably, this is the first Civ game since Civilization 4 in 2004 to launch with a completely new game engine – one Firaxis says it’s designed from the ground up with deep mod support in mind." (link)

 

[Deliverator]

Certainly possible, very interested to see the source.

Firaxis' most extensive comments on mod support so far that I've seen are here:

The team have already teased the potential for modding with this new engine. Can you talk about that in any more detail?

We are going to be talking in more detail about modding further down the road. We haven't announced when we're releasing our tools, it'll probably be soon after release. A lot of the tools will be similar to what we had in Civ V. [Even] if you're a highly skilled modder, it was extremely difficult to work on leaders, and our elite units in the game, so we're paying special focus to that. We are bringing mods to multiplayer as well. As usual, a big emphasis on modding.

I would expect they are planning a big reveal on mod support sometime in the next couple of weeks before the release date.

 

[Quinzy]

I mean import the Poser model file of the civ3 unit graphic, not the flc file.

If that were possible, I would dive headfirst back into unit making.

 

[Gilgamensch]

Im excited to learn for civ 6 and try to plop out new Civs without having them crash my game 200 times.
When you wait 3 minutes for civ 5 to load, 2 minutes for mods to configure, 1 minute for the map to load, and 30 seconds for the game to crash.