Ogre3D vs. Torque

#1 The last time I looked at Ogre it was basically a graphics engine. I'm not sure how much it's changed, but it doesn't seem to really have support for actually making a "game", but it would be cool for tech demos. For people interested in making games Torque has quite allot already setup and if you want to do something unique you'll of course have allot to add, but most of the ground work is already done for allot of it. I looked at Ogre before I went with Torque and I'm pretty happy with my decision to go with Torque.

#2 Ogre as nice as it is, is only a Graphics engine. They have added a couple of addons for physics and such, but it's still only a Graphics engine.

They are currently making tools to help build scenes. But its still only a Graphics engine.

Ogre           Torque
Graphics                Direct9        OpenGL
Sound                   Addon          BuiltIn
Physics                 Addon          BuiltIn
Scripting               Addon          BuiltIn
BSP Rendering           Q3 Kinda       BuiltIn
Networking              none           BuiltIn
Integrated
   Landscape/interior   none           BuiltIn
Portal Engine           Kinda          BuiltIn
Tools                   Being Made     Already existing
Proven Track Record     low key        Tribes 1 & 2, plus a
                         indy games    whole host of others
Community               Great          Great
Learning Curve          Steep          Mountain

Don't get me wrong, I like Ogre. As a Graphics Engine as Sinbad has claimed it rocks. I was very active for a while with itm then I found Torque. Have not gone back to Ogre since.

#3 Hi,

I have to reply to this because I use OGRE constantly and yet still am a torque owner.

OGRE is built to be a graphics engine and nothing more and it achieves it immensively well. It has an extremely well thought out design which allows you to create any kind of application on top of it (or should I say: using it). OGRE demands a lot of work from the developper because of its open nature but imho it is well worth the trouble.

If you want to make a game, it really depends on what you are trying to achieve and in what time frame. OGRE has the best graphics quality hands down compared to torque and even when TSE fully comes out, OGRE will be better at graphics because of its versatility.

OGRE is not a sound engine, or a physics engine, or a game engine or anything else for that matter. It just wants to render 3d graphics and give the developper as much control as possible. Torque on the other hand is a full blown game engine with everything integrated and working together. That is a major feat to do by yourself even if there already are tons of libraries over there.. tying things up together is a big job.

Do you have the time necessary to tie a sound engine, a physics engine, a scripting engine and all that to be able to make your game? If yes, you should really consider OGRE. Torque is great to just start building your game right away and hacking at the features you need to have for your game.

To clarify some things a bit from what Simon posted and just some other things:
- Ogre can use the direct3d7 and direct3d9 render systems as well as openGL.
- Ogre does run on windows, linux and macs very nicely and I heard it has been built on solaris too.
- Ogre is not a game engine and thus will never have a good input system, or a sound engine or any of that other stuff.
- With OGRE you are free to customize any part of the rendering and add new terrains or change how the render queue operates without having to hack into the library.
- The tools that OGRE really needs are there and are getting better. It has exporters for 3ds max, maya, blender, lightwave, wings3d, ac3d. Not all of them are fully functional though.
- OGRE doesn't have the track record that torque does but it appears to be getting there as a lot of people are producing things with it. A few universities are using it to make games or use it as a 3d visual engine for other things than games. And there are quite a few cool screenshots to dig up over the web.
- The community is truly amazing just like here and they truly do help you when you need help.
- The learning curve i'd say is pretty easy but it mostly depends on how comfortable you are with C++. When I first started using OGRE I did not know much about 3d graphics at all.

My post is mostly about OGRE because I know a lot of people are going to respond on how torque is. I would say that OGRE is truly a good graphics engine to use and I am using it myself. Hopefully in the next few months I'll post a plan showing off what I've been doing for the past year by myself, until then it's back in 3ds max trying to model my main character. You gotta love programmer's art :)

#4 I've really got nothing more to add to Simon and Lois's comments, I just want to say that I agree with them.

I used Ogre prior to buying Torque, and I'm looking forward to purchasing TSE when it's released :-)

#5 Another option, if you REALLY want to go mad, is to drop the renderer from torque and use OGRE for the render part only :)

Ok.. silly, but its do-able!

#6 Currently there are tons of middleware solutions an indie can use for free and custom build their own engine using whatever they tools they want to integrate into their graphics pipeline. many of which are better than what comes with Torque. Of course you can allways integrate these into torque too if you wish. But if you look around you may find other solutions allready support things like Tokamak, ODE, verlet, coldat for colliosions. Middleware lighting tools like Giles etc.

I've noticed some talk about these things in the forums, but not sure how many are allready avaliable for TGE developers. But the tools are out there and avaliable to developers using a lot of different languages and tools in many different ways. GG has their network library avaliable to developers, so there's no reason someone couldn't just use that with ogre, ODE, Fmod and create an engine of their own in a relatively short time to create somethoing more powerfull than any of these things as they stand on ther own.

Just mentioning it as someone I work with has just done something like this and the results are very impressive indeed allthough he created the GL graphics renderer himself.

#7 Having done my own 'engine' as opposed to using torque there's a world of difference between a 'working' good looking engine and a battle tested AAA engine.

I can get a renderer showing the latest shader effects in less than a hour. I'd guess at a few months to get even close the the level of Torque's networking. Choose your priority...

Jay, I know what you're thinking. The 'loss of control' from not having written everything (or at least the glue) is quite daunting at first. However I think that's something most programmers need to get over. You'll save a ton of time using torque, and don't worry there will be plenty of stuff to do if you want to ;)

It was the lack of docs that originally scared me off, but even they're a lot better now.

Can't add more to what anyone else has said about Ogre. Zoom's idea probably has more merit than he let on.

#8 I knew when I posted here it was about like asking people their opinion of President Bush at the Democratic National Convention.

I think Gareth knows where I'm coming from. I'm no stranger to building 3D game engines. Void War is about to ship with my custom homebrew engine which I've already grown to resent :) There's a lot more to it than just putting pixels on the screen, as I'm sure everyone here is aware. It's not particularly HARD, just time consuming... and when you are looking at adding importers from different file / tool formats, Mac OS-X support, all the little UI widgets that make the interface kinda-sorta standard with the native platform, and just dealing with all kinds of unique hardware issues, and the BUGS that come up in immature code ... it's a lot of "butt-in-chair" work when you'd really rather be making a GAME.

I have been leaning in Torque's direction. But the TSE is still too much of an unknown, and I've been scared off a little by some comments about how much of a pain in the butt it is to get into the guts of Torque's systems to enhance it. Ogre seems like it's more feature-rich on the graphics end, as well as being easier to extend. It also *seems* (correct me if I'm wrong) to have a broader tools path. I'm not sure about the UI Widget support on either end. Torque's advantage is it's maturity, perhaps more robust (if narrower) tools support, and a lot of built-in functionality that I wouldn't have to create myself or bring in from a third party. So the control-freak part of me is concerned :)

Of course, I've browsed a little bit of Ogre's documentation, and of course done a feature-by-feature comparison, but that never tells you much. In about two months I'll have SOME time to dive in and really explore these questions in more detail. I was just looking for as much of a heads-up as the seasoned vets here could offer. Since projects 2 and 3 are on the drawing board, I'm trying to get as much early, easy planning done as possible.

Thanks for the help and advice, folks!

#9 Ogre is actually not as flexible with shaders as TSE is. Since TSE can procedurally create the most common shaders, there is no programming involved - even across different hardware types. With Ogre on the other hand, you have to hand code every shader material for every type of graphics hardware. That adds up pretty quick when you are making a game.

Their GUI system and terrain are also pretty crude compared to the TGE. It's one thing to code up some nifty looking effects, but it's a lot more work to integrate them properly into an engine or a game.

#10

Quote:It's one thing to code up some nifty looking effects, but it's a lot more work to integrate them properly into an engine or a game.

Right on! To add to this, it's one thing to make some nifty looking art for a tiny little bit of a grand design. It's another thing to make all of the grand design's art and hook it all to gameplay.

If you want to make an engine, I wouldn't plan on making games anytime soon. If you want to make games, you have to finish them.

#11 @Jay:

In response to your post, I saw Brain download the Ogre and take a look at it. Like he said, there are some nice little effects, but it is far from an engine. From my perspective, i.e. seeing it run next to the TSE, it looks like stone age tools compared to space age weapons. I'm sure this will incite some kind of flame war, but I just know what I see.

-Jeff Tunnell GG

#12 Jay,

TSE is really just TGE with new rendering code. Which means that the other 80% of the engine is the same... And the new stuff is very similar to the old stuff. We've made sure it's easy to modify and extend (given we know we're going to sell it, unlike the original authors of TGE :).

#13 Heh - Joshua - your DOOM builder stuff is part of what interested me in going to Torque on this in the first place. Not that it couldn't be done for some other engine. But the idea of whipping out "roughed out" versions of interiors at DOOM-level creation speed is... well, kinda intoxicating.

Jeff & Ben - That is reassuring. I guess now it comes down to when the TSE will become available :) Though I think it will probably be within my timeframe. What that DOES mean is that I'll probably want to hold off doing any gut-ripping of Torque's engine until then.

#14 @Jay: The DOOM level to brush reprentation stuff works really, really well for roughing out interiors. We are still finding the game's art very time consuming for the detail, scope, and quality we want to achieve.

Rendering speed in only one kind of speed. As an independent, a balanced must be found between quality and construction speed. Our engine, "The Prairie Engine", is hand tuned for us to make our game. I have put WAY more work into the engine/tools than I thought on the front side. I, myself, sometimes have a hard time believing an engine based on QuakeII is capable of what I am playing.

I am really glad it is damned hard to make independent games. It increases the value of my product.

-J

#15 I played w/ ogre too, but i think the issue is what everyone else is basically saying. Do you want to spend a year making a game or do you want to spend a year integrating tools into a tech demo to make it into a gameable engine? IMHO Your efforts would be better spend working on TGE's rendering or even better yet, gameplay.

-s

#16 Hi,

I have to disagree with Brian Ramage and Jeff Tunnel on how OGRE compares to TSE but I have not seen TSE so I cannot say much about it BUT OGRE is extensible and you can write shaders in HLSL, CG and ASM and if you write your shaders in cg it will work on every card that support the versions of vs and fs that you told it to compile it for and it will work in directx and opengl. It would also be pretty easy to add a plugin to get the gl shader implemented too (glslang I believe? )

Don't get me wrong, I love torque and I use it if I need to prototype something very fast. I have used it that way and it works great. Though as a hobbyist I prefer OGRE because I feel it is much better on the graphics side than Torque is. OGRE is not a toy.. it is a very well designed and powerful engine IMHO just lke torque is. They have different purposes though.

Anyway, we all need to remember to use the right tools for the job and for some people it is OGRE and some others it is Torque.

- Lois

EDIT: Also the gui of OGRE is indeed not as good as torque's and it is being remedied but that's definitely one downside right now. OGRE has a lot of terrains most of them made by the community. OGRE takes a lot more work than torque and like I said before, act accordingly.

#17 Lois: I think you missed the point. OGRE is not an engine, it's a graphics layer with some tools that are NOT built in.

#18 Hi Stefan,

I guess I must have worded myself wrong because I know very well what OGRE is and isn't since that's what I am using for my main project.

OGRE stands for "Object Oriented Graphics Rendering Engine" so OGRE is indeed an engine but not a game engine like torque is. OGRE is not just a graphics layer in my opinion since it allows you to do so much. Tools are not built in OGRE because that would force the people using OGRE to have to use these tools which is why the the OGRE's GUI is a plugin and not really part of the core. OGRE really wants to give choice to the developpers and it does that admirably well.

Sorry if I wasn't clear.

- Lois

#19 I don't think anybody on this thread confuses Ogre with a "full game engine," folks. I think everyone seems to know the difference (and if not, it's been made abundantly clear a few dozen times... :) )

But if it was ONLY about having a working game engine, I'd stick with the one I've already spent 18 months building and refining. Part of my intent when I started this project (before I'd done serious work discovering the indie community) was to build working libraries that I could then leverage into future titles --- making later titles MUCH easier to develop.

And that remains an option. Integrating what I have with an improved, multiplatform, more complete graphics engine is a reasonable possibility. Especially when I hear things like how it takes six months just to get a handle on Torque... I don't want to spend that much time getting a handle on the engine.

But there's going to be pain either way. As was said earlier, you have to use the right tool for the job. And if all you have is a hammer, EVERYTHING looks like a nail.

#20 The learning curve for torque is steep. But is far less than what it would take to write, test, debug and document similar features. You might spend 6 months learning the ins and outs of torque, but you can become productive almost immediately -- and that is not something that can be said for the build your own libraries approach.

Perhaps what you are saying is that secretly you don't want to make games... you want to design and build libraries for others to use? A very noble goal. And a profitable one if done properly.

However, if that really is where you heart lies I would look at Torque (and TSE when it is released) and things like Ogre and see where you can make substantial contributions. Perhaps create a programming pack like the lighting pack that was just released. There are areas where this engine is showing its age. People are struggling with the edges all the time. AI, audio, physics, and the editor are all being looked at as well as the current graphics and terrain. The number of resources people contribute will give you some idea of where interests lie.

And such libraries can be put together without the assistance of large numbers of artists and modelers, making it a good target for individuals and small teams of programers with a passion to see good games created with what they have created.

Food for thought.