Ogre - Why waiting for 2.0

[FrameFever]

Hi there,

in the last months we have seen some big changes in the AAA game engine market:
- Unreal is now free to use (free for free games; or pay 5% per quarter for commercial)
- Unity cost 75$ per months
- CryEngine 10$ per month (only API C++)

I have not yet found the time to read up all the documents about Ogre 2.0 and what is new.
That's why I am asking here, I hope it's not so provocative:

Why should a new user should choose Ogre 2.0 over Unreal, Unity or CryEngine?

But to make things more fair, I would like to compare only the render engines itself.
So what are the benefits of Ogre 2.0 over these engines, besides of course the much better license model.
Will it be the fastest render engine on the market?
What key features can we exactly expect when Ogre 2.0 is finished end of this year.

When we have some nice points, we should put it in the front of the website, so the average user can see easily what the advantage is.
Otherwise, maybe many people will not start using Ogre 2.0, because everybody is starting to use these other engines.
Don't get me wrong, I like Ogre and I would like to start use it again with the new version, but first I would like to hear what I can expect from it now.

[xrgo]

Hi! In my case this are the reasons to choose Ogre:

1. Control over everything. I can choose the components that I want (physics library for instance, bullet over physics, etc). And easily integrate other libraries, Qt, sqlite, etc. This way I can make my engine to fit my needs.
2. Open source. I like to (try to) understand whats happening on the insides, and modify it if necessary.
3. Performance. Its going really good on Ogre 2.1, it seems better that other engines, but I can't really guarantee this since it still lack a lot of features (GI, probes, ssao, etc).
4. Linux support. I use only Linux, to develop and to deploy. UE4 editor works under linux, but works really buggy and its really slow.
5. Completely free, for ever.

[Thyrion]

this for ogre generally:

1. Control over everything. I can choose the components that I want (physics library for instance, bullet over physics, etc). And easily integrate other libraries, Qt, sqlite, etc. This way I can make my engine to fit my needs.
3. Performance. Its going really good on Ogre 2.1, it seems better that other engines, but I can't really guarantee this since it still lack a lot of features (GI, probes, ssao, etc).
5. Completely free, for ever.

additional:
- you are realy independent

but why 2.0 and not 2.1?

[xrgo]

but why 2.0 and not 2.1?

dont know... I used 2.0 like 2 weeks, then jump to 2.1 as soon as it was available

[FrameFever]

One additional note, with the term Ogre 2.0 I mean it synonym to the new ogre.
Not confuse with Ogre 2.0 & 2.1 I don't get it why there is already a 2.1 anyway, when 2.0 is not yet even released. But this is another story.
I am more interested in the key rendering features. What advantages will the user get?

@xrgo
2. You can also look into the source code of the unreal engine. So no advantage.
So only point 1, 3 and 4 seems to be an advantage, although a performance comparison is not yet available.
Remain 1 & 4

[xrgo]

2. You can also look into the source code of the unreal engine. So no advantage.

yes I know, but only for UE

Remain 1 & 4

and 5 =)

[N0vember]

2. You can also look into the source code of the unreal engine. So no advantage.

When we stress that Ogre is Open-source we don't mean it in the too literal "you can look at the source code".
We mean in the sense that "you are free to do basically whatever you want with the source code". Modify it, enhance it, distribute it, take it apart, make a paying product on top of it, freely, with virtually no constraint.
It's an order of magnitude of difference.

I guess using Ogre is close to a political choice for me. It's about transition away from proprietary software. And not willfully enslave yourself / your software to any kind of third party.
I know you're not looking for this kind of answers, as you want to look at the technical side. It is true that being open-source alone doesn't mean anything if the technical side is weak. But it still matters a lot, to some people.

[cybereality]

To be honest, for any beginner that asks me what engine to use, I would tell them to look at Unity or Unreal. You would need a good reason to use something else (and in that case, you probably know what you're doing).

There is certainly a flexibility that comes from combining different libraries, or doing custom work, so I won't argue that. And open-source is a plus, for sure.

But really, for probably 90% of the types of games people want to make, you get basically everything you need with an all-in-one engine like Unity or Unreal.

Maybe you are in the other 10% that has unique needs or requirements, in which case something like Ogre may make more sense. I don't know.

[syedhs]

I think for someone who is quite well versed with graphics programming.. Ogre is very nice because this is the place where people can make full use of their expertise. Each source code file is accounted for, well understood, scene managers can be modified to each probably unique uses - it is satisfying to programmers' soul like what John Carmack one time said "Low level programming is food for programmers soul" <-- something like that, I dont remember the exact words.

However, there is one big stumbling block.. it is the world/scene editor. It has to reach to certain level where it is usable, feature rich and easily modifiable by the communities. Constructing world editor and game engine at the same time can be quite frustrating because programming world editor can deviate one's time a lot from actually creating a game.

[Klaim]

The 3 reasons I use Ogre (because it's the actual question you are asking):

1. My game's design imply inserting special code in between (graphic, audio, etc) engines comunications, or glue-code. I cannot use a full game engine for this even if they provide the source code because the whole game engine is always designed by taking into account how the whole works, not like if the different parts could be separated (which is both an advantage for most users and a major issue in my case). This is specific to my game and not usual requirement at all for most games though.

2. I need total control and flexibility of the rendering system, mainly because in my game a lot of graphic things are driven by "plugins" (by design) which mean that I can't guarantee much about what you will get in the screen. The graphics can be "hacked" by some part of the game temporarily and this mean that that part of the code needs to talk to the graphic system directly. My experience with UT and Unity is that they enforce specific idioms to you so as soon as you want to go outside of them you are stuck.

3. Open source and license allows me to modify Ogre if I want to provide a specific version with my game that have some game-specific fixes or modifications. I'm not stuck or forced to go around some limitation of the engine, I can dive into it's code and change things, with my limited graphic rendering knowledge as an actual limitation which I still can push by paying someone do the thing for me.

As you can see it depends both on the game and what the team is capable or ready to tackle. You can't generalize the choice of tools. Except if you go the other way arround: you chose a set of tools (say Unity) and you do everything within the limitations of this tool. This is a great aproach too, in particular if you don't have a specific idea of what you want to do. It's more explorative than knowing your starting point and global goal and therefore finding what tools match your needs.

[TheOnlyJoey]

Personally I think comparing Ogre to a game engine is not relevant with the simplest reason that Ogre is not a game engine.
If you don't see the advantages of using Ogre, there probably is none for you since both cater to different users.
UE or other game engines are created for people who want to jump in and create a game, Ogre is not made for that.

Whenever you are creating custom software and need a good graphical base, or want to do create something from scratch without the need of reinventing the wheel, Ogre is a really good choice.
If you are creating a game engine, or are doing something with visualisation and not need the bulk of a entire framework (ie, not needing physics, animation, 2d framework etc etc) using Ogre has a lot of advantages over using a pre-built package.

This is just a quick perspective from a software designers point of view, but imho the most relevant for this comparison.
It is not about the prettiest out-of-the-box image also, but what you can do with it.
If you want to point and click some things and create something pretty, use the tools that are made for that, but do not expect something like that from a software library (which Ogre still is).

Also the whole 2.0/2.1 rave (although really interesting and technically impressive) is not something 'that big' as some seem to put forward.
Ogre has been useful in many stages and it does not matter which version you use, if you know what you are doing with it, you will create something beautiful.
Many shipped Ogre games can show you this!

[Kojack]

Sadly with the increase in use of Unity and Unreal, we might have more programmers making acceptable games, but we are breeding a generation of programmers who don't have low level skills. At least for learning development, I think assembled frameworks using ogre, sfml, etc are superior to Unity. It might not be as friendly, but it doesn't hide the technical issues. I've seen it first hand, students who started with sfml and ogre were better unity programmers than students who only learned unity. Of course someone on their own would have an easier time learning Unity, I'm talking from a teacher point of view.
Same thing with languages.

Unity and Unreal are very much game engines, which makes them bloated for non game applications.

We could say the same about a lot of tech and theory. Why learn how path finding works when Unity has a nav mesh built in? Why hasn't Havok gone out of business when Unity and Unreal use Physx? Why do DSLR cameras still exist when everybody has a smart phone.

although a performance comparison is not yet available.

From my own simple tests, Ogre 2.1 is far ahead of Unity 5 for heavy PBS scenes, and UE4's editor had a heart attack before I could even finish making the scene.

Some of my main issues with Unity:
- commercial and closed source (although the free version is getting better)
- can't remove things I don't need in a project (physics, audio, etc)
- I don't like C#, UnityScript or Boo. I prefer C++, Lua, Falcon, D and others. (C# has some nice features, but it also annoys me a lot)

[frostbyte]

When we have some nice points, we should put it in the front of the website

the plan is to put a really nice demo...as we know... 1*good_demo > 1000*words

so the average user can see easily what the advantage is.

what is an avarage user?
if you refer to non-programers/script-kids/artist then OGRE being a c++ lib is out of scope anyhow...

Otherwise, maybe many people will not start using Ogre 2.0, because everybody is starting to use these other engines.

it will/is happen anyway...you can't control the "herd" and you can't beat unity/unreal in terms of advertisment
neo-axis, shiva and other capable game-engines exist for a long time now...still they have only a fraction of unity users...

also...
5% royalties != free
can see source/closed source != open source
game engine != rendering engine
game editor != rendering api lib

in terms of whos fastest/features etc...i guess all engine will eventualy straighten lines since there is no quantom physics/black-magic involved here
probably each engine will have it strength/weakness...its too early to compare...also comparing engines is not a trivial task as it sounds to be...

[FrameFever]

mhm... some people here don't read the posts exactly.
I said, only compare the render engine and also don't compare the license.
Furthermore, I never said I want to make a game.

Anyway, at the moment I would say we have only following points:

• - lightweight engine; you can add or swap components on your own very easily
  - flexibility

I hoped to read things like:

• - nice API
  - very good documentation
  - following render features... (e.g. the best shadows, particles..., you name it)
  - performance is very good for scene like....

And the end I would say, we need someone who has taken a deep look into the engines mentioned above and knows exactly in what state the ogre development is and also how it will look like when it is finished.
Maybe dark_sylinc can answer this, I would think out of curiosity he has maybe at least taken one look at the unreal sources.

[Klaim]

I think you might be misunderstanding the "not a game engine" point. It IS a major advantage, as several people pointed, in particular the ability to compose your engine.
It's not something to be dismissed just because it's comparing apples and oranges, it's a big difference that can sell something like Ogre, SFML and other engines.
It's a selling point.
Game or not game, it's all 3d visualization of some kind of dynamic modelisation, and the point stay important.

I wouldn't say "nice API" because it feels like 90s code, but it's good enough and certainly modernizable later.

[frostbyte]

like kojack said...its too early to compare since some features are still missing...
anyway ogre2.1 internals were inspired by the "azdo optimization" talk, so performance should be about as good as it gets...
ogre is becoming gpu bounded( in contrast to cpu bounded ) which means it will operate in direct relation to the level of the users gpu with minimal driver/cpu overhead...but still every engine has its unique optimizations( example: unity has umbra occlusion culling )
so repeating myself...its too early to compare, and probably every engine has its strengh and weakness

btw...i never saw a real comparison between engines...
the ones i did see always ended up with flame wars and apologizing for not knowing how to fully optimize the scene for the "losing engine"
and i'm talking here about the simplest scenes containing only a box or some 2d sprites...

i think, you make a good point about the need to have a features list and maybe a bit of self PR
but comparing OGRE with other engines just feels wrong...(... unless it wins )

[Thyrion]

what i would choose for what usecase:

UE4:
- ArchViZ (no royalty and it seems no one can beat the realtime realistic rendering of UE, Blueprints are also perfect here)
- First person shooter games
- Story driven games
- Console Games

Unity:
- Mobile games(UE forum is "full" of bad mobile performance threads)
- maybe VR (UE forum is "full" of bad VR performance threads)


OGRE:

- if u dont want to use PhysX (UE users seems to have the most problems with the physics (precision))
- make something special (full control)


i still try to find something you can't do with UE