Engine

[Madball357]

this looks okay.

yet I hope they will use a modified version of GemRB

http://sourceforge.net/projects/gemrb/

features not supported:

Hardware acceleration

hmmmm

That's the Infinity engine dude, I would be glad if they used that for Wasteland 2.

[dmazz]

Some questions: Just how important would tech support be for an engine given Wasteland 2's short 12 month development time? The Unreal 3 engine appears to have the best support so perhaps it justifies the expense, if support will be used often. For choplifter HD, fargo says his team was fairly self sufficient when it comes to working out engine programming issues, but he did use their support nonetheless.

I'd also like to know how important middleware is to a project like this. Speedtree for example would make vegetation creation a cinch and there are tons of other high end middleware that will provide the tools for rapid creation of content. So is an engines support for middleware a big factor in this case given the short development time available?. Will the short time available push them towards using alot of middleware products?

I now think Unreal 3 is gonna be the engine, doesn't matter that it's the most expensive option. For a few reasons - the support available for questions, recent previous experience with the engine, extensive middleware support and no upfront costs. linux/osx porting ain't a problem. Unreal 3 engine may be overkill and really really expensive in the long run, but it has no equal in fast development speed, crucial for an RPG that only has 12 months to get made.

Unigine has a nice renderer and easy basic tools to get up and running but scarce middleware support, scarce technical support or docs given the engine is 2.5 years old. So for any hardcoding you've got no support, and little middleware support forces you to build your own tools or use their inadequate ones. Increasing development time. Unigine is also not used much with big complex games yet so they have no experience providing support.

The best option seems to be getting a engine with decent graphics, tech and doc support, (Unity engine graphically would be the minimium) then using heaps of middleware (physics, sound, scripting, animation, artificial intelligence. lighting) for fast content creation. It would be expensive initially but be much cheaper in the long run than using a AAA game title engine like CryENGINE, idTech or Unreal.

Gamebryo looks like a good commercial canditate. It's got good support, documentation, decent graphics and alot of middeware support. Gamebryo 4 was released this month so the company is still kicking, and supports opengl for linux and mac. Fairly cheap at 30K for it's source I've heard. With the best middleware money can buy the total engine layout will be between 50-70 thousand dollars. Not cheap for a game with this budget, but not expensive either. In addition Fallout 3 uses the Gamebryo engine and Brian Fargo might have some contacts in Bethseda who worked on Fallout 3. Or if not who are diehard Fallout 1+2 fans, who'd love to work on a project like Wasteland 2. They will know the gamebryo engine intimately having worked with it for years, shortening the learning curve period for his team.

An unlikely and unorthodox approach would be to get the source code for a rpg+turn based combat game and do a total conversion into Wasteland 2. Possibly needing to recruit one of it's ex-programmers to serve as technical support for the project. Some canditate games for this would be Neverwinter nights 1 or 2, Star Wars Knights of the old republic basically all the games based on the Aurora engine which also supports Linux. Bioware as a publisher might be sympathetic and receptive since this may be Biowares chance to get into the post apocalyptic genre.

Another possibility is Interplay gives them the source to The Bard's Tale which InXile developed for them. That game engine may be too limited though, I'm not even sure if it's combat is turn based.. But Inxile are familiar with the engine and that will speed up development.

I'll add that real time with pause turn based can easily be modified to traditional turn based. What matters is that the hardcoded number crunching is happening in the background. Another candidate is the European games based on the silent storm engine, the most prominent being Jagged Alliance 3, stuck five years in development now. This is a long shot cause they live on another continent, but love for turn based combat runs deep and Silent Storm/Jagged Alliance 2 are now not only the deepest examples of the genre but related in a business and development sense. A Wasteland 2 RPG using either games source code in a total conversion will be seen as complimenting not competing with Jagged Alliance 3’s urban combat sim when it get’s released. So they may get a warm welcome to this idea.

The longest shot candidate would be getting the source code for turn based combat RPG games which use the german Nebula 2/3 engine, which is open source.The game source code I have in mind is a series of games that won best RPG game of the year in germany. Also seeing as the publisher and developers live in Europe, hiring a programmer familiar with the engine will be very difficult. Also the Nebula engine being maintained or supported right now by a publisher is questionable.

[iMik]

What about Obsidian Entertainment Onyx Engine.

[temotodochi]

Just a short GPL reminder for people:

using a GPL engine would not mean that the whole game would need to be GPL as well.
Only if the engine is modified in any way, those changes would have to be released.

GPL license is quite detailed on such matters and i would recommend to read it, just once and you're set with GPL.
You can sell software which has GPL license. Just the parts licenced with it must be freely available as well.

Sorry if this was already addressed. I read through some 10+ pages and didn't see it.

--
temo

[geezer]

What about Obsidian Entertainment Onyx Engine.

Already suggested multiple times. Doesn't Square Enix own the rights to that?

[Brother None]

What about Obsidian Entertainment Onyx Engine.

Already suggested multiple times. Doesn't Square Enix own the rights to that?

Heck no, Obsidian owns it. I just don't know if it's well-suited/documented enough for usage by another studio, especially one with a short dev cycle.

[geezer]

Well if Obsidian owns it... That seems like a good candidate for a first shot to me. Obsidian will probably be friendly to kickstarter projects. It might be possible to license the engine for a reasonable price. I'm not sure how Brian and Feargus or Chris would get along though. They might hate each other or something. I do think the graphics in DS3 are beautiful. It's an awful game, but the graphics are much better than we have any right to expect from Wasteland 2. I keep hearing conflicting views about the difficulty of adapting real time engines to turn based play. So I don't know what to think about that. But I certainly hope inXile at least considers the idea.

[dmazz]

Well my opinion on the matter is turn based combat, whether traditional or real time has a complex, rule based system underlying the hack and slashing on the screen. Meaning if you change that complex combat system you change the nature of combat. Action real time games though have none of that, their combat engines are based on mainly collision detection and reflexes. And let's not forget A.I, far more complex in real time turn based games. All this results in action hack and slashh games offering little to nothing for a turn based RPG even though they may appear to resemble one superficially.

Changed my mind about the Unity engine now that I've read more about it. Although it's graphics aren't as good and neither is it as featured, it's alot cheaper than Gamebyro. The only thing that concerns me is it's middleware support. As I now consider GUI and A.I middleware essential. Other great middleware available to speed up development would be for FX and sound both rich immersion tools that should be used extensively in this game. And implementing weather effects would be greatly sped up with vegetation and physics middleware. Scaleform a GUI middleware comes with a UI 'kit' for MMO games. If used a good chunk of the GUI work has already been done, great.

Fargo said in a recent interview . http://www.nma-fallout.com/forum/viewtopic.php?t=60856

We do have the advantage of this being a top down game which saves tremendously on the art creation which in turn allows us to script out numerous outcomes without the concern of creating graphics for every possible situation. Artists can spend months on a single 3D model in a 1st person game and that would make a lower budget title impossible.

From the sound of Fargo in this interview, where he seems to indicate his knowledge of how time consuming 3D can be and him previously mentioning top down graphics. I suspect the graphics will be 2D top down view/isometric view world exploration and combat, but the engine will be in 3D. Perhaps there will be animation of characters, but I doubt it, most characters will be in 2D with the bare minimum in animation. This will save him untold hours of time in art creation, as all he will need is one drawing for the background in an area. And a few drawings for a character. The use of a 3D engine will enable him to use the latest tools and knowledge to speed up development, and with middleware rapidly add advanced immersion enhancing things like weather effects, explosions, dust, lighting, complex sound effects and vegetation.

The game will look and function (except in combat) like Commandos 3: Destination Berlin. More top down than isometic 2D background. Sparsely animated characters. A sprinkling of 3D art assets, and 3D effects applied over it all to give it more mood. http://www.youtube.com/watch?v=x9sUarEh ... re=related And the game will somewhat resemble the original Wasteland in spirit too.

This also explains the importance of the 6 month preproduction stage. In that period they'll design and create all the art in the game cause it will be in 2D, including all the GUI's. And also design and balance the game mechanics. All this completed, they will then start development and immediately begin hardcoding the most important game mechanics, while the art gets digitized, placed in the game and enhanced with 3D effects. (lighting, dust blowing, leaves rustling, branches swaying, insects buzzing stuff like that)

So we might be surprised at how simple the graphics are, but they will enable Fargo to concentrate on the gameplay, story, game mechanics and combat system as much as possible. Which is after all the core of this game, and not worry about graphics quality. The use of a modern 3D engine will make the game visually, easily upgradeable in the future though.

I guess now with the modest graphics, use of the open source infinity engine GemRB is a possibility. But I doubt it, the engine is too limiting in it's tools for the game mechanics envisioned for Wasteland 2. Requiring hard coding anyway, which will be time consuming given the open source not well supported nature of the engine.

[Xiao]

From the sound of Fargo in this interview, where he seems to indicate his knowledge of how time consuming 3D can be and him previously mentioning top down graphics. I suspect the graphics will be 2D top down view/isometric view world exploration and combat, but the engine will be in 3D.

That's what I thought at first, that he meant he's going 2D, but when I read more carefuly he said:

Artists can spend months on a single 3D model in a 1st person game and that would make a lower budget title impossible.

Note the "1st person game" part, I think Brian meant that character models in Fallout 3 is going to be way more complicated than for example unit models in Warcraft 3, where you're looking at them from a top-down view. So he's not ruling out 3D, just complicated 3D models.

Plus, later on in that same interview you quoted he explicitly said he's not decided on 2D/3D yet, and the only thing for sure is they don't have time/budget to develop their own engine.

[geezer]

Fargo's statement doesn't make a whole lot of sense to me. I can't figure out whether he is leaning toward 3D or 2D. The first sentence claims the advantage of 'top down' is less art creation because there are fewer "situations". Keep in mind that he seems to use "top down" to mean "not first person". But perspective has nothing to do with 2D vs 3D art. In fact some of the first graphic adventure games were both first person and 2D (i.e. Wizard and the Princess or Cranston Manor).

If anything I would interpret the first sentence to imply that he is leaning toward 3D, because the main advantage of 3D is that you don't have to draw multiple perspectives for every character. There is a lot of semi-repetition with 2D art drawing bitmap after bitmap which represent only an alteration in perspective with each drawing. With 3D you draw a model once and that's it. All the viewing angles are taken care of. Yet his second statement makes it sound like he is leaning toward 2D art because he seems concerned with how much time it takes to make a nice looking 3D model. Definitely confusing.

As far as being able to use a real time engine for a turn based game, I don't think an "opinion" is particularly useful. For a particular engine it is either true that you can use it for turn based combat or it isn't. I've never used a game engine myself, but I believe the basic idea is that it is a library of objects or functions. So it certainly seems plausible that you could simply use the functions that are useful and ignore the ones that relate to real time combat. It may also be possible to call an engine function from within your hand coded TB combat functions. That may depend on the particular engine.

As far as 2D art being less work, some apparently experienced posters have claimed that it is just the opposite. That 3D is less work than having to draw so many bitmaps for each character. That seems plausible enough to me. I do know that 3D art and 2D art require very different artistic talents, although both require an artistic imagination. Pretty much any traditional artist would be capable of 2D drawings. 3D art requires a specialized workflow involving modeling and texturing software.

I'm not sure creating art assets would be considered "pre-production". That sounds like regular production to me.

GemRB does run on Windows and OS X and Linux. It was already discussed in this thread and there was a strong objection to it. Perhaps it was the GPL. Some people are saying that the GPL shouldn't be a problem. So perhaps that is a possibility.

As far as the Onyx engine does anyone know how portable it is? It seems to support consoles, but it probably doesn't support Linux or OS X. Too bad because the DS3 environments really are pretty. Particularly indoors.

[Xiao]

Read through all the previous post (not gonna do that sort of thing ever again)....

Just feels like a mess.

You get expensive, budget sucking 3D engines, and you get open source/cheap crappy engines.

My first thought was the heck with it and just use GemRB and rely on pre-rendered / painted graphics instead of trying to impress with engine capacities, forget 3D and focus on delivering a great story + good game mechanics + nice 2D graphics.

Then I thought, what about other recent indie games with decent graphics?

In the interview here about Gaslamp Games (Dungeons Of Dredmor), apparently they rolled their own engine. Even though it's "held together by duct tape", the end product looks great. I wouldn't mind that level of graphics - though of course different art style.

And if you focus on 2D, there's more choices out there besides open source engines, there are also open source games which has their own custom engine that might be able to be used for Wasteland 2.

Off the top of my head Wesnoth's engine might be able to be adapted for this, Aquaria (from the first Humble Indie Bundle) is also 2D and open source, there are surely more games like those that's open source but might not be considered at first glance because they are not themselves "game engines" but have the potential.

[Bryce777]

Dungeons of dredmor I hope you're joking. There's plenty of 2D stuff out there, if they go 2D don't even worry about it.

GemRB is also pretty well crap mostly because the IE is complete crap, but it doesn't even properly do everything that the IE did. It has a really terrible scripting system, too, and there's no support for turn based.

There's a big difference between general engine and game code. The stuff like combat and tools and all that is going to be useless for making an all new game unless it's almost exactly the same as what you're making.

[Badunius]

http://youtu.be/8MMaA8z0zsg

[destraht]

I would love to see Wasteland 2 use the open source FIFE Engine (Flexible Isometric Free Engine). After all it was developed to produce Fallout like games, so it is quite fitting. There are already three games using it and it supports Windows, Mac and Linux. It has qualified for the Google SoC (Summer of Code) at least a few times already. It is licensed under the LGPL which means that it can be used commercially. I'm proposing that if the project gets to $3 million that inXile open sources the code and asssets in 4-5 years. That way this project will forever be a part of our culture.

Fife Engine Features
http://wiki.fifengine.net/Features

Can I use FIFE to make a commercial product?
http://wiki.fifengine.net/Frequently_answered_questions

The following basic rules apply concerning the used LGPL:

If you modify the FIFE C++ core and release a binary based on it, you'll need to release your modifications under the LGPL.
If you modify Python scripts that ship with FIFE you'll need to release your modifications under the LGPL.
If you just link your custom code against the FIFE C++ core (libFIFE) but don't modify the FIFE core itself, you won't need to release your code under the LGPL.
If you write your own Python scripts but import functionality from FIFE Python scripts you won't need to release your custom Python scripts under the LGPL.
You can use proprietery libraries in combination with FIFE if you don't integrate their code directly into the FIFE C++ core or the FIFE Python scripts.

I've been watching the FIFE Engine for years now hoping that I would get another Fallout type game.

Here is an example of the FIFE engine (Flexible Isometric Free Engine) at work
http://wiki.fifengine.net/File:Zero_pro ... shot_1.jpg

It sure looks like Fallout to me. Slap some proper assets in there and you are looking at a real game. Better yet the Zero Projekt post-apocalyptic assets can be used as place holders early in the development process.

[Mort2]

Isnt an open source engline means, they will have to release their source code....

[Elerond]

Isnt an open source engline means, they will have to realize their source code....

No, If they use LGPL 2.1 licensed engine (as proposed FIFE is) they must only publish changes what they have made in engine. But all code which only uses engine (which is propably most of the code) can use any license and publication form what they want to use.

But I would not use FIFE anyhow because it is not mature enough to be valid choice.

[mmx]

I would love to see Wasteland 2 use the open source FIFE Engine (Flexible Isometric Free Engine). After all it was developed to produce Fallout like games, so it is quite fitting. There are already three games using it and it supports Windows, Mac and Linux. It has qualified for the Google SoC (Summer of Code) at least a few times already. It is licensed under the LGPL which means that it can be used commercially. I'm proposing that if the project gets to $3 million that inXile open sources the code and asssets in 4-5 years. That way this project will forever be a part of our culture.

You're joking right? Fallout 1 looks way better than this!

[Elerond]

I would love to see Wasteland 2 use the open source FIFE Engine (Flexible Isometric Free Engine). After all it was developed to produce Fallout like games, so it is quite fitting. There are already three games using it and it supports Windows, Mac and Linux. It has qualified for the Google SoC (Summer of Code) at least a few times already. It is licensed under the LGPL which means that it can be used commercially. I'm proposing that if the project gets to $3 million that inXile open sources the code and asssets in 4-5 years. That way this project will forever be a part of our culture.

You're joking right? Fallout 1 looks way better than this!

That is because they had better artist than team which has made examples for FIFE.

But still there is better engines for this game, which are more mature and which have better tools and which give more flexibility for developers.

[Prosper]

I would be willing to make the entire engine. It would not be proprietary. Not entirely at least. I am an experienced programmer and have probably the best use doing an engine for this kind of game. The only pay I would require is 500 USD a month. Or 25USD each day for pizza.

Joking aside, I am willing to help out with tool creation. If you pick an obscure engine or a well known but underused engine (like FIFE) I would be able to devote myself around the clock to creating your toolset. Payment in pizzas would be nice though..

Sincerely,
Prosper

[Badunius]

I am an experienced programmer...

prove it