In 2011 I started to make a Game Engine (Just as a hobby), at the time CryTek licensing was incomplete and UDK wasn’t the direction I wanted to follow. In Feb 2012, I joined up (or trans-recruited) a far more experienced developer from a telecom’s company that I owned. Who had a lot of experience with game engines, we proceeded to generally make huge strides in terms of getting it towards completion but at some point we gave in and started seriously looking into Unity to develop our games as it had matured greatly. Unity is great no doubt, but we have a slab of tech with 1000’s of hours of time, pain and sweat sitting around doing absolutely nothing. The Engine itself isn’t complete, but contains some powerful in-built features…
Myself and the senior developer met up for drinks tonight and wondering what to do with it all, a lot of passion went into this project… So what would you do? Leave it in the past or License it out to customers and finish it off? It’s not really a competitor for Unity, it only supports MAC and PC for one. With UE4 around the corner, it seems harder to think of reasons to go in this direction if it wasn’t for this:
A) The design of the engine is made for large open environments and does it well
B) It’s heavily focused towards ease of use, because we ourselves are a small team
C) I do have access to many more engineers from our main dev company… As work load balances between the two, I can shift as necessary.
D) It costs me less to use our own engine and it could help increase revenue for future releases.
I’ll do a demo when I can figure out why it decides to blue screen whenever you open the editor. The last thing we did (or tried to do) was change over to a 64-bit editor and it’s not playing nicely…
As a full engine it would be tough, no matter what direction you go. Mostly because you will have continue to sink development costs into it to get it to a ship state. If you have it final and ready to go today, you could probably recoup sunk costs, but if not ready to go, how much more development is required to get it to final state, and can you make that back?
Given these two points, what about the possibility of turning into a tool for primarily those things? Either as standalone tool that generates assets for other engine use, or a modular tool that could be added to another engine (along the lines of Unity pluging, but for other engines). That way you might be able to focus on the unique aspect of what you have built. Just a thought.
Good luck, it is always frustrating to see a lot of work go to waste. I hope you can find a way to leverage it.
Take your existing codebase, and try to improve unity with it. So make portions of your engine compatible with unity (via plugins) and then sell them on the asset store. Assuming it was written in c++ that is.
Good points there: Here is the thing, by the time we get a portion of it complete… New improved tech comes out that does what we need it to do… Also we have been stuck on some system breaking bugs for a long time:
We heavily rely on Nvidia’s Physx and APEX. Last time we did anything implementing Physx 3… Here is the notes list:
Outstanding issues / to do:
64-Bit editor (Causes system crashes, tracing stack call shows potential issue in Multi-threading)
Direct X (DX9 and 11 support) and OpenGL libraries (Implemented, HLSL check needed for continuity in default shaders)
Terrain system (Has issues with slope layers, LOD) (but nearly done)
Data clustered streaming algorithm (Memory allocation issues, needs fixing can cause crashes)
Animation system not yet implemented
Vector based GUI system + Dialogue renderer, still not implemented
Issues with .FBX support, (geometry not rendered correctly)
Performance issues with Volumetric particle systems (Water / fog)
Network capabilities (Not yet tested with multiple third party options)…
Working:
Language support for C++ and C#, works fine… Direct API call and Rpath correctly integrated.
Track animation with cutscene system
Deferred lighting and HDR support
Post effects suite: DOF, BOKEH, Bloom, Motion Blur, Colour correction, DSSDO, SSAO, Edge detection, AA
Real time GI (Voxel based, spherical harmonic indirect lighting works correctly… Needs a performance review)
Distance based Culling
Internal runtime compiler
Nvidia based Physx
Nvidia Apex (destruction tested, works correctly)
I thought Unity Pro was $1500.00 :), thanks for the feedback though… Yes there is a mac editor, there shouldn’t be an issue releasing a free SDK with either a free for non-commercial use or some slightly crippled features like Unity has…
This engine (Originally) was aimed specifically aimed at CryEngine (Crysis 2 Era) with easier workflow…
Now this is the other option, but here’s a couple of caveat’s:
I’ll be releasing a lot of the post effect’s for Unity, as for the rest… Well I’ll be straight up honest, I’m not 100% sure what Unity is capable of inside and out. I know what my own engine does, it’s strengths and weaknesses so I’ll put it down as a maybe.
The asset store is good for revenue anyway, so thanks for the suggestion.
Will do, let me just try and fix a few quite major bugs… Then anyone else who is brave enough to try it can have the full API for free, Disclaimer: If anything happens to your Mac’s or PC’s… This is a warning, it might happen and I accept no liability :)…
I wish I could, most likely this engine will never see the light of day with the stack of issues we or should I say (I) don’t know how to solve… Our first priority is our game in Unity… As a technical flex it was interesting, I’ll let people have a look at it. But whether or not it ever makes any traction, is another matter.
It looks like I’ll try and use some of the source to improve Unity like others have said, so it wasn’t a complete waste of time.
The only engine that I know of that renders large open worlds is Grit Engine. That is open source at least.
But it is very feature incomplete and very messy.
@ShadowK - I would be willing to help up with some art assets to showcase this engine of yours. I would personally like to see it open sourced. But with a licensing model similar to Monodevelop.
EDIT: Just realized something - Cry Engine Still supports Windows ONLY apart from the major consoles.
I think that if you cleaned up the code and made it ship able that you would have a awesome product on your hands.
At the moment this is how it stands, I’ve got to get the dialogue system up and running in our Unity game this week… Then I’ll have one last attempt at sorting it out so people can look at it (If I can fix it), get some feedback and if it’s ok (With some tools disabled) I’ll continue work on it… If not I’ll rip all the tech out / try and get it into Unity.
