Torque Game Engine Advanced
Windows, Mac OS X
Languages Written In:
- User Reviews
The Torque Game Engine Advanced (TGEA) is a fully featured AAA game engine that has been used to
make countless commercial games. Building upon Torque's award winning multi-player network code,
seamless indoor/outdoor rendering engine, skeletal animation, drag and drop GUI creation, a built in
world editor, and a C-like scripting language, TGEA also incorporates modern shader features, supporting
normalmapping, detail maps, specular, cubemapping, glow, refraction, reflection, texture UV animation,
and more. As pioneers of the Indie license, the TGEA engine with full source code is offered at a low cost
to independent developers. The version 1.8 beta now also includes support for MacOS.
- Object-Oriented Design
- Save/Load System
- Multi-Platform and a journaling and Integrated hierarchical profiler
- Efficient memory manager with extensive debugging features; Resource Manager and a File/Stream IO
- Unicode international strings support
- True spherical distance fogging (for terrain, buildings)
- Spherical distance clipping (for terrain, buildings)
- Environment Mapping
- Lens Flares
- Particle System
- Integrated particle system with scripting engine.
- Supports texture and scale animation as well as particle generators which can emit multiple particle types at once
- Layered volumetric fog
- A water engine that supports dynamic waves and multi-pass wave crest and shoreline textures
- Animated lightmaps
- Control lighting via World Editor
- Vertex and multi-pass lighting
- Shadow Mapping
- Projected planar
- Shadow Volume Projected object shadows (clipped against the environment)
- Volumetric Multi-pass texturing with detail and decal support
- High Level Drop custom shaders or effects into the shader interface or have them procedurally generated
- Mesh Loading
- Continuous level of detail based on a progressive mesh algorithm
- Damage and collision detail level
- Supports Milkshape, 3DStudio Max, and Blender plug-ins, including an exporter and helper objects
- Occlusion Culling
- Portal based interiors with seamless integration with the terrain engine. Buildings can be instanced, placed, manipulated and scaled with the world editor
- Spatial Database
- Quark and WorldCraft converter and lighting
- Support for Cartography Shop and almost all .map-compatible brush editors
- Inverse Kinematics
- Skeletal Animation
- Animation Blending
- Supports, bone, mesh, texture bitmap, texture coor. and visibilty animation
- Mesh vertex deformation and multi-bone mesh skeletal animation
- Simple interface to multi-sequence animation manager
- Continuous, seamless, LOD mesh generation and Aggregated tile mip-mapping
- Dynamically generated blend tiles and Altitude based fog banks
- Seamless integration with the interior engine
- Basic Physics
- Collision Detection
- Rigid Body
- Vehicle Physics Built-in physics engine Sample Implementations for major 3rd party physics libraries: PhysX, Bullet, ODE
- Award winning client/server architecture, which uses UDP and TCP
- Bit level stream, message, and string packing
- Uses the optimal Notified Delivery Protocol
- Packet stream and Object ghost manager with partial object updates
- Finite State Machines
- Scripted Highly customizable through TorqueScript
- 2D Sound
- 3D Sound
- Streaming Sound
- SFX/Music driver using OpenAL
- 3D sound support; panning, volume, doppler, cones
- Multi-channel prioritized SFX manager
- Built-in Theora video codec playback
- FMOD supported
- Easy integration of third party libraries
- World Editor allows you to manage terrain generation and object placement as well as construct, place, size, scale and rotate objects visually
- Terrain Editor, Terrain Generator, and GUI Editor
- C++ like syntax
- Local and Remote Debugging
|License Name||Price in $US||Source Code Included?||Additional information|
KEEP AWAY FROM GG - Fraudulant behaviour!
This review is a little different to those others I have just read. It seems to me I have been ripped off....and most likely illegally!! I bought a professional T3D license back when it was still in Beta form. Several months ago our username in the forums was myself and my husbands name as we are a husband and wife business and we use this in every forum we're in, despite only one of us using whatever product. This was until some smartarse decided to accuse me of using T3D illegally on multiple machines just from the username. Mind you this same person we have seen put other people in the forums down. Naturally I put in a complaint at being victimised and guess what.....Garagegames has taken his side and told us to change our username to something.....so I did.....our dogs name! I have no way provoked this and have never been rude to anybody, in fact I've working in IT Support most of my life and being a female programmer I know what it is like to receive abuse and I would certainly never give it out, it was completely un-provoked! Several months later after being disgusted...I went in to post a question a few minutes ago and I have been banned from posting in the forums! It's bad enough they have un-finished and bad documentation where you have to rely on the forums for help and now I can't get diddly squat!!! So now we have an engine we cannot use. I am at present awaiting a response from GG and have told them if the fix isn't satisfactorily I'm demending my money back. I am trying to get a game released commercially but now I doubt I can rely on GG as we cannot trust a company that does things like this to their customers.
On a better note the engine is good....when it's working. I feel the coding is a mess and unnesaccerily long. I have looked at other engines and I know what one I will be using if GG don't get their act together. The features are good but many lacking what they claim to have. Pretty disapointed when Wet Maps didn't even get a show. That's just one example of things they claim to have but can't deliver.
I just wanted to get my view out there and warn people in the hope you tread lightly before committing.
DON'T TRUST GARAGE GAMES!
Just an update to all your potential GG Buyers from our previous one, we have contacted GG via email. They have refused to give out a phone number to discuss why they have banned us from posting in the forum and they refuse to refund our money. They are accusing us of breach of EULA based on assumptions made from a fellow forum member. All we did was use both our names in our user ID and they have assumed we are using it on more than one machine and banned us. We've only been told this morning they almost banned us completly from the web site including the software we have rightfully purchased and with that purchase comes forum access. What good is a piece of software who's only support is the forums if you're banned from the forums. So we have now reported this to the FBI as internet fraud. KEEP AWAY FROM GG!
GarageGames and their Marketing Superiority
Where to begin?
Features: The rendering engine looks ok. Nothing special and certainly not as pretty as C4, Unity, Leadwerks, etc. The networking in Torque is top notch and IMHO the best out there.
Ease of Use: This kills the engine. The art pipeline is as easy to use as putting out a forest fire with a squirt gun. Their DTS format is utter crap and not well supported. The DTS format itself is extremely dated and in major need of an update.
Stability & Performance: Engine crashes (especially while using the editor).
Support: Sorry GarageGames, you can't count the community forums as your only means of support. YOU need to provide support.
GarageGames is ALL about marketing and not about delivering. Torque Shader Engine....or now Torque Game Engine Advanced is already dead. This engine looked to be so promising from their milestones. I was one of the suckers who trusted GarageGames and purchased an early adopters license. When GarbageGames realized that their spaghetti code for Tribes (that they call an engine) was incapable of implementing all of the goals they had for TSE, the just erased them from the milestones page. Needless to say everyone was VERY pissed off. I hoped that they would continue to improve upon their engine and deliver an outstanding product. Well what do you know, TSE/TGEA is done. They have now moved onto Torque3D, and you can even be an early adopter and purchase Beta access. Sound familiar?
GarageGames justification for the new engine was they didn't have enough resources and need to charge more for their engines. GarageGames associates (henchmen) even try to make us feel guilty in the community about how nice GG has been giving us free update for so long in TSE/TGEA. Here is the reason why you don't have enough resources GG, ready?
STOP TRYING TO DEVELOP AND MAINTAIN SO MANY ENGINES AT ONE TIME!!!!
Lets see how many engines you have been supporting:
1) Torque Game Engine
2) Torque Shader Engine
3) Torque Game Builder
4) TorqueX 2D
5) TorqueX 3D
6) Torque3D (development)
This list doesn't even include the mountain of starter packs (hacks) you support.
***WHAT I FIND TO BE THE FUNNIEST THING OF ALL, ALMOST ANYTHING USEFUL OR EYE CANDY THAT COMES OUT FOR TORQUE, COMES FROM THE COMMUNITY.***
LOL, what do your programmers even do? Your brand new Torque3D your are working on, is developed in the most part by your associates (henchmen).
So all your suckers out there who believe the marketing BS of GarageGames, go ahead and purchase Torque3D. Be prepared to pay BIG TIME for your updates from now on. More importantly, be prepared for migraines from horrible source code, an inaccessible art pipeline, and dishing out the $$$.
Not high on my list
Its been a little while since TGEA has been available, yet, its still in need of attention. Unfortunately, GG has already began shifting to their "next big product" and the expectation is that the latest version may very well be the last.
TGEA has failed to deliver on a majority of the plans that were mentioned in Milestone 1,2,3, and 4. Those plans have been for the most part rolled over to the next product with a larger, non-indie free price tag. GG use to be innovative for its pricing structure and how it worked with the "Garage" mentality, but it has since lost most of that.
Its feature set looks pretty, but if the community is any indication, its not going anywhere anytime soon
Sorry, but I must say that this engine is very poor.
Engine is very old.
For support there are only forums, where you can get answers only for simple questions from other users.
Documentation? It´s big pain of all garagegames.
Easy of use??? No no no ... I dont know who made this, but he never saw a good engine. THIS IS VERY UFRIENDLY.
Scripting?? Whaaat a big joke.
If you want to do a simple things in a hard way than this is right choise.
If you really want to do some games without so much pain you shuld try some other engines.
I hope next version will be much better.
I'll briefly highlight some of the issues with this engine.
* Editors are not stable. Many simple tasks will crash the engine or produce unexpected results. No orthographic viewport for world editing.
* Art pipeline is horrible. BSP based interiors don't build correctly and seldom receive correct lighting. Tools to convert to native Torque formats don't work well (if at all) and setting up a scene in Max is an overcomplicated and problematic task.
People seem to have mixed opinions about the DTS format. It does work; once you learn it you will be able to bring 3D content (including animations) into the engine however I feel it is a cumbersome setup. I've used much more graceful solutions that provide all of the features of the DTS format (LOD, animation blending etc) in a much more intuitive way. The DTS plug-in has some strange limitations too such as a limit on number of nodes, details and polygons one can export. Only one UVW channel can be used and the scale setup is hard coded into the exporter. Combine that with the array of broken features (additive blend animations, triggers that mysteriously move position when not initiated on the first frame etc) and the steps you have to go through to correctly set up a scene (hierarchies, helper nodes etc) and it makes for a poor art pipeline.
A new tool named constructor was released free of charge by Garage Games to help remedy some of the art pipeline shortcomings however this tool has followed the usual trend, i.e. it is riddled with bugs and many of the advertised features simply don't work. Also, source code for this tool and most others is not freely available. You can request the source code and if successful will have to make an additional payment for it.
The Atlas terrain system, which is a paging terrain system that supposedly allows one to create huge terrains, is severely crippled. You cannon edit, create, or paint terrains in game. Instead, you have to use a bunch of 3rd party tools and import terrain information into the engine. Even with these 3rd party tools the engine has trouble generating large terrains. The terrain doesn't cast or receive shadows and does not work across a network.
The standard work flow of TGEA is quite poor. It may be passable for hobbyists but certainly not adequate when time is money and your staff is wasting time with external tools in a vein attempt to efficiently bring content into the engine.
* Lighting and shadowing is not indicative of a "next generation" engine. Low resolution, hard coded, lightmapping. No support for infinite lights, no stencil shadows, self shadowing looks horrible and has a significant performance hit. Only one object type supports self shadowing and they do not cast shadows onto each other. The lighting in this engine is reminiscent of a late 90's engine.
* Performance in a high poly scene is less than desirable. Performance in general is less than desirable. The engine will choke at any sign of moderate complexity. Very few optimization and visibility determination techniques exist in this engine (scissor optimizations, culling, zones and portals etc).
* The OpenAL audio layer is disgusting. Fall offs and attenuations do not work correctly, if at all, and haven't done so since the engine's release. Streaming large audio files will cause the engine to hang.
* Documentation is non-existent or outdated and much disorganized, difficult to locate. Although recently the community has been working hard to improve this.
* The source code of this engine is an abomination. It is extremely messy and disorganized. It's apparent very little forethought was put into the design and structure of this engine.
The engine reached its official release on 15th February 2007. The quality of the product does not reflect this though, i.e. it may not be in beta anymore however the engine is still plagued with bugs and has some serious design issues.
Thankfully they have a full featured demo available for download; I strongly recommend thoroughly trying it out before committing to this engine.
Dont waste your time...
This engine appears very capable on paper, but in execution it is far from it. While many modern features are supported in this engine, nothing works like you expect and most of the implementations are half-baked at best. Overall the code is very hackish and the engine is not at all stable. To put it simply, don't waste your time.
There are some benefits to the engine, it does in fact display 3d models. It supports shaders. It has a scripting language. It has a GUI editor, particles, physics, etc. etc. Thats all fine, except its a pain to use in every sense. Read on to see why.
First and foremost, the documentation is incomplete, outdated or flat-out wrong in some cases. Everything is contained on a wiki, but the community are expect to update it. Many of the newer features of the engine are glossed over, or essential sections are blank (its a wiki). At first it appears to be a wealth of information but as you learn the package more you will realize how worthless a good 90% of it is. Much of the information is for the original TGE and may or may not work in TSE/TGEA. There are only a few starter tutorials, nothing really explains the system in any practical manner. Much of the docs just explain some illogical hackish method needed to get something to work in torque, yet fails to give you code to make it work. For any real help you must go to the forums, but half the time people are just as in the dark as you are. The employees are sometimes helpful but many times even they don't have a simple solution to a common problem. You can imagine how confusing this engine must be if even the developers of the said product don't fully understand how it works.
In terms of art, where do I begin. You can't just import a building from 3dmax or maya. It has to be in dif format which is like a compiled bsp. So you have to use obsolete programs like QuArk (used for quake1) in order to model levels. Then you must convert to torque's proprietary dif format. The program needed to convert from a quake/halflife map file to a dif is buggy and not really supported. It will mangle up models of even moderate complexity. This is just the beginning of the problems. For exporting dts models, from a program like maya or max, it is just as problematic. The exporter sometimes hangs for long periods, 5 or 10 minutes when exporting a moderate detail mesh. You can't just build a model and export it with one click. You have to setup the model with a special name, special helper dummy objects, mess with the graph editor hierarchy. All sorts of things just to import a simple static object. Then to get the model to display in engine, you need to write a material script. The syntax on the material script changes all the time, it follows no kind of uniform design. The information of the wiki is wrong, forum help can be confusing. The syntax for setting a texture exported from maya is different from that of max, for example. If you get it wrong, expect to crash the engine.
The engine is not at all optimized. It will run fine with a big terrain and nothing else, but once you start dropping in some real models it slows to a crawl. If they are animated, even slower. It does not take advantage of even basic optimizations like backface culling. If you are looking to push the graphics on your game, this is not the engine.
The editor, which is the main feature of this product is not intuitive to use and frequently crashes. Many times it will crash when trying to import a new model. Even models that display fine in ShowTool (torque's model viewer) can and will crash the editor. Once your game gets a certain size it will take ages for the editor to start, possibly as long as 5 minutes. This is just to start the program, I'm not even talking about compiling anything. This becomes a very big issue if you are an artist working with torque because the way its designed you need to be a coder to understand all the hackish methods needed even to do something simple like load a 3d model. For example, theres no simple color picker. In some cases you have to put in a normalized vector just to pick a color which is ridiculous. Moving objects is not precise, there is no orthographic view so you have to do everything from the freecam perspective. Sometimes you will move an object, spend your time to get it perfect. Then for no reason at all it will jump to some random location. This software is not stable, crashes all the time. Even just alt-tabbing can crash it.
The lighting and shadowing on this engine could also be a lot better. The renders tend to be very washed out no matter what type of lighting you use. It doesnt work with transparent surfaces too well. The lightmapping on static objects is very low resolution and its hardcoded into the engine, so theres no easy way to load lightmaps from max/maya or other commercial applications. Modifying lights in the editor is a hassle. In fact, just about modifying anything in the editor is a hassle.
The physics in this engine is totally worthless. The rigid bodies all act like marshmallows under water. The vehicle functionality is barely passable. Collision detection works as expected, I'll give them that.
There are so many more problems with this product I can't even begin to explain. The developers, GarageGames, are just a bunch of spin-doctors. They have great marketing, but don't back it up with the goods. This engine has been in development for years and it still not officially released. The level editor, called Constructor, has been in development for over 2 years. Still no release. If you buy this product, expect to be waiting a long time for any meaningful updates or bug fixes.
Please, do not waste your time with the sorry excuse for a 3d engine. You will regret everything about it, the product is so bad it borders on false advertising. I have been programming for close to 10 years, modeling in 3d for the past 4, and I've used a few different engines in that time. This is by far the biggest waste of time, money and effort I have ever encountered. Pretty much any engine will get you further than this piece of trash. Seriously, don't buy it. There are better engines you can get for free.
Where to begin.
Having used TGE, TGEA, and T3D, all I can say is that it used to be good. But right now as it stands it definitely isnt worth the price.
Features: It has some really great features, but like Dark Basic, expect to pay for the many packs and additions that are available to get what is standard in most other engines out there.
Ease of Use: Its pretty good, not as bad as some people make it sound, but it isnt great either. If you're a programmer it will be easy for the most part.
Stability: Is so so. It is very buggy, and therefore this is where the source code will come in handy because basically you'll have to fix the engne bugs yourself. Because if you wait for GG to fix them, then you'll be waiting a long time (if they even fix it at all).
Support: Is okay, but not great. The community forum can get very rude at times. The documention in some aspects is very lacking, which is its biggest downfall.
- Performance can be very low and require very high spec requirements, so having an end product run at acceptable speeds will be a daunting task for many to accheive succesfully.
- Price, its definitely overpriced, especially with T3D (since it basically should have been a TGEA upgrade instead of a rebranding to T3D)
- Licensing agreement is steep compared to other engines out there, and basically isnt worth the price you pay.
So overall if you're a hobbyist that has extra cash its an okay solution. But if you're a serious indie looking get into some serious game dev and release a product to market, look elsewhere, you'll have more hassles and lost time than need be with Torque.
Good spots, making for a dissapointing whole.
Unlikely to make friends for this, but: looks like the marketting guys for GG ignored the actual engine again in spots. Which is a real shame, since there *are* good points to the engine. It's just hard to take a company seriously when they post things like:
• Multi-Platform and a journaling and Integrated hierarchical profiler"
When last i checked, the journaling was shattered,
• Local and Remote Debugging
I've never seen code for remote debugging, nor an implementation thereof, though that could verry well be burried in some of the newer documentation,
• Built-in physics engine Sample Implementations for major 3rd party physics libraries: PhysX, Bullet, ODE
which should read: built in physics engine and several user made examples of experimenting with plugging PhysX, Bullet, and ODE into that version, which this is not, most of which do not take advantage in the slightest of networking,
• Projected object shadows (clipped against the environment)
only counts for terrain, and .diff (proprietary BSP model format) objects as oposed to say, thier polysoup implementation for example,
• Spatial Database
which unless they're talking about an internal memory resident version, is also completely undocumented, which is why theres community rescources to plug in just about any sql derivative known to man, also by thier nature kept up as the authors have time,
"Inverse Kinematics, Skeletal Animation, Animation Blending: "
when, at time of writing, IK is non-existent as well, though there are a few folks working on it,
and at that point, i'll leave off on the negatives because I really don't want to come off as trolling negativity. There are some dedicated people doing quite impressive work that makes it in in fits and starts improving the engine over time, hampered only by thier obsession with full big releases and filling in marketting bulletpoints to the exclusion of excellence in the fine-grained implementations, and as long as you take every word from the marketing department with a heavy, and let me stress this one: HEAVY grain of salt, you'll likely be pleasantly surprised. But please. Please. grab a demo and try to pull off an equivalent of what youre aiming for before shelling out to the company as a whole. It's the only way to see a bit more integrity on thier part again.
All that being said, I have shipped a game with this, and since it does come with full sourcecode, and I do mean full, not libraries with a few headers, if youre willing to put a major ammount of elbow grease into the engine in the spots that it either a- doesn't cover, or b- covers poorly, it is possible to work with it. Just don't expect anything untill you've seen it do whatever you need it to do first hand, or are willing to devote some heavy time into doing it yourself.
And GG takes another (deserved) hit.
So, somewhat recently, GG announced their move to an MIT Open Source license.
Not sure what motivated this move as I don't know what goes on behind closed doors at GG. I have a feeling their inability, or unwillingness, to properly support their "product", leaving it to the community to fix their problems has finally caught up with them. The GG Boys are too busy working on yet another new engine to sell to support the ones they already have, apparently.
Other Indie game engine companies create one, maybe two (though I've rarely seen it) engines, and then continue to develop, expand, improve and most importantly, support that engine over time. GG is the only company I've ever known who will pretty much abandon their current tech every time they decide an upgrade (or sidegrade as the case may be) is warranted, which is roughly every 2 years, considering how many different engines they've pumped out in their 12 years in business.
That's not including the absurd amount of add-ons, content-packs and starter-kits they produce at the same time.
No wonder their engines are always bug-addled messes, they're too busy "making new stuff to sell!".
The disdain, disrespect and dismissive attitude they hold on the forums for anyone who isn't firmly in their fan club is palpable.
Here's what a GG employee had to say in response to reviews here on DevMaster. This is a direct quote. You can find the source by going to their forums and doing a search for "DevMaster".
Begin Quote :
"I think the idea of an engine database can be useful, but no matter what we do we are going to get a lot of jaded users of previous engines commenting on Torque 3D. Some have not even used an engine since TGE or TGEA Beta, so their comments and ratings will be misleading. I can even see users of rival engines posting just bring down our engine.
Maybe our new hire will have some insight and suggestions on combating that. "
: End Quote
So there you go. That's the attitude taken by GG employees about criticism leveled at them here on DevMaster. It's just a bunch of jaded people who obviously (since they're criticizing it) don't really know what they're talking about and are just being misleading. It must also be some posters from rival engines.
They were also awaiting a new hire to "combat" (read: spin, BS and censor if possible) the negativity.
Well, 2 years on and we can see how that worked out for them. My guess is their "marketing guy" took one look at the complaints here, compared them to those on the forums and maybe even asked the GG guys about their accuracy, and then decided the amount of lying they'd have to do to make GG look good would border on false advertising and lying to customers. Least that's the most likely scenario I can imagine.
Folks, the complaints that people post about on these reviews are right in line with many of the same complaints you'll find in their forums. They are legitimate issues that are brought up many times in the GG community - even by experienced users - and either go unanswered or are reacted to with rude dismissal by some of its community members. GG's attitude toward fixing their own issues is "We gave them the code. Let them fix it".
No, it's not "jaded people" posting "misleading things". It's people who've paid GG money expecting proper professional support on a complete engine that works as advertised; a completely reasonable expectation.
It's a shame to say, but GG deserves any and every bit of poor luck that has come and will come to them. Maybe someday they'll get a clue, look at how other, more successful companies develop a *single* engine, and actually provide proper support instead of leaving everything to their community to fix. Unity3D would be a great example, considering they're pretty much what GG wishes they were. And no, I'm not a Unity3D customer, nor do I even particularly care for their product or licensing. I'd be dishonest, though, to say they aren't doing something right considering how successful they've become over time.
Good and bad
For it's price its a good engine but don't expect an easy ride.
The best part of this engine is its network code and its well known for it, yet this feature is also a problem if you want to make a single player game because everything gets doubled especially on the physics side.
Physics gets calculated on both the client and the server which is essential on multi player but a resource hog in single player mode. The physics itself is OK for basic requirements but not spectacular for anything advanced. No built in rag doll or joint support which is essential for any character based game involving weapons.
The engine has gone through many stages of evolution from it's earliest V12 then TGE then TSE then TGEA and now the upcoming T3D. What has changed? Well most visible is the addition of shader support. TSE was an acronym for Torque Shader Engine but then the marketing people decided that it sounded too much like TGE with shader support (which it was) and it got renamed to TGEA (TGE advanced) It also included the Torque Lighting Kit which was an attempt to fix many of the engines lighting deficiencies.
Right from the start it was a bad design choice in that the developers involved were Microsoft fans and decided to make TSE/TGEA use directX only. Imagine that, a community of 1000's of TGE people most/all of them being openGL and Mac/Linux supporters being expected to upgrade to a Windows only engine. It did not go down well with the community at large and adoption of TGEA was slow. The company was also trying to get in on the Microsoft gaming channels and produced TorqueX which was a failure from a business perspective.
Then GG got bought by instant action and a lot of time and effort got put into that, leaving TGEA owners with a bad taste in the mouth because the product was incomplete but sort of functional. All this internal 'hunting' for a viable business direction had taken its toll and the forums became much quieter. Many of the old TGE brains of the community had left and GG business advisor had told staff to withdraw forum support and charge a fee for support. As of then you got no forum support from the people who knew the product best. The only support you get now is from other community members who might be bored enough to answer a question assuming they have the knowledge. The instant action acquisition also seems to have pushed out the original brains of the company for reasons known only to them and the mature management got replaced by younger decision makers who are taking the company in a different direction.
Examples being the 'decision' to change the web-site from a bright and cheery 'happy place' to a dark and dreary uhm, place. The reasons are technical apparently, the new design has better features but its a work in progress. Along the way someone decided to sweep all the art exporter data into a single artists 'corner forum', thereby loosing any time continuity in the forum posts and mixing old dated posts with newer exporter data to the degree that its now foolhardy to trust anything you read about exporter documentation, assuming you can find it in the first place. Which brings us to the exporter problem. GG never provided any official support for exporters and it was up to community members to write their own exporter for the complex dts file format. The attitude is: I'll sell you a car but no one makes tires for this car so you will have build your own tire factory because making tires is not our core business. Have a nice day.
The newer T3D will circumvent this problem by supporting the Collada file format. At least that's a good decision. Another good one was to go back to their roots and stop kissing Microsofts backside because the latest TGEA 1.81 release supports openGL on the Mac platform. Still nothing on Linux, not even a dedicated server which means you will have to shell out for more expensive Windows hosting for your games.
The biggest problem this company has is not listening to or caring about its community. These are their customers but they get treated like unwanted annoyances.... I gave you the source code now get lost while I continue to shoot myself in the foot. I have to keep doing that until I figure out why it hurts so much.
T3D now has an early adopter temptation but don't rush in blind. It might look good and shiny but all its new features are new and won't become stable for years. TGEA 1.81 still has unfixed bugs which will probably remain unfixed because company attention is now on T3D. The ability to use zipped data files is broken which means a game shipped with TGEA will be a mess of files, exposing all your art assets to public abuse. Zipped assets worked fine in TGE and one would expect that when you upgrade, these things will continue to work. Re-skinning a mesh is another example, it used to work but no more. These 'bugs' get hidden in a forum which you only have access to once you have bought the engine upgrade. There should be a PUBLIC known problems list so that you can make an informed choice before paying your money for an upgrade. Making it public will also force the company to give attention to fixing these problems instead of sweeping them under the carpet.
In general, the code base is still the V12 or Tribes2 engine which has been heavily modified and added to in an attempt to modernize it. But its a mess to work in. The code has become something like a Xmas tree. An old framework with lots of shiny things 'attached' which make it look good but if you want that shiny red ball to bounce off the shiny green ball then you have to follow the string down the branch, criss-cross other branches and shiny things and create bridges where none exist and hope that other shiny things are not gonna break along the way.
It needs a ground up rewrite with a core framework that's designed around current game requirements and it needs to be done in 6 to 8 months because the game world changes fast. That means it will require a large team of game professionals on a project to get it done in that time frame. That makes it expensive which pushes it out of the Indie budget. GG are working on a plugable engine architecture based on design patterns and have been doing so for years but its still not ready.
Will there still be a market for it by the time its done?
Is T3D a rewrite? Nope, it's still the Tribes2 Xmas tree with yet more new shiny things and some old broken not so shiny things swept under the rug under the Xmas tree. The question is, can you get something better at that price. If you can, go for it.
An Indie Developer using TGEA
I am currently using TGEA 1.7.1 to develop a retail game. So I review the engine from experience since I also license many other engines. The cons are to demonstrate that this is more of a engine for teams that include programmers, as opposed to an artist/ producer, & designer oriented team trying to make a retail game.
1. The community is great and you can find bits and pieces of code resources.
2. Final games built with TGEA look fantastic.
3. The engine is great for programmers who can script using TorqueScript.
4. It is easy to get artwork into your game project.
5. Has 3rd party add on's for audio (FMOD), and VoIP.
6. Comes with a very basic FPS template.
7. Water looks nice.
1. If you can not script using TorqueScript and no one on your team can, then you will not be able to develop a game with it. A programming is required to develop a game.
2. Getting characters into the game is hard, since each animation needs to be exported out as an individual DSQ file. Time consuming in comparison to other engines character animations requirements.
3. There is only the FPS template that comes with the engine, therefore it takes a lot of work to build a RPG. No official GG approved RPG working template available.
4. Very few 3rd party code and component add on's available. Requires programming time to develop.
5. Entire engine is Programmer oriented.
6. No physics engine integrated. Unable to make objects fall or bounce, (i.e. rag doll) until you code one of the industry standard physics engines into the engine.
7. Video does not by default play out of the box with easy insertion points using tools. A programmer must code to play, and stop Theora video when appropriate. Few examples made available. Should be in the fps template to make it easier.
8. When engine updates occur it is difficult to upgrade your game to the next version of code without time consuming programming changes. i.e. going from 1.7.1 to 1.8.1.
9. Some 3rd party code Add on's are too costly for small indie developers to purchase.
10. Allows you to only define a diffuse and a bump map (normal) image for textures. Can not use and define images for your specular map, ambient occlusion, or displacement maps.
11. World editor is limited in comparison to competition.
12. Most helpful suggestions come from other developers, and not the engine developers.
13. Must pay additionally for support from engine developers.
14. No real AI engine, for complicated logic it must be written, or a 3rd party library licensed and integrated.
15. Existing books are all TGE oriented and not TGEA specific.
16. A real fully working game should be provided as an example so that developers can take it apart to learn from. Not a demo with no real gameplay.
2008 engine of the year!
It's a wonderfully engine for beginners to season vets. Tons of content packs to help you get your project started. The community is great! The documentation is fantastic and as far as the engine (for the price) it speaks for itself.
In my opition it has really made an improvement from 1.03 to 1.8.1
Check out a demo for your self http://www.garagegames.com/products/torque-3D
If your not happy with TGEA then wait until T3D it'll be released later this year!
Top-notch indie-oriented game engine
This is a great engine for serious independent game developers. While it's not a simple "Game Creation Kit", it has all of the features that you need to develop cutting edge games.
The rendering engine is a little rough around the edges, but with a little elbow grease you can accomplish all kinds of next-generation rendering effects. The GFX2 rendering layer abstracts the rendering process for cross-platform, cross-API rendering with DIrectX and OpenGL support. The procedural material system generates shaders for nice generalized per-pixel lighting, and there's an integrated glow-buffer for nice glow effects. HDR lighting effects are also supported out of the box.
The physics system is adequate, though for any substantial physics effects you will be better served using a physics middleware solution.
Integrating FMOD audio is painless, and produces very nice 3d audio.
The shining light of the Torque series is the network layer. I would put the Torque networking code up against any AAA game engine on the market. Making multi-player games is just as easy as making single-player games.
While the documentation has been a problem in the past, the guys at GarageGames have listened to the pleas from the community and hired somebody to work full-time on revamping the documentation, and it's now a very nice and useful collection, and improving daily.
The content formats are fairly cludgy. The DTS format, while versatile, is often a pain to work with, because it strives to be everything to everybody, which is never a winning formula. For animated skinned characters DTS is fine, for anything else it's too much. The DIF format is an outdated relic that I wouldn't bother with.
The good news is that implementing your own custom formats is quite simple. So give the DTS format a quick trial run, and if it doesn't suit your needs, don't get frustrated. Not only do you get a great game engine, but you get the source code, so don't fight a losing battle, just dig in and customize it to your needs.
It's not the best rendering engine, not the best physics system, not the best audio system, but as a complete package with source code included, at a reasonable price, nothing on the market comes close.
Update Many Improvments (Time to take another look at this engine)
I have had a chance to check out the full documentation on the newest non beta release of TGEA (1.7) (had been still working off the beta docs) and work with the engine a bit more in practice so I thought I'd update on my last review.
One of the major complaints people had with this engine was the documentation. However in addition to the community docs and WIKI it now comes (for the first time I believe) with a very good, complete set of official, documentation which is all relevant and current to the release it's bundled with. It includes screen shots , and a thorough explanation of every aspect of the engine , as well as defines and illustrates the workflow for art/content creation and import.
It is a complete professional manual for the product and it makes a world of difference in being able to effectively use the engine. I found that many of the issues that were causing me frustration were due to misconceptions I had.
With the improvement in the documentation (and thus usability of the engine), the engine stability, and it having features like Polysoup collision, mega terrains, Atlas terrains, portals, zones, BSP interiors, dynamic lighting , shaders and network multi-player support, this engine offers a package that you won't often find with such freedom of licensing and for the price.
If you keep in mind it's not a point and click game creation system (though it offers many WYSIWYG utilities and a straight forward scripting language) for the money, you won't be disapionted.
This is an engine that could be considered bleeding edge oriented in that they release new features while still a bit rough around the edges for those with the knowledge and desire to start exploring them. However in recent times it appears they have thankfully devoted an equal amount of time and effort to standards and documentation to ensure the core functionality is solid and accessible.
Because the developers were listening to the community, (and perhaps facilitated in part by some of the upcoming commercial games they appear to be developing or publishing) the engine, tools and documentation have matured considerably in the newest release. Though it was a wait, they took time to get it right this time around. With the new documentation , editing tools and utilities, this newest release is definitely worth checking out.
Fit's the bill
I won't sit here and blow smoke, nor will I waste your time using this space to vent my frustrations with various aspects of this engine. What I will say is that TGEA is a viable engine for most peoples projects.
The engine utilizes .dif format for brushwork which is similar to the BSP style of format. They have released Constructor, which like any similar program comes with afew bugs, but is fully functional. If you so decide that Constructor is not your thing, you can always use almost any other .map supported app ie gtkRadiant, Quark.. etc.
Model support comes int eh form of .DTS format, which In my opinion is cumbersome / overly complicated to get working right, but IS fully functional. Scripting can be done through their torquescripting language which is not all that difficult to catch onto, or you may write things in C++.
The community overall is very helpful, and will become your biggest asset while learning the ropes. Documentation, while they do have some, and continues to grow, is still lacking. Also, there are many GG provided / community provided packs you can purchase to make things easier (several items have been provided by the community for free).
My team has been working with TGEA for approx. 8 months give or take, and our thoughts on the engine has ranged from total frustration, to very pleased. More recently due to afew fixes, and ourselves learning the engine (which does have a curve) I am more inclined to recommend TGEA. I warn you now, if you think you are going to walk into this engine, or any other engine without having to solve afew issues, it is not going to happen. If your eyes are wide open with appropriate expectations, you will be fine, and eventually very happy with TGEA.
It's pretty good
I'm sure there are other engines out there that are better (Unreal, etc), but for the price, this one's pretty good. There is still work to be done in the art pipeline and documentation, but it's stable and it's easy to program once you get used to it.