Hi all,
just a quick question;
is Lightwave an efficient tool for creating 3d (character and environment) for gaming? I was recently involved in a project as art director; lw is my first choice 'cause i know it bettere than other packages. But i don’t want to choose a wrong tool at this stage. Can someone suggest me somethig also abotu the process of creating 3d for gaming (just stills and animations in my portfolio now).
thanx!
I’m not a lightwaver, but sure you could use lightwave… Some people working on unreal tournament and serious sam use it so i don’t see why not 
Game art more heavily relies on textures, rather than 3d geometry although as time goes on these “limits” go away… While creating characters you should keep a target poly count in mind, and let the texture do the detailing wherever possible. Most games out right now use around 3 to 5 thousand faces for a character model, some games use bump and spec maps and some don’t but gradually were seeing it used more and more… Thats about all i can think of right now
LW is very popular in the gaming-industry, because it seems to be one of the most powerful tools for polygon modeling (i am not a LW user… i prefer maya…).
the UT2k3 models are done with LW and id sofware did most of their doom3 models with it (even the high res ones)
Lightwave can be used for games yes, but what’s more important is to check whether the game engine you’re using has an exporter for your 3D app.
For example, Croteam wrote a Lightwave plugin for it’s SKA system (skeletal animation models). So you basically take your lightwave scene file and it’s character animations and export them directly into Croteam’s format. It’s really easy.
However, Croteam didn’t write a exporter plugin for 3DSMax, never wrote a plugin for Maya, or any other app as well so you’re dependent on other people having to write those plugins for you, and they can be extremely buggy or not even work at all. For example, even Milkshape doesn’t export Croteam’s SKA right.
So my feelings are that if the company of the game engine you’re working on doesn’t have it’s own exporter for the 3D app that you are using, don’t use that app unless you know there will be reliable 3rd party plugins available.
As for creating environment geometry, Lightwave is good for creating the actual geometry, but you can’t texture your levels in Lightwave as most 3D game level editors don’t import UVs along with the geometry. In Serious Sam, you can use models for world geometry, and then use invisible CSG for collision. There are lots of tricks, it just depends what engine you’re using.
Just my experiences from using Croteam’s stuff :).
Thank You all!
Really precious suggestions. So the point is the interface between the 3d sw and the engine…
The project is on a eraly stage: still scouting for the engine. I think I will partecipate this market screening =)
G.
dead-eye’s comments are pretty right on. Some game development toolsets are pretty tightly bound to a particular 3D app although the degree of dependence varies from simple model importers to deeper scene and animation info support.
Theoretically, virtually any of the better 3D apps is pretty much capable of producing the needed data.
dead-eye, if you stop back, would you comment more on your experience with the Serious engine? There is an interesting related thread:
which had not discussed that particular toolset last time I looked. That discussion is mostly about open source engines, but also of commercial ones so long as financially within reach for “garage game” developers.
My experience with the Serious Engine is just with the modding scene. It’s too expensive for me to license… $20,000 to $100,000 depending on the level of support that you need.
Their entity scripting language is almost just like C++. It’s not like JavaScript as Unreal’s UScript is. In fact, if you’ve ever used MFC, you’d think Croteam worked for Microsoft as their code looks just like MFC code (the style, same syntax, and believe me, just as bloated too lol).
The Serious SDK for modders comes with about 75% of the game’s source code, including all the math, polygon, and BSP tree stuff. It’s really nice to look at how a good part of a successful commercial engine was built.
The netcode for the game is extremely bad. If you have 2 players in two different sectors (rooms basically), the game becomes virtually unplayable unless you have a 1 MBit connection. This game lags badly even when playing on a LAN with just one buddy. If network play is an important feature, this engine is not for you. Anyone who licenses this engine better have a good WinSock programmer available… because they’ll need him 
.
There’s no decentralized AI. I like the way the Monolith (NOLF LithTech) engine does AI, as all the AI and enemy classes are seperate. In Serious Sam the AI is all in a single enemy base class so all the enemies pretty much behave the same way. If you make one change to the AI, it’s going to affect ALL your enemies (unless you coded totally new enemies).
Currently, the Serious Sam Engine is only good for Serious Sam type games. Making any other type of game would require an extreme amount of coding. The main reason people license engines is so that they don’t have to do alot of coding. This is why the modding scene in Serious Sam is very poor. Virtually all teams end up saying, “We’re moving to so-and-so’s other engine because the Serious Engine just doesn’t have all the features we need.” (But oftentimes these teams just use that excuse to cover up for the reality, “We really don’t know what we’re doing.”)
The game physics are also whack and there’s no support for vehicles. Croteam’s also abandoning their MDL model format (model has to be in three peices) for SKA (model can be one peice), for which SKA player models aren’t directly supported in the current engine. So basically everyone’s waiting for Serious Engine 2, including me :).
Another thing is that with the Serious Engine you’re stuck using Visual C++ whether you like it or not. It’s not like the Sourceforge engines like CrystalSpace, OGRE, or NeoEngine that can be compiled on different compilers. The Serious Engine uses Microsoft-specific template code and keywords such as __forceinline and inline assembly code that can’t be compiled on say Borland C++ or Metrowerks CodeWarrior.
EDIT: Wow did I really vent. Sounds more like a bunch of complaints lol. But it’s a nice engine that’s really fun to play with if you’re a bored C++ coder like myself :). The level editing tools that also come with the game are awesome and easy to use. SeriousEditor runs 100x’s faster than UnrealEd on my machine. It is just so much fun to play with Croteam’s stuff.
>>rant…
No, you told me pretty much what I wanted to know. Thanks!
As a fellow LightWave’er, my primary interest is the strong LW connection – most of the others discussed on gameboy’s thread tend to center around MAX or Maya. I guess I’ve just gotten used to LW and think it theoretically should be a pretty good game tool. Anyway, the $$$ you mentioned make it a true commercial engine, not something for the casual experimenter.
My experience with the Serious Engine is just with the modding scene. It’s too expensive for me to license… $20,000 to $100,000 depending on the level of support that you need.
I was thinking… it sounds a bit too cheap.
well if you want an engine thats not cheap the Unreal II engine is $350,000 for a single comercial use license plus ROyalties hehe. I am in the same boat I am looking for an Engine that I can use LightWave in. Crap everone heloing me on my team except for one perosn uses LightWave :surprised I hope I dont have to change engines but if I can’t get an exporter/importer I might have too…
Originally posted by TheWriter
I was thinking… it sounds a bit too cheap.
You’re absolutely correct. Compared to licensing the Unreal or Quake engines, it is a bargain :). Keiyentai, what engine are you currently using?
I did not know about Quake’s royaties ontop of their other price listing. Man, that sucks for quake developers.
Can someone post me the exact specs on their price?
This thread has been automatically closed as it remained inactive for 12 months. If you wish to continue the discussion, please create a new thread in the appropriate forum.