TV3D vs Torque Shader Engine

[alpha_counter]

the TSE dancing monster demo seems quite impressive...

TSE and TV3D are two of the more affordable 3D engines out there.

does anyone know what TSE is like?

given a choice of TV3D and TSE,  which one should I buy?

what are the pros and cons of TSE?

 :?:

[AriusMyst]

the TSE dancing monster demo seems quite impressive...

TSE and TV3D are two of the more affordable 3D engines out there.

does anyone know what TSE is like?

given a choice of TV3D and TSE,  which one should I buy?

what are the pros and cons of TSE?

 :?:

I havent used TSE, an i'm infamously biast towards TV3D, so i just thought id say "welcome to the board"  

[BlindSide]

I hadn't planned on posting about this quiet yet, but I figure now is as good a time as any.

Soulos and I actually had a chat about this very thing on the IRC, and we decided that instead of comparing screenshots/feature lists between TV (6.5) and TSE, we would make a tech demo in both using the same level, same textures, sounds, models, etc.

We havn't gotten too far as of yet, but when we do I'll post some info on the board.

Btw, the TV one will be open source, hopefully ready by the time 6.5 goes public. The TSE will be closed source due to licensing restrictions.

[potato]

Well, here's a quick rundown from my point of view:

TSE
===
Pros:
- Comes with a full game engine, if your purposes are game-related, this is good.
- Comes with some exceedingly crazy netcode.
- Supports shaders (this one is a no duh)
- A LOT of gamecode is built for you, this is very good, and depending on your skill level, this will cut your development time significantly.
Cons:
- C++ only, unless you want to learn their own little TorqueScript language. Eww.
- Seems unecessarily geared towards big-outdoor FPSes. Doing another genre will require an insane amount of source modification.
- Lack of basic features - last time I checked there wasn't even lightmap support unless you buy a 3rd-party code "upgrade"
- Convoluted art pipeline - your artists will hate you
- From what I've seen, slower than TV, but this is a qualitative statement, I have not benched the two together.

TV3D (I'm talking 6.5, since that's the one I play with most often)
===
Pros:
- Insanely fast.
- Usable in multiple languages (VB6, Delphi, VB.NET, C#, C++...)
- Clean structure makes it easier to understand and pick up
- Supports a lot of common features transparently. Lightmaps, for example, require NO work on your part to support.
- Phenomenal support. With how many engines can you contact the devs directly, and talk to them in chat?
Cons:
- No netcode/media engine yet - that will come later
- No game engine to wrap it around - this is only a con if you're doing an big-outdoor FPS , since Torque, IMHO, is just as weak with any OTHER type of application.
- Lacks the "proven" track record that Torque carries.
- No source code with engine - but for this price, can you expect it?

[Oxygen]

"Lacks the "proven" track record that Torque carries. "

You mean like Tribes 2: Failure on an epic scale? I'm a HUGE fan of the Tribes franchise and T2 is it's bastard child.

Another con of the Torque engine is it all looks the same. I've yet to see one game that doesn't "look" torqueish.

TV for the win again.

[Javin]

I think it's a bit unfair (for torque)  to compare it against TV, really.  I mean, TV isn't a gaming engine.  It's a 3D Engine that can be used for anything from real-world simulations (using its built-in physics) advanced graphics interface applications (which I use here at work all the time) and game development is only a fraction of what TV can be used for.  Additionally, TV can work in nearly any language of choice.  

Torque is a gaming engine used for making a specific type of game, that can only be used in C++ or its own scripting language.  From what I understand, it's also not even as fast as TV (as many parts of TV actually use ASM and are highly optimized.)

The only realistic con for TV that was listed was 6.5's lack of network code.  Yet, 6.5 hasn't been completed and released, and the network code is definitely on the way.  

Where's the comparison?

-Javin

[nacra]

* TSE is planned to be cross-platform (Win, Mac, Linux), like TGE.   Some TGE titles have been ported to XBOX so I think the plan is to sell Torque titles on XBOX Live.  Or potentially  through the normal console distribution methods.

I second potato's comments on Torque being locked into FPS-style mode because of its legacy.  And the art pipeline *is* interesting.  I mean difficult.

[potato]

I had a hard time picking out a large number of TV's cons, but I had to make the comparison somewhat equal.

You mean like Tribes 2: Failure on an epic scale? I'm a HUGE fan of the Tribes franchise and T2 is it's bastard child.

It was commercially successful, and any gameplay failures certainly cannot be blamed on the engine.

TSE is planned to be cross-platform (Win, Mac, Linux), like TGE. Some TGE titles have been ported to XBOX so I think the plan is to sell Torque titles on XBOX Live. Or potentially through the normal console distribution methods.

The market for Linux games is INCREDIBLY limited, especially knowing the fact that many Win32 games can be played with emulators on Linux. Mac/Xbox is the only cross-platform-ness that's truly valuable, and TV's already got one side covered.

One advantage of Torque: OpenGL renderer, makes the engine more portable.

[Oxygen]

It was commercially successful, and any gameplay failures certainly cannot be blamed on the engine.

Very true.

[GoodVillain]

I liked the game. Even the first one, Tribes Vengence, and Tribes on the PS1. I heart Tribes.....

[Spetsnaz]

I posted this elsewhere, but our game deve group is  trying to see what else is out there besides TSE. They are taking for frigging ever to make it stable and to get to the final milestones.

My problem with TV3d is the development pipeline. Theres no bloddy level editor! You have to use a modelling program to make your levels. Ewww.

We're open to being preached to. change our minds!

[Borys Pomianek]

Umm i dont want to change your mind.

Actually i would say: if your thiking about releasing a commercial game.

Create your own engine and software.

That will a lot better than tse or tv becouse it will be customized and created specialy for one game. for specyfic needs.

Now TSE main problem is that working with it is like moding a game.

With tse you will spend a lot of time moding the out of the box thing before you even start to write a single line of code.

Now, moding someones code is not bad or anything but its all about what you like.

I for instance discovered that creating my own engine is a lot more fun that using someones. Still when 6.5 goes public beta iam gona jump on and use it (as it will support c++ more).

Now i dont want to use tse becouse i cant risk 300 bucks to discover that its not for me.

With tv you can check the software, use it how you like and then make a decision.

Yeh sure tse has a demo but that demo crashes on my system lol.

So actually tse is loosing a lot by releasing out broken demos or demos that are not smooth on really fast systems like mine.

Now i understand that my graphics card is not intended for games but duh, if i want to make something diffrent than a game i want it to work on proffesional hardware and not only popular gaming hardware (wich is not bad for proffesional work but usually lacks features for that kind of use)

Just make your own choice, you can eseally create your own world editor if you have time.

You can create your editors, create your game then sell your game and sell your editors for that matter.

Generaly if your thinking about some mature development, each company usually have its own aps for some stuff customized for them.

Creating tools is the first step to launching a big scale project.

BP

[Spetsnaz]

I dont think you realize how unrealistic it is to just create a level editor. Maybe it's easy for you, but it would save us a lot of time to use a system that already has one instead of having to create software just to even get to start doing the work.

 I guess TV3d requires modelling environments? Sigh. Ok. Thats a bit of a let down.

[billythekid]

Its not THAT unrealistic. And Im not sure how you would get custom content into the game without using some modelling tool like Max or Maya. The editor would just allow you to place that content in the game world, the content still needs to be made.

It sounds like you guys want to make a mod for one of the top engines like Source or Unreal.

[Borys Pomianek]

If your serious about doing anything Spetsnaz you have to understand that without content there will be no game.

You have to use a modeling enviroment with EVERY engine out there even those tha include level editors or any other kind of editors.

That is becouse you actually HAVE TO MODEL your MODELS.

Becouse 3d world consists of models !

a world editor is an app just as billy told you to take X number of models do X number of things with them and put them together and create (generate) a level out of that preferably in your own format.

You can at the same time use 3rd party bsp editors to create your levels but not other not static stuff in that level.

You can use olmost all bsp editors out there.

The same you would use if you would create a mod for one of the mainstream game.

gosh iam tired of typing that over and over.

Just read a thing about creating content for games and then make stupid statements.

BP

[BlindSide]

Making a level editor is not very difficult - I made one that encapsulated almost every tv feature (including full physics support) in 3 months. I do this in my spare time, between 3 jobs and school - a level editor is definately doable by a dedicated group of programmers.

Using a canned level editor, especially with an mmo (I'm working on one as well) is NOT going to work. Sooner or later, you're going to come to realize that you are restricted. Especially with an MMO, restriction is a bad thing. Then, you'll end up writing a level editor anyway.

Additionally, max is going to allow you more creative freedom than any map editor out there. What I would suggest is model everything in max, and have dummy objects for things like triggers, lights, npc's, etc. Then, assign those dummy objects specific textures only used for dummy objects. When loading the mesh, check the textures, and put in a light/npc/trigger in place of the dummy object.

[Mincetro]

Why does everyone here think that Torque can only be used for FPS's?

I've had torque for a year now and I've used it for Adventure games, Side Scrollers, Wireframe Flightsims and A small RPG OOTB by modifying the source code and Torquescript.
There's an RTS starter kit that includes thousands of lines of code and resources that are used in RTS's (character picking in realtime environments, minimaps, attribute assigning, Unfocused free camera, etc.)
You just need to get to know the engine a bit beforehand.

TV3D uses precompiled closed-source libraries (which means no custom-written extensions) which are linked to .Net or Delphi to assist with using DirectX functions. I used TV3d to create a small Sim City clone a few months back in .Net, but it seemed a bit less suited for RPG's and FPS's and more to first-person adventures or DungeonCrawlers. I havent tried 6.5 (I don't have a license) but I can't see much comparable about the engines.

I'm considering upgrading from Torque to TSE, but I"m researching as much as possible for the upgrade.

[Borys Pomianek]

Yes tv is a library.

You write your own code and use tv just for graphics or use tv media and tv net olso if you want.

You can write custom extensions. you can write anything.
A library is just a library people.

If it lacks something for your projects you just write some of your own code that can but does not have to use tv.

While torque is a ready game code. A project with a sourcecode and some docs.

The main argument against torque is that you dont work from the bottom up creating your game but you mod things.

Its not that you can only do fps. Add some simple rpg logic and you have an rpg out of an fps. That is not the point.

Point is that i have a game design document and to realise it i would have to overhaul the whole engine becouse i dont want to just mod existing code.

moreover torque has a lot of bugs wich include broken avi playback so no cutscenes.

Most of typical, general features that games have are writen down and documented either on the net or in numerous books so having torque just becouse it includes some very general and widely know game code is not a good option for most people.

Some things in torque are more worth the time like the networking but actually most people never go to the level of online games.

For me its more important to create the content itself so that it will be what i want rather than trying to do an online game just becouse its online.


I would want to use torque too if i could evaluate the engine itself !
But i cant !

Iam not risking 250 bucks (torque + TSE early license) to be able to check if TSE is for me.

If garagegames gives a full evaluate version (with source code and everything that the normal license includes !) for 10 bucks i would check it out and then pay the 250 if i like it.

BP

[Mincetro]

Fair Argument.
I'll probably be getting TSE sometime early this year, I'll read the EULA and see if I can make an evaluation EXE.

UPDATE: a New, Unofficial TSE demo using the starter.FPS demo to show off the shaders has been released here: http://www.garagegames.com/mg/forums/result.thread.php?qt=33724