Other > Game Development

The thread of programming questions, advice, and secrets.

<< < (2/4) > >>

If you're going to study OCaml, then i'd definitely go with F# instead. Nicer syntax (less rigid from what i hear), and you get access to the entirety of .NET (and you can use VS which is the best IDE bar none).

I've looked at it briefly from time to time, and it seems like the sort of thing that has a lot of potential to me.  My other favorite language other than C# is TSQL, and I feel like F# takes a bit from both of those existing languages and mixes them together with some completely new concepts.  I'd have to really do some nontrivial work in it before forming a full opinion, but at the moment there hasn't been a compelling reason to do so.  I'd hate to think that I'm becoming a programming stick-in-the-mud already, but I don't yet have a case for anything that can be done in F# that I can't do just as fast with C# at the moment, with my library of past-developed functionality and my history with it.

In that case, you might be interested in http://stackoverflow.com/questions/2785029/in-what-areas-might-the-use-of-f-be-more-appropriate-than-c :)

What do you think of open-source engines like OGRE?  OGRE was used to create torchlight which was a fairly popular dungeon crawler available on Steam, and a full 3D game.  Could such an engine be used for 2D games similar to how your using the Unity engine?

(Note:  I don't bring it up to persuade you to move away from Unity by any means.  I like Unity and it's obviously done well for you with Tidadis.  I'm just wondering if you have any thoughts about it, though you most likely haven't looked into every obscure engine.  :)  )

Sigh, the forums apparently went through a couple days where they did not notify me of some threads, which is really annoying.  So I'm only seeing this now, I wasn't avoiding it.

At any rate, we evaluated the following platforms when we decided upon Unity for AI War to move to, and to use for Tidalis, plus a couple of others:

SlimDX / .Net 3.5 (current platform)
OpenTK / Mono
Torque Game Builder
Unity 3D
Torque X 2D
XNA Framework
Tao Framework / Mono
Blitz3D SDK
FMOD (Audio Only)
TrueVision 3D

For all of the above except for Unity, there were various serious drawbacks for Arcen specifically.  One or more of the following was true of everything on this list except Unity, in the end:

1. Overly difficult/long installation process.
2. Buggy and uneven support.
3. Lacking in the latest features and/or not frequently/recently updated.
4. Poorer performance than we were willing to accept.
5. Funky workflow requirements that we hated (Unity actually had this one as well, but we managed to work around that).
6. Lack of support for platforms beyond the PC.
7. Lack of support for productive/powerful programming languages and scripting (we are biased toward C#, but could have accepted an equally-functional alternative with a similar level of verbosity).
8. Price (Unity is $1200 per developer, but some were far more than that).

I think there were also a few others that we considered, like the LOVE engine for example, but some of them were so early in development and had such limited functionality at the time that they didn't even make it onto the full consideration list.  But, some of those -- LOVE specifically -- is looking like it will be really great for certain games in the future.  We also investigated FLEX to a heavy degree, and discarded it for various reasons.

I was nervous as heck about Unity, and I hated their way of loading assets and their workflow in general with an intense passion upon first using it.  However, I managed to completely bypass that style of workflow within Unity and approximated something more like what we had with SlimDX, except even more efficient and effective.

I can't stress enough how much I love SlimDX, I'm really sad to be moving away from that.  It's a great library, and if it wasn't for it's issues with #1 and #6 above, I'd want to stay with it.  And #1 really isn't its fault, it's more of a .NET 3.5 thing. 

My impression of OGRE and SDL.NET and most of the other open source C# ones was that they were at various levels of polish and completion, and that many of them haven't been updated much in the last 2-3 years.  I don't remember which ones were which, now, and certainly the C++ versions of several of those were much more active and had larger communities surrounding them.

In the end, everything on that list is a great product for certain purposes, or many purposes, as the case may be.  But, we needed something that matched with our really specific needs and our established (and semi unusual, but super effective) style of workflow.  So just because something doesn't work for us, doesn't mean it won't work for you.  Some of it even boiled down to us having this focus on 2D in 3D, a lot of those products would have been much better for 3D purposes, which is what many of them were designed for in the first place.

I couldn't be happy with how Unity turned out, though, and at this stage I love it pretty much the same as I do SlimDX.  Both really solid products.


[0] Message Index

[#] Next page

[*] Previous page

Go to full version