Would it be possible to duplicate COX with another engine?

[Kaos Arcanna]

Matt Miller's said more than once that he didn't think that anyone else but Paragon Studios would be able to deal with the spaghetti code that was the Cryptic Engine for COX.

That leaves me curious about something that hopefully more computer literate people can answer--

As the subject heading says, would it be possible if someone did own the IP to be able to remake COX using a different engine?

[navyrayne]

Yup

Anything from Unity, which I've worked with twice in the industry, to unreal, and even designing another engine.

The "spirit" of CoX was the mechanics, not the engine.

[The Fifth Horseman]

The only questions really are
1. How long would it take and
2. How horrible (or not) the resulting code would be

[navyrayne]

You build the game in C++ and C#, the code all the games I've worked on are being built in.

[The Fifth Horseman]

Those questions were rhetorical. :p

[Manga]

As the subject heading says, would it be possible if someone did own the IP to be able to remake COX using a different engine?

The short answer:  Yes, but it would not be trivial.

Longer answer:  The 3D objects and textures would be trivial to import.  The city maps would have to be meticulously measured and re-built entirely in the new game engine, one object at a time, which would take hundreds of hours of time.  Animations would have to be re-done in a format compatible with modern skeletons.  Outdated character models would have to be replaced, and all of the costumes re-done (because none of them use modern rigging).  And then, finally, all of the powers, systems, AI, NPC behavior, etc would have to be re-coded in game engine compatible script.  The powers data, drop tables, enhancements, etc, anything number-wise, would be importable.

There's a lot I'm forgetting, but you get the idea.  It's possible, but not easy, and highly risky because it would be a lot of work just to risk losing everything if NCSoft sues for copyright infrigement.

[srmalloy]

There's a lot I'm forgetting, but you get the idea.  It's possible, but not easy, and highly risky because it would be a lot of work just to risk losing everything if NCSoft sues for copyright infrigement.

The question made the assumption that whoever was doing the conversion had acquired the IP, so NCSoft suing wouldn't be a consideration.

For the static world geometry, though, it should be possible to write code to process the geometry automatically, but that just gets you the polygons; although you have all the textures and can map them back to the structures directly, doing things like the environment reflections that the devs added in Ultra Mode are going to be different in each engine, so what they did might need to be redone by hand, going back over all of the buildings that had environment reflection enabled. Without knowing how both CoH and whatever engine you're replacing it with did the reflections, I don't know how procedurally complex defining reflective surfaces would be.

But you're right; re-rigging all the mob models and their animations and power FX would keep the art department busy for quite a while.

[Eoraptor]

Doing the basic skeletal animation would be... perhaps not easy, but it could be accomplished. however, as others have said, COH runs on many levels. a basic "game engine" which controls the ver-all rules of the world and shapes of the beings in it. then on top of that, is a physics engine. that controls the water, the capes, whether or not you can affect gravity, the ragdoll physics, etc. On top of that are yet more layers of code controlling things such as mob propagation rules, aggro rules, door instances and dialogue timing... and none of that takes into account the layer that controls accounts functionality so that you don't suddenly log in as someone eles's character or on the chinese language server.

in the end you're talking about trying to replace the base layer of a large and complex software ecosystem which saw nine odd years of tweaks and additions. Hence the Spaghetti metaphor. it literally would take less effort to just build an entirely new game, which is what people are doing.

[Mantic]

Anytime programmers try to clone something there is variation; every game aspect is a problem that the programmer(s) must find a solution to, but they can be solved many ways. Even using the same method, barring you're calling the same standard libraries, will wind up slightly different; it could only be moreso if you are working with different libraries and algorithms entirely. So, even if you could eliminate all vanity from restoration efforts (as if), you would almost certainly lose the "feel" of the game in many ways by doing such a conversion or clone.


Regarding the original CoX engine and it's supposed spaghetti code: a lot of those problems had already been addressed, or bypassed, via the implementation of a new script interpreter. It is not so certain that the system as it existed in late 2012 could not be managed and even extended by new people. The last couple of major overhauls were particularly geared to making the engine easier to extend. CoX was finally at the dawn of a golden age just when NCSoft pulled the plug.

The client also need not be updated significantly to avail new content production; the major part of such is just adding data files.