A hopefully constructive engine comparison

#21 I'm still not sure a comparison topic is the best place, unless you compared implementing a Thief shadow detection system in TGE, BeyondVirtual, Unity, and Unreal 2k3, for example to compare the engines and what it would take to do it--though for the most part, you can quite easily fake it with triggers on that front in any of the engines mentioned. The render quality will be different, but the technical engine side of the gameplay can be done without much worry, in script.

Fighting games are so specifically complex in terms of control and animation integration, that you would have to do some significant tweaking to make it work nicely in any middleware engine. Especially if you work with combos. Melee in, say, Oblivion is several thousand degrees simpler than even very bad, buggy fighting games.

Performing a raycast to determine visibility isn't much of a big deal in any of the engines I've mentioned. I haven't done anything like it in Lawmaker, but I believe the farm example level has an example of line-of-sight AI. I've done similar things to see if a trigger is nearby and visible (for a switch) in Torque. Unity and BeyondVirtual have simple means of doing this as well. The algorithms behind "seeing" something are not that complex (especially in 2D), but interpreting and reacting accordingly can be. That's the area where LawMaker's AI functionality plays nice.

The memory card issue in GT is an old, old-school trick. It takes up multiple blocks and has fixed binary data (I also believe that it was streamed from the card in GT1 and 2 and decompressed, but I can't remember if that is the game that I read about). It stores garage data and such the same way that old-school RPG's stored character data: in a fixed binary format. It makes it easy to read and parse. It was also why your garages could be screwed up if you turned off your game while it was saving between races. It made GameSharking it, or hexediting in old RPG's, quite easy.

Tuning and tweaking the performance would be a horrendous job for the programming and testing team. Creating a garage and adding parts that up, say, acceleration, is simple. You could do that now in script. Have three engines, each one a little faster. But tweaking it in a "realistic" manner would be extremely painful and time intensive.

Didn't Path of Neo use a custom engine that was designed specifically for the "unlimited" things the player could do? Much like Red Faction's engine was designed around the carving concept? I can't remember. I hated that game, though. So I think I blocked it out of my head.

In the early Tomb Raider games, the blocks were all certain sizes, so the grab-able calculations were easy. I did something similar a long, long time ago with "edge bars" in TGE. I had the mount points for a cliff edge (a line model with a alpha texture) and then changed the animations for a "pull up". There's no reason that these should be "stock" since they are so specific (and the implementation is especially specific) to different games.

I'm still stuck on keeping this thread about a comparison, not a "cool resources to add to TGE" topic. That would be a nice topic, and there are resources for swimming, ladders, melee, etc.

#22 .

#23 It's not so much about modern-looking games but games of specific genres, which moves from middle-ware into the realm of only creating Tomb-raider games. Right now Torque has the reputation of making only FPS's because of its history and starter kit, but most of the indie games made with TGE have *not* been FPS's.

It's your topic and you can do what you want with it, but I was just thinking about people coming to it to find engine comparisons and then finding all kinds of things which don't really belong to the topic at hand. Useful things, but not relevant to engine comparison. Also, there is no telling whether it would be indexed in the search engine correctly with a topic change in the middle. It might be more useful to people searching for useful resources to have a topic with what they are searching for.

#24 .