Daggerfall Workshop Forums

This is mostly a question for Interkarma, as he knows so much about the internal workings of Daggerfall, but I would also appreciate answers from those who have helped with Daggerfall Unity...

If you were to create a spiritual successor to Daggerfall - or a game of a similar size and scope - that included new and original content (so you wouldn't need to use Daggerfall files), which game engine would you choose and why?

Welcome to the forums.

I think any of the major engines could do this easily, at least within an indie scale. With the right amount of care and attention to detail, most users probably wouldn't know or care which engine its running on.

I had some good reasons for choosing Unity for this project and I will probably use Unity again for my next game now that I'm so comfortable with it. But if someone was starting from scratch on their own game, they could choose whichever engine they liked. Engines by themselves aren't particularly interesting - they're just a toolkit. It's what you build on top of the engine that really matters.

When I last compared the most popular game engines a couple of years or so ago, only Unreal Engine 4 could handle massive terrains by loading and unloading each section when needed, and resetting the world origin to avoid floating point inaccuracies.

Cryengine was going to add a similar feature, but I didn't see any mention of it in the feature list on their site when I checked a few weeks ago.

Most of the game engine sites only mention being able to handle "large terrains" without mentioning how large they can actually be. Even when they use the word "unlimited" it generally turns out that they would have problems handling a map the size of London, let alone England.

Did you have to modify Unity's source to allow it to handle Daggerfall's map size or was it done with scripting alone?

When Daggerfall Unity is completed, you should consider writing a book on how you did it and what you learned. It would be an invaluable resource for so many people.

Wysardry wrote: ↑Sat Jul 07, 2018 4:21 am Did you have to modify Unity's source to allow it to handle Daggerfall's map size or was it done with scripting alone?

It all done through scripting. The streaming world setup handles all the relevant parts like paging terrain tiles in and out, object pooling, dual coordinate system to fix floating origin, and stitching adjacent terrains together at the edges.

I'm happy enough with the setup for this project, but I'd do a few things differently if I was starting again knowing what I know now. Will get a chance to do better on my next large game after this one. But I might do a couple of smaller things first just to decompress.

Unity has the best GUI among the ones I've tried, and is the easiest to expand among those, AFAICT.

OTOH, it's also the priciest, overall, since you *need **plugins with it, and a subscription doesn't get you access to the source code.

* Technically, theoretically, you don't; but practically, you do; either to buy or code your own.
** Code that expands the functionality of the Unity Engine, rather than your game, which is admittedly a bit of a hard line to draw.

I'm kind of leaning towards UE4, as it has included world streaming features for longer. The only downside to its system is that it doesn't support world origin shifting in multiplayer games as standard.

If Unity's large world system does support multiplayer "out of the box" that would be a big plus, as I'd like to allow up to 8 people to join the same game for cooperative or competitive play.

My only problem with UE4 is that all the game look the same on it. It has no soul. That is ofc very personal opinion.

Though I'm not a coder, it's still interesting to see what all Chris Roberts and ClG are doing with Lumberyard. I'd not even heard of the engine until last year. Before that, they were using CryEngine 3, if I recall correctly. But I guess they decided it lacked the flexibility they needed to develop certain in-house tools. Maybe Lumberyard is more modular, flexible, or just bare-bones so they can build the tools they need to tackle that massive project of theirs. (whether or not it ever makes it to becoming a full game isn't something I'll go into here, suffice it to say I'm neither a fanboy nor a naysayer, but keeping a close eye on the development).

Lumberyard is a modified cryengine that has been licensed out to amazon from what I understand

If you were to create a spiritual successor to Daggerfall - or a game of a similar size and scope - that included new and original content (so you wouldn't need to use Daggerfall files), which game engine would you choose and why?

There is not sufficient detail in the question to provide an answer.

What level of detail are you aiming for? What kind of project - commercial or open source? What kind of features are you looking at?

Have you investigated the DFTU source? Would a modded version of it provide what you need?

Have you looked at OpenMW?

You need to lay out a proper use case instead of just saying, "Daggerfall successor." If it is just a Daggerfall successor then I'd say you are in the right place already as this engine can already drive a game the size and scope of Daggerfall, and improve on it from there. You just need different content.