TGB Questions Again?

[Sean Doherty]

I've been playing with the trial and here are some questions and initial observations:

1) Looks pretty good and complete (lots of tools).
2) The initial learning curve is high, but I think you would gain back the time once you got going.
3) I don't think you can create just any game without a lot of extra work. It is not as flexible as a something like PTK.
4) What types of games is it best suited for? What kinds of games is it not suited for?
5) Those using TGB, what kind of game are you making?

Thanks

[amaranth]

I'm using TGB to build my game, which is a casual SIM. So far, I like it. There are a couple of quirks with it, but over all it's great for rapid development. The scripting language feels a lot like C/C++, and so far, I've not needed to modify the engine files to do what I want to do. The editor is also nice, but all of the functionality in it can be directly scripted if you decide not to use it. One of my favorite features are the datablocks, which let you create templates for objects.

What sort of game are you making? If you're making a RPG I definitely wouldn't use it. In my opinion, its best for making arcade games and puzzle games.

I've heard a lot of good things about PTK, but I've not used it yet. (I've grown up with engines that have editors. From what I know PTK is more of a library, right? Please correct me if I am wrong!)

There is one other thing to consider about using TGB... the marketing side... The Torque people are loooouuud. Also, I don't know of many casual games that have been completed with it. If you build a game with it, there is a good possibility that they will help swirl up the publicity for you. It's a win-win situation for both parties.

[zoombapup]

I've worked with the torque codebase (which TGB is an extended subset of) for a long time now (what, somewhere near 5 years).

The good parts about TGB:

It has a much more user friendly interface than it had.
The gui system (which although it has a load of quirks, rocks still)
Performance
It is VERY useful for rapid development turnover
Great scripting/C++ interface
You can get source
Physicsy stuff
Great community
PC,MAC,Linux

Downsides:

Source modification can be a bear if youre not au-fait with how it all works
I get the impression that the physics can be a bit picky.. I have vowed to NEVER MAKE A PHYSICS BASED GAME AGAIN so thats ok for me :)
Lots of stuff to dig through
Lots of whining community too (inevitable with the scale of it)
Requires hardware renderer

There is a new "adventure kit" that enables a lot of stuff you'd need for a RPG.

At the end of the day, you cant really compare TGB and PTK, they are completely different ends of the spectrum. PTK seems like a really excellent framework library for starting from scratch. TGB seems like a great container to mod the engine to do what you want it to. Its a different approach and I actually like both methods. PTK has more appeal to those people who have a NIM mentality. TGB appeals to rapid protyper dont-care-what-its-made-with types.

We are actually using TGB to produce a quick game after shipping air ace. Should be a simple "casual" wargame. Ok, that doesnt sound right, but it will be a simple puzzle/action game based on the concept from an old wargame. I'd expect it to come in under a month, because the whole team has loads of experience with Torque.

Oh, if you DO go with TGB, might I recommend you get Torsion. I mean, recommend it so strongly that if you DONT get it, you are using the tool completely wrongly. Torsion is *the* ide for script debugging/game making in TGB, nuff said.

Amanda: Well, the games I know about using TGB... are Gold Miner Deluxe and Puzzle Poker. I'm aware of a few others, but cant recall the names.

[zoombapup]

Oh yeah, and GG themselves have a casual game coming that uses it. Cant be a bad recommendation.

[TimS]

1) Looks pretty good and complete (lots of tools).
True. Though I can't use too many of 'em cause I'm still using Alpha 2.

2) The initial learning curve is high, but I think you would gain back the time once you got going.
Very true. Sometimes you'll have to bend it to your will, though, if it turns out to not do what you want.

3) I don't think you can create just any game without a lot of extra work. It is not as flexible as a something like PTK.
I can't think of a 2D game that you couldn't use it for easily... can't comment on PTK because I haven't used it, but I suspect TGB would easily do any 2D game I ask it to...

5) Those using TGB, what kind of game are you making?
Arcade / Sokoban (a.k.a. real-time non-grid-based sokoban).

As long as you consider it for what it is (an engine and a series of useful libraries and not a "game builder", despite the title) you can easily use it to create just about anything. If you just want to use the editors to click and drag a game together, however, I have no idea what you could create... but the odds are it won't be very compelling. Maybe a simple platformer or a scrolling shooter could be done this way.

Anyway -- best of luck with the search!

-Tim

[woo]

We're using it for an online, turn-based, strategy game (with light RPG elements). I ported the game to mac in literally minutes. We have at least one more project coming up using it too. I highly recommend it.

In terms of learning curve, there's a learning curve to making games. The learning curve to using TGB itself is actually quite trivial in comparison. As long as it's a mostly 2d game, TGB can get it done.

@amaranth: While I understand that building certain types of RPG's would be easier in a tool that was built from the ground up for making certain types of RPG's, I don't necessarily think that it's a bad engine for RPG's in general. Its built in tile system/editor isn't quite as scalable maybe as it would need to be for a game like yours and it doesn't have a dialog tree system out of the box, for example, but with some fairly minor tweaking (mostly in toolkit stuff, not really in the core engine) it could make a pretty cool RPG pretty easily.

edit: ack - I second zoombapup's torsion recommendation.

favorite torque game builder feature: there is a common core among all the torque tools so that while there is a learning curve to moving to TGE and TGE:A, you can build up your familiarity with a very well rounded engine so if you ever want to break out of the 2D box, get more advanced networking support or mess around with high-end graphics/shaders, you don't have to start from square 1.

edit: I'll also throw in the great work from tentons on games like Fashion Fable and Little Gods (as well as an educational title I believe) that use TGB.

-Andrew Douglas
http://theoreticalgames.com

[Sean Doherty]

@Tim,

[quote]
As long as you consider it for what it is (an engine and a series of useful libraries and not a "game builder", despite the title) you can easily use it to create just about anything. If you just want to use the editors to click and drag a game together, however, I have no idea what you could create... but the odds are it won't be very compelling.
[quote]

Are you saying that unless you modify the source your unlikely to create a good game? Or are you taking about writing scripts?

[TimS]

Are you saying that unless you modify the source your unlikely to create a good game? Or are you taking about writing scripts?

I'm just talking about script here... no engine modifications will be necessary for most games I'd think. Mainly I'm just saying don't expect to be able to click together a game without some script in there to make it go.

-T

[Sean Doherty]

I'm just talking about script here... no engine modifications will be necessary for most games I'd think. Mainly I'm just saying don't expect to be able to click together a game without some script in there to make it go.

-T

Can I make thinks like an action adventure with dialogs or a level editor without twisting the engine to do my bidding? What about a game that has a flocking algorithm where units fly in a formation?

[woo]

it depends on how many objects are actually swarming and what swarming algorithm you use, but it'd obviously be more efficient to write it in c++. It wouldn't require anything serious in terms of code modifications, and you may be able to use some existing resources for what you want to do...
See, the thing is that the way the scripting is setup, it can communicate fairly easily with methods defined in c++, so 100% of your game's logic can still be in script but if you really need a method for calculating the flocking behavior of an ai character to be really screamingly fast, it can be written in c++ and just called from script. Same thing goes for things like embedding sqlite or tying into other 3rd party api's.

If you already know you are going to want a lot of performance oriented code or integration of 3rd party libraries, then go ahead and get the source version of TGB.

-Andrew

[TimS]

Can I make thinks like an action adventure with dialogs or a level editor without twisting the engine to do my bidding? What about a game that has a flocking algorithm where units fly in a formation?

absolutely. Dialog and a level editor I'd do 100% in script anyway (did, in fact). Flocking behavior would be faster in C++, but like woo was saying, it'll depend on how many things you've got flocking. The odds are good that script logic won't be the limiting factor on your framerate unless you get really crazy with it. A dozen boids surely won't bog it down if scripted efficiently. A couple hundred might, but you could always make 'em less intelligent about their flocking.

The bending of the engine that I was talking about were more like when you decide you need it to play flash movies... and midi... and that you really want one of the 'mount' types to be a 'spring force'... Pretty much if you're picky and the feature list doesn't have what you want.

[Greg Squire]

I'm using TGB currently to make a shooter game. Some things are a piece of cake to do, and others require a better knowledge of the engine to do. I've run into a few little "gotchas", which can be "worked around" once you've found the answer in the docs, or in TDN (Torque Developer Network), or on their forums, but still has a little bit of a learning curve to it (not as bad as the older Torque engine though).

Garage Games recently released a game using TGB (though it source was modified). It's called Rack'em Up Road Trip (http://www.garagegames.com/mg/snapshot/view.php?qid=1296).

[Sean Doherty]

I see there is a TGB book. Is it worth buying?

[TimS]

I see there is a TGB book. Is it worth buying?
There is? Where? That's the first I've heard of it.

[Sean Doherty]

There is? Where? That's the first I've heard of it.

Maybe it was a TGE book. Sorry

[TimS]

Maybe it was a TGE book. Sorry
Darn... I was hopin'. At any rate that TGE book has been highly praised by those who are new to Torque, and it is very likely that a TGB-specific one will come along fairly soon.

I might just have to buy it when it comes out, too... :D

[S2P]

Hi,

we bought the TGE book back in the days of just starting out, but the references to the codebase are pretty much the same, if not the same.

We use TGE and TGB and highly rate both of the engines. For an MMORPG type game in 3d, there is a 3rd party kit available at MyDreamRPG (http://www.mydreamrpg.com) - requiring TGE or TSE as your base platform.

TGE - we are developing a large scale open planned FPS, then possibly moving to TSE.
TGB - a 2d side scrolling martial arts game.
Realm Crafter - for an MMORPG