[FVG]: Choosing an Engine: Final Poll

[Astronouth7303]

Since no one seems to want to debate this, I assume that's all the ideas there are. so we have:

1. Crystal Space
   Price:Open source (free)
2. Genesis 3D
   Price:Open source (free)
3. Unreal
   Price:$350,000 plus royalties; 1 platform; $50,000 for additional platforms.

   $750,000 royalty-free; 1 platform; $100,000 for additional platforms

4. Torque
   Price: $100

You have 14 days. Votes are public.

[Tytus Gerrish]

why are people Voting for unreal? do they want to donate all the $ to use it?

[Yan Wang]

Tytus, the question asked what engine SHOULD be used. It says nothing about what WILL be used or the practicality of getting any such graphics engine.

[Heretic121]

the unreal engine is FREE!!!!!!!!!!!! i mean... it comes down to wethere we want to make our own game and make it seperate, or just use the unreal engine and make a MOD for ut2k4... (i have seen a golf sim mod for ut2k4, so n e thing is possible) =D

[ngreen]

I don't have the money for unreal and like open source and think I prefer crystal space. It seems like the best option without shelling out dough. And it can be made so many people, not just thoughs with UT, can take advantage of the game. Just reading opinion this seems to be the path that those actually going to do a lot a work for this project want to go. Not just making a good hack of UT but actually building something. I think crystal space looks like a good building block.

[Denman]

Which unreal engine is this? The totally first one????

[Bcahn836]

I have no clue so i think i am going to stay out of this one. LOL

[Heretic121]

Originally Posted by Denman

Which unreal engine is this? The totally first one????

no... i would persume the 2k4 engine w/ the unreal2.5 engine... BUT... we COUld use UT99, IF it has an Unreal Editor with it... but then it wouldnt be as modern and we wouldnt be able to use high poly count objects =/

[Tristan Lall]

If you guys are really serious about this--and that means that you want to mess with the engine, rather than just making mods (which is fine too!)--Descent and Quake II (or was it Quake I?) were open-sourced, if I remember correctly. Descent is missing some proprietary sound driver code, but other than that, is complete (and runs under DOS or an emulator). Quake, I'm not sure.

But really, a modern engine like Unreal, with a well-developed modding community is probably a better choice, if it can be used without cost. You will have to consider people losing interest in the project (it always happens with mods), and will have to evaluate the technical aspects of what you want to do versus not only your abilities, but also the capabilities of the modding layer of the engine. (What do I mean? You can't mod Warcraft II into Splinter Cell--though for your purposes, a 3-D shooter engine should serve nicely).

Consider which engines or games can be used to approximate real-world physics. Unreal Tournament isn't so great out of the box, but Raven Shield (an Unreal-based game) has rather good physics.

Also consider that these open-source engines, which may not have actually been used in any products on the market, are (by and large) unsupported and unproven. Will they work? Hopefully/probably. Does Unreal (for example) work? Yes. (And can you afford to buy the source code? No. Would they even sell it to you? Probably not.)

But take a good hard look at the potential for people to back out of the project. If the only person who can handle the advanced C++ and DirectX (or whatever you're coding for) needed for a full-fledged game engine quits out of frustration (or petty bickering, or illness, or school, or life in general), what happens? You need to make sure that the project depends on no one person, or that every person will absolutely refuse to quit (which is unlikely, at best). Wouldn't it be a safer bet to mod a game, where there is comparatively little skill required, and where tasks can be assigned to just about anyone? (You probably shouldn't assume that someone who just completed AP Comp. Sci. is able to write--or even read--code to support DirectX, but maybe they'll know enough to adjust parameters in data files, or learn other skills from modding tutorials.)

In short, don't get in over your heads, and you'll be fine. (And if you still want to go ahead with tweaking the game engine itself, just keep these observations in mind.)

P.S. The link to Crystal needs an extra "e" in "sourceforge"....

[Ryan M.]

I'm still all for Crystal. I get what you're saying, but I've already stated my feeling about using less than open-source code...

[Astronouth7303]

Follow the links. The Unreal engine is just the engine; they license it to various groups. (The current version is 2) The prices posted are the normal ones, no special offers.

And if we just mod UT2k4, the cost would be to buy UT2k4 for everyone. Not real cheap, either.

One other thing: There is no sponsor. Anything that we buy is going to have to come out of our pockets. And I don't think that Epic licenses to private people, just corporations.

[Ryan M.]

Originally Posted by Astronouth7303

...
One other thing: There is no sponsor. Anything that we buy is going to have to come out of our pockets. And I don't think that Epic licenses to private people, just corporations.

Remember that, even if you aren't part of the team that works on this...

[Tristan Lall]

Originally Posted by Astronouth7303

And if we just mod UT2k4, the cost would be to buy UT2k4 for everyone. Not real cheap, either.

It really isn't that expensive, and you do get a perfectly good game out of it. (I have of course been referring to modding--it's obvious that you can't afford to buy the code for the engine, sponsor or no sponsor.)
Doing so means that you have a engine that is known to work properly. If you use something else, there is a higher probability of something beyond your control going wrong. What guarantees that it will be fixed in a timely fashion, so as not to inconvienience you greatly? And why not make use of the Unreal modding community? That's arguably better than getting an SDK and reading the documentation, and fighting with low-level code, but still ending up with a product that's based on a sub-par or dated backend.

If the price is such a big deal, why not use the old Unreal Tournament? It's exceptionally efficient, very scalable, runs on anything (of consequence), and is "known good". The techniques to mod it (note--once again, I doubt it's worth it to even think of buying the code!) are established and widely available.

Whatever you do, don't paint yourselves into a corner by choosing an engine for which you will be debugging errors totally unrelated to your project--let someone else do that for you! (Open source is ideologically nice, but do you have the time and energy to bugtest someone else's work in progress, while you simultaneously try to develop an implementation of the same thing, using ever-changing builds, and without a guarantee of stability? It's a crapshoot.)

One other thing about open source (or any source, really): once you do see the code, will you know what to do with it? Not to put too fine a point on it, but that's not exactly easy stuff to interpret, especially the first time around. If you have real expertise (high school comp. sci. probably doesn't constitute expertise), or are willing to invest significant time to learn, then it may be reasonable; but if you want to work on a FIRST video game, and not spend your (not-unlimited) time learning the nuances of a particular engine, the best route is still modding someone else's work.

It's a matter of perspective--what do you really want? To learn 3-D game design/implementation, or to produce a FIRST game? They aren't mutually exclusive, but neither are they the same thing. Another thing: if you're recruiting volunteers, make sure they know in advance your answer to this question--if they assume that you're going to have a game within a certain time period, and find that you, the leaders, are still tinkering with the intricacies of the engine when you "ought to be" working on the game itself, they may just quit out of disillusionment.

[Astronouth7303]

I never said it would be easy, but considering that I don't have Unreal in any form, I am going to say no to it.

There is also the fact that it would be a mod, not it's own program.

If we would to break Unreal somehow, all we can do is hope it's fixable. If we break CS, we can modify the source if need be.

Running unreal would be cool, but the strings atached would make it almost not worth it. Using the Unreal engine would be no more dificult than using an open source engine.

And if we spend any money, a huge infrastructure may need to be set up. I certainly don't have a whole lot of cash.

[Tristan Lall]

Originally Posted by Astronouth7303

If we would to break Unreal somehow, all we can do is hope it's fixable. If we break CS, we can modify the source if need be.

Running unreal would be cool, but the strings atached would make it almost not worth it. Using the Unreal engine would be no more dificult than using an open source engine.

The point is, you can't break Unreal, without tampering with source code that you don't have. If you're confident in Crystal Space, that's not a problem, but be aware that modifying source is what can lead to the most trouble. If you're up to it, and keep backups of every build (so you don't lose as much time hunting down erroneous changes and subtle bugs), you end up with a more flexible and arguably more effective platform. If you're lazy, or lack the resources to be methodical about source control, or any number of disadvantageous situations pop up, you're in for a pile of painful debugging. Also, if someone else's bug is the problem, are you sure that you can fix it? Are you sure that they can fix it (in the near future, without invalidating your code)?

With an open-source engine, you're not coding for a mature platform--you're coding for something that's in development. That's a lot harder than modding a well-known, well-documented, versatile, and stable game.

Originally Posted by Astronouth7303

I never said it would be easy, but considering that I don't have Unreal in any form, I am going to say no to it.
...
And if we spend any money, a huge infrastructure may need to be set up. I certainly don't have a whole lot of cash.

It isn't expensive to buy Unreal Tournament ($20 CAD in the old games section), nor is it particularly expensive to buy Unreal Tournament 2004 ($60 CAD). Not free, but not all that much cash either. It is not a huge infrastructure. (To reiterate: you can't buy a copy of the source without a corporation and suitcases full of money. So it's irrelevant!)

Originally Posted by Astronouth7073

There is also the fact that it would be a mod, not it's own program.

Yes it would be a mod, so no, you couldn't sell it. (Were you planning to sell it? Who would buy it? There aren't that many FIRST people, you know, say 30 000, most of whom will not buy anything of the sort. And if you intend to convince the public to buy it, that's excellent in that it provides exposure for FIRST and for you, not to mention money; but are you sure that you can pull it off? Maybe that's just a wild extrapolation, though; disregard it, if you had no such intentions.) More to the point, what fundamental objections do you have to modding a game?

Originally Posted by Astronouth7303

Running unreal would be cool, but the strings atached would make it almost not worth it. Using the Unreal engine would be no more dificult than using an open source engine.

So what's the problem? To which strings are you referring? You're implying a similar level of difficulty, so either you're intending a (relatively simple) mod for Crystal, or you're making minimal changes to its code. (In other words, why do you need/want the source code?)

Or is it that you really want to use SourceForge, but can't justify it if it's a commercial game engine? I hope it isn't something as miniscule as that.

We're still not really clear on one thing: are you focused on getting a "product" (saleable or not) out in the world, or are you concentrating on developing your own skills? I'd tend to agree that the more challenging option (Crystal Space) is better for skill development, but that as a practical matter, you're more likely to end up with something substantial to show for your efforts if you mod Unreal (or the like).

And like I said before, since you're promoting this project vigourously, and recruiting help, people are going to be disappointed if you find Crystal Space to be too difficult, or too limited, or otherwise unsuitable. By opening this up to the world, you're now bound by their expectations, at least to some degree.

All that said, give it your best shot--who knows what will come of your decisions, but at least you'll have learned something....