engine for outdoor environments

Ryno

I've got a little game design in mind, and was wondering which engine you guys would recommend, taking my criteria into account.

1. Must easily handle different player classes, where classes have special abilities, different armor, footspeed, etc.

2. Be able to outdoor natural environments well. Either handle lighting well in engine, or make use of lightmaps. Good water a plus.

3. Good particle effects support.

4. Be able to support numerous characters on screen at the same time. (20+)

5. Game should scale well to play on older PCs with reduced graphics quality.

I'm aware that a lot of this stuff hinges on custom-tailored code, and just good, pretty, smart use of art assets. But some engines would make these tasks easier than others. Please make suggestions based on your production experience with any licensable engine. Thanks!

lkraan

[ QUOTE ]
5. Game should scale well to play on older PCs with reduced graphics quality.

[/ QUOTE ]

How old and with what kind of graphcis possibilities?

On not-so-old-pc's there is the Battlefield 1942 engine which supports big landscapes and a large number of characters on screen.
The UT2003/4 engine does that too I think.

An older game (engine?) that I can think of and that does this is Tribes 2.

Ryno

Not ancient, but it should be able to play on a 3 year old mid-range system. As of now, this would be about a 1.2 GHZ machine with a Geforce 3 or so. Given, it a player running with this wouldn't get fancy shader and water effects, but they should be able to play at lower resolutions.

Yeah, I was considering the Battlefield engine, with the Vietnam variant in particular. Of course, if the Battlefield 2 engine scales well...

cep

Torque Game Engine sounds like a good choice, wich was used in Tribes 2.

It's only 100$ us.

And it meets all your criteria.

JKMakowka

Nahh, Torque is really a bit old by now (although it has a nice network code that was GPLed recently), so if money isn't that much of a concern than go for the UT2004 engine.

I definitly wouldn't advise to use the Battlefield engine, because it is a buggy pice of &%$§# and doesn't run very well on slower computers (even though the graphics arn't really that great either).

If you are on extreme low funds you could prbably also use the open-source engine "NebulaDevice" which was used in the fullprice game "ProjectNomads" two years ago I think. (but it has improved since then).

Edit: Of course there is also the FarCry engine, but that one doesn't scale very well

Ryno

Price really isn't a tremendous concern right now, I'd really just like the best engine. What I am trying to do is to polish up my design, and maybe make a test level with extremely limited, or no gameplay. Really this would just be used so that I could make a better pitch of the idea to my company. So any in-game proof of concept beauty shots will be very helpful.

So the engine needs to handle my criteria, must look good, and needs to be modable so that I can do mock-ups of stuff, preferrably without having to pay for a license ahead of time. If I get the go ahead, then either my company or our publisher could handle getting the engine license.

Hey JK, what specifically do you find to be bad about the Battlefield engine? My boss was one of the leads at Dice (who developed Battlefield), and I know that he'll be suggesting that engine if this little project gets the greenlight. I'd like to know what issues this might cause me ahead of time, if you could fill me in. Thanks.

AstroZombie

Torque:
Player Classes - check
Outdoor Environments - check
Particle Effects - probably dated but "check"
20+ players on screen - check
scalable for older PCs - check

Unreal:
Player Classes - check
Outdoor Environments - check+
Particle Effects - check+
20+ players on screen - possible, but most likely pushing it
scalable for older PCs - probably not

BF:
Player Classes - check
Outdoor Environments - check
Particle Effects - check
20+ players on screen - check
scalable for older PCs - better than Unreal, not as good as Torque

I'd say Torque is your best choice for all of the features above.

eepberries

Perhaps the Torque Shader Engine? It's a new adaptation of the Torque engine with new shader things added, it sounds pretty promising. The original Torque engine was able to handle outdoor environments pretty well . An example of the original engine's application is Tribes 2.

TSE:
http://www.garagegames.com/pg/product/view.php?id=28#screenshots
http://www.garagegames.com/pg/product/view.php?id=28

kleinluka

yeah torque has shaders, torque has this, torque has THAT. Doesn't change the fact that it looks outdated and ugly. Move on, people!

Ryno

That would have been my kneejerk reaction to Torque as well. If anyone can point out a really good looking game that has been made using Torque, please do so. No disrespect to Tribes intended, as it has really fun gameplay, but even when it was new it didn't look particularly stunning graphically. Perhaps if it was just poor artwork that was making it look worse, and not the engine's rendering that made it look bad, I would consider it if it's easy to work with. But then again, the same thing could be said about the Half-Life 1 engine, as they effectively did Team Fortress Classic with it, which is in line with my goals, but just doesn't look good enough.

At this point I am leaning slightly to UT, as I have some experience with it, it looks good, and has good mod support. Sure it might not scale to my specs now, but two years from now, it should still look good and run well on last year's mid range systems.

Of course I have yet to make a firm decision, so if anyone has more suggestions, plz shoot them my way.

AstroZombie

What about Source? If what they say is true then Source should be very scalable.

Otherwise the Unreal Engine is always a good choice. It does have very good support and documentation for mods.

ndcv

I don't see why people are saying UT doesn't scale well - I have one of your "older spec" systems (1.3Ghz, GeForce2 Ultra) and UT2k4 runs great on my machine. Probably doesn't look as nice as some other systems, but it runs just fine. Sometimes in huge onslaught maps you can notice it start to slow down a bit, but still totally playable.

Ryno

ndcv, that's good to know. Thanks. I hadn't ever tried it on a system that old, so that's good info.

AZ. Yeah, I'd definitely consider Source, as from what I've seen it looks great. And knowing that they'll hopefully port TFC over at some point is encouraging to me. My only concerns are that the engine has not been seen in a real world environment yet, and that there may be some lack of documentation initially. But if it lives up to the hype graphically, and scales reasonably well, it would be at the top of my list. Hopefully HL2 will ship with their editor so that we can all test it out. Anyone knowing if this will be happening with the release?

Badname

Ut2004 works better on older pc's than BF. Bf engine code is REALY REALY B A D even on many good pc's it gives a shocky fps.

JKMakowka

[ QUOTE ]

Hey JK, what specifically do you find to be bad about the Battlefield engine? My boss was one of the leads at Dice (who developed Battlefield), and I know that he'll be suggesting that engine if this little project gets the greenlight. I'd like to know what issues this might cause me ahead of time, if you could fill me in. Thanks.

[/ QUOTE ]

Can't really give you exact error descriptions, and I didn't do any kind of mod work with it, but from a players point of view:
1. Runs quite bad considering how medicore the graphics are
2. Netcode: where is all that traffic going??? (lots of people have a volume limit)
3. Had quite a few freezes, exspecially with earlier versions (and frigging 100 something MB patches arn't really great either)

Sa74n

the unreal 2.x engine has already been successfully used in a similar project to what your expectations are - take a look at lineage2 http://www.lineage2.com/.

otherwise i'd recommend contacting valve on the possibilities their source engine offers.

Ryno

Very nice, Satan. Unreal looks like it would work fine.

Since I'm just doing preliminary design and R&D, I wouldn't want to contact Valve just yet. Seeing as how I work for a very small developer, it would be nice to be able to just mod the demo first, then license the engine once we had something substantial to show the publisher.

Anyone deal with licensing from Valve in the past? If so, comments?

Downsizer

The only issue with torque is the lighting model, which can be replaced with a lighting pack created by synapse gaming or the sorts. I have no idea why people put torque down for graphic capability, the skill of your artist is what defines how it will look. They just dont have a dedicated art team.

If your making your own game, you would have to use torque or another inexpensive engine like quest3d. Unreal is probably out of your price range for a commercial title.

Ryno

This would be a commercial title, as I work for a small development studio. It's just that I don't want to have to license the engine for initial proof-of-concept, but would like to be able to mod some of this gameplay together so that people could really see what I'm shooting for. If I could get even a little bit of this modded on my own, I can make the pitch to the art director and manager here, basically showing them "Look, it's not that hard. I even managed to get this done on my own time."

This would help to show them that with very little financial commitment on the studio's part, (other than a few people's time) we could put together a little game demo, and would help to encourage them to let me make a run at this little project. Once we had a little demo and some more artwork to go along with my design docs, we can make the pitch to one of the publishers that we work with, try to get the engine properly licensed, and get some more funding for the project.

My goal right now is to just figure out the engine that would be best for the gameplay that I envision, and do be able to mod-test some of this gameplay, to avoid making a mis-guided pitch to my company. I really don't want to start out with one engine, then have to switch to another if this gets the greenlight, as it'll just be a big waste of work if that were to happen. So, seeing as how UT is very moddable, has lots of documentation, community support, etc., it has the lead right now. I'm still not set on it though, so if anyone else has another recommendation...